r1024: Add optional graduations to waveform.

[cinelerra_cv.git] / doc / cinelerra_cv_manual_en.texi

\input texinfo  @c -*-texinfo-*-

@c Cinelerra CV Manual - ENGLISH
@c Cinelerra Community Version website: http://cvs.cinelerra.org
@c Documentation coordinator: Nicolas MAUFRAIS - e.conti@gmx.net

@c Licence:
@c You may redistribute the Cinelerra CV manual and/or modify it under the terms
@c of the GNU General Public License, as published by the Free Software
@c Foundation; either version 2 of the License, or (at your option) any later
@c version.

@setfilename cinelerra_cv_manual_en.info
@documentlanguage en
@documentencoding ISO-8859-1
@settitle Cinelerra CV Manual
@afourpaper
@set EDITION 1.00.EN
@set VERSION 2.1

@finalout
@titlepage
@title Cinelerra CV Manual
@subtitle @b{Community Version} @value{VERSION}
@subtitle Edition @value{EDITION}
@author Heroine Virtual Ltd
@author Cinelerra CV Team
@page
@vskip 0pt plus 1filll
Copyright @copyright{} 2003, 2004, 2005, 2006 Adam Crossfire - Heroine Virtual Ltd.

Copyright @copyright{} 2003, 2004, 2005, 2006, 2007 Cinelerra CV Team.@*

This manual is free; you can redistribute it and/or modify it under the terms
of the GNU General Public License as published by the Free Software Foundation;
either version 2 of the License, or (at your option) any later version.

This document is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with
this program; if not, write to the Free Software Foundation, Inc., 59 Temple
Place, Suite 330, Boston, MA  02111-1307  USA
@end titlepage
@headings off
@evenheading @thispage @| @| @thischapter
@oddheading @thischapter @| @| @thispage

@setchapternewpage odd

@contents

@ifnottex
@c cincvdoc_node_number_1
@node Top
@top
@end ifnottex

@menu
* Introduction::                      Cinelerra in brief.
* Installation::                      Making Cinelerra work on your system.
* Configuration::                     Adjusting the behavior of Cinelerra.
* Project attributes::                Changing the way the media is displayed.
* Loading and saving files::          Moving media between disk and Cinelerra.
* Program window::
* Compositor window::
* Viewer window::
* Resources window::
* Sound level meters window::
* Transport controls::
* Timebar::
* Realtime effects::
* Rendered effects::
* Ladspa effects::
* Transitions::
* Keyframes::                         Making effects change over time.
* Capturing media::                   Moving media from the real world to disk.
* Rendering files::
* Tips::                              Unusual applications of Cinelerra to common problems.
* Troubleshooting::                   Problems with Cinelerra.
* Plugin authoring::                  How to write new effects.
* Keyboard shortcuts::                How to accelerate most commands with the keyboard.
* Copying::                           The GNU General Public License
@ifnotplaintext
@ifnothtml
@ifnotdocbook
* Index::                             A menu covering many topics.
@end ifnotdocbook
@end ifnothtml
@end ifnotplaintext
@end menu

@c cincvdoc_node_number_2
@node Introduction
@chapter Introduction
@cindex Introduction

@menu
* About Cinelerra::
* The two versions of Cinelerra::
* About this manual::
* Getting help::
* Tutorials::
@end menu

@c cincvdoc_node_number_3
@node About Cinelerra
@section About Cinelerra
@cindex About Cinelerra
@cindex Cinelerra, about

For years, some people have wanted a way to edit their audio and video in one
place as fluidly as writing text.  Cinelerra tries to be a single location for
all your audio and video editing needs.  All the recording, editing, and
playback are handled here.  It can be used as an audio player.  It can be used
to record audio or video.  It can even be used as a photo retoucher.

There are two types of moviegoers: producers who create new content and revisit it
for further refinement, and consumers who
want to acquire the content and watch it.  Cinelerra is not intended for
consumers.  Cinelerra has many features for uncompressed content, high
resolution processing, and compositing.  Producers
need these features in order to retouch many generations of footage,
which makes Cinelerra very complex.  Consumers should consider other tools such
as MainActor, Kino, or Moxy.

@c cincvdoc_node_number_4
@node The two versions of Cinelerra
@section The two versions of Cinelerra
@cindex Cinelerra, the two versions of

There are two branches of Cinelerra.  One can be found at
@uref{http://www.heroinewarrior.com} and the other at
@uref{http://cvs.cinelerra.org}.  This documentation is focused on
@b{Cinelerra-CV (Community Version)}.

The official Cinelerra source code is developed "upstream" by Heroine Virtual, Ltd (HV).
HV shares its code base with a community version of Cinelerra (Cinelerra-CV), but does not
actively participate with the community of developers responsible for Cinelerra-CV. 
HV likes to work on its own copy
of Cinelerra, releasing code on a periodic basis every 6 months or
so. 

Cinelerra-CV was founded by developers who wanted to extend the functionality
and fix bugs inherent in the HV code base.  They decided to develop Cinelerra
in a community fashion and not create a separate fork of the original HV code.
So, the Cinelerra CV code is very similar to the official release.  CV coders
apply bug fixes (@uref{http://bugs.cinelerra.org}), enhancements to the SVN and
compliance fixes.  Programmers occasionally send patches upstream.  In this
way, Cinelerra CV has a number of features that the official version does not.

Unlike other programs, the HV release can not be described as "stable".  After
HV's Cinelerra is released, there are often bugs or unusable new features.
When there is a new release, a CV member (j6t) merges HV's code with Cinelerra
CV code, taking the enhancements from HV and reformatting the CV code (white
spaces, function naming, directory naming) to be more similar to HV's with
slight changes to implementations.  After the merge, the latest Cinelerra CV
release is a little unstable as users find bugs.  Time permitting, the CV
programmers will address as many of these bugs as possible.  In this way,
Cinelerra CV can be seen as the community's attempt to stabilize HV's release. 

As mentioned, the community adds new enhancements to the HV source.  Members
will comment on each other's implementations in order to create a more fully
functional and stable product.  Occasionally, HV will give feedback on
implementations that the members of the CV submit to it.  However, not all of
the enhancements that the community create make it upstream; for example, YUV
pipe rendering.

Given the above discussion, obtaining the SVN just before a merge will
generally be more stable than a post-merge CV version.  Be aware that existing
project description files, or Edit Decision Lists (discussed below), may not be
compatible with the newly merged CV version.  With any version of Cinelerra,
the task of finding bugs is relatively easy.  However, clearly and concisely
documenting these bugs for the community that fixes them is a task that we ask
of all users of the software.  The community is very responsive.  Please help
them by creating well-formed bug reports.  You may join our mailing list at
@uref{http://cvs.cinelerra.org}.

@c cincvdoc_node_number_5
@node About this manual
@section About this manual
@cindex Manual, about this

This manual edition is @value{EDITION}, for Cinelerra CV version
@value{VERSION}.  You may redistribute it and/or modify it under the terms of
the GNU General Public License, as published by the Free Software Foundation;
either version 2 of the License, or (at your option) any later version.

This manual originates from "Secrets of Cinelerra", an excellent primer written
by @w{Adam @sc{Crossfire}} from @w{@sc{Heroine Virtual Ltd}}.  In 2003 Alex @sc{Ferrer} created a Wiki based on
that manual and added many screenshots and topic descriptions.  At that
time, Cinelerra CV still did not have its own manual and information regarding
the Community Version of Cinelerra was scattered across the Internet
(mailing-list, IRC, websites, wiki, etc).  In 2006, Nicolas @sc{Maufrais}
combined the original "Secrets of Cinelerra" with the contents from Alex @sc{Ferrer}'s Wiki
into a unified document.

Cinelerra-CV documentation maintainers: @*
English: Nicolas @sc{Maufrais} (coordinator) @*
French: Jean-Luc @sc{Coulon}

Other contributors: Alexandre @sc{Bourget}, Kevin @sc{Brosius}, Carlos
@sc{Davila}, Rafael @sc{Diniz}, Pierre @sc{Dumuid}, Mike @sc{Edwards}, Martin
@sc{Ellison}, Scott @sc{Frase}, Joe @sc{Friedrichsen}, Gus Gus, Ben
@sc{Jorden}, Nathan @sc{Kidd}, Marcin @sc{Kostur}, Valentina @sc{Messeri},
Herman @sc{Robak}, Dana @sc{Rogers}, Jim @sc{Scott}, Andraz @sc{Tori}, Raffaella
@sc{Traniello}.

Thanks to the GNU project team, and particularly to Karl @sc{Berry}, maintainer
of GNU Texinfo, for the precious help he gave us during the elaboration of this
manual.

@itemize
@item To fetch the manual sources, install cogito and git-core on your computer and
run: @*
@command{cg-clone git://scm.pipapo.org/cinelerra/nicolasm}
@item You can participate on editing this manual by making changes in the
Cinelerra-CV wiki: @*
@uref{http://cvs.cinelerra.org/docs/wiki/doku.php}
@end itemize

@ifnotplaintext
@ifnothtml
@ifnotdocbook

@b{Note:} This manual is intended to be duplex-printed.  Therefore, it is
normal in the PDF manual for some even pages to be left blank.
@end ifnotdocbook
@end ifnothtml
@end ifnotplaintext

@c cincvdoc_node_number_6
@node Getting help
@section Getting help
@cindex Getting help
@cindex Help, getting

Help can be found at:
@itemize @bullet
@item @b{IRC channel:} #cinelerra on Freenode
@item @b{Mailing list:} @uref{https://init.linpro.no/mailman/skolelinux.no/listinfo/cinelerra}
@item @b{Cinelerra CV website:} @uref{http://cvs.cinelerra.org}
@end itemize

@c cincvdoc_node_number_327
@node Tutorials
@section Tutorials
@cindex Tutorials

Some tutorials are available on the internet:
@itemize @bullet
@item @b{Cinelerra Tutorial - Getting Started}, by Rob @sc{Fisher}@*
@uref{http://www.robfisher.net/video/cinelerra1.html}
@item @b{Guide d'utilisation de Cinelerra}, in French@*
@uref{http://www.funix.org/fr/linux/cinelerra.htm}
@item @b{Cinelerra video tutorials}, by The Source@*
@uref{http://www.thesourceshow.org/node/11}@* 
The first tutorial is in Episode 6 - "The Return Of The Pixel".@*
The second tutorial is in Episode 1 - "The Filesystem Menace".
@end itemize

@c cincvdoc_node_number_7
@node Installation
@chapter Installation

@cindex Installation
This is the general contents of all Cinelerra packages.

@itemize @bullet
@item @b{Foreign language translations} - These go into
@file{/usr/share/locale}
@item @b{Cinelerra executable} - This goes into @file{/usr/bin}
@item @b{Cinelerra plugins} - These go into @file{/usr/lib/cinelerra} in 32 bit
systems and @file{/usr/lib64/cinelerra} in 64 bit systems.
@cindex Soundtest
@cindex Sound card buffer size
@item @b{soundtest} - Utility for determining sound card buffer size.
@cindex mplexlo
@item @b{mplexlo} - Multiplexing of MPEG elementary streams without
standards conformance but more efficiently.
@cindex mpeg3cat
@item @b{mpeg3cat} - Utility for reading an MPEG file from a certain
standard and outputting it to stdout.
@cindex mpeg3toc
@cindex mpeg3dump
@item @b{mpeg3toc, mpeg3cat, mpeg3dump} - Utilities for indexing and
reading MPEG files.
@cindex mpeg3peek
@item @b{mpeg3peek} - Utility for displaying the byte offset of a
frame in an MPEG file.
@end itemize

@menu
* Hardware requirements::
* Software requirements::
* Installing an RPM::
* Compiling Cinelerra CV::
* Running Cinelerra::
* Debian::
* Ubuntu::
* Gentoo::
* Knoppix::
* Redhat::
* Mandrake::
* Slackware::
* Suse::
* MacOSX::
@end menu

@c cincvdoc_node_number_8
@node Hardware requirements
@section Hardware requirements
@cindex Hardware requirements
@cindex Requirements, hardware

Cinelerra is demanding on all PC subsystems, as reading, decoding and playing
video can be quite taxing.  Thus, performance and usability of Cinelerra are
directly proportional to the video format (SVCD/DV/HDV/HD/etc) used and the CPU
and I/O bus speeds and video and memory bus architecture of your hardware.
Therefore, it stands to reason that a less powerful system will be sufficient
for users working with audio only or lower resolution video formats.  However,
that same system may slow down considerably when playing back a higher
resolution format, such as DV video.  Effects and several tracks of audio will
compound these problems.  Given these constraints, here are some suggestions
for running Cinelerra:

@itemize @bullet
@item @b{CPU speed} @*
At least 500 MHz CPU speed, anything less would be useless.  Dual-core and
SMP processors greatly improve Cinelerra speed.
@item @b{Memory} @*
When working with video, a large amount of free memory available can help speed
up operations by avoiding unnecessary disk swaps and keeping material ready
accessible.  Have at least 256 Megabytes of memory.  To really use Cinelerra for
higher resolution video formats and larger projects, greater than 1 Gb memory space
is suggested.
@item @b{Storage} @*
Video editing can be quite I/O intensive.  Storage requirements are based on
your particular video editing needs.  If you expect to produce long pieces in
uncompressed or larger resolution formats, you should have large (>200 Gb) and
fast (<10ms) disk drives.  For example, DV uses about 3.5 Megs per second, or
12 Gigs per hour.  For smaller projects you might get away with 1 Gb.  RAID0
(stripe set), RAID1+0 (striped/mirrored) or RAID5 (stripe set with parity) will
also speed playback
@item @b{Video adapters} @*
Since version 2.1, Cinelerra benefits from OpenGL hardware
acceleration.  Make sure the video card you use supports OpenGL 2.0 in order to
benefit from that acceleration.  Nvidia series 7 (ie. 7600GS) are known to work
well.  Unfortunately, ATI's Linux drivers do not support a complete
implementation of OpenGL 2.0.  If you are going to send a composite signal
directly to a TV or video recorder, make sure your video card supports it. 
@item @b{Multiple monitors} @*
You can use XFree86's Xinerama features to work on multiple monitor heads.
This feature can be a very effective way of increasing productivity.
@item @b{TV-Out} @*
If your Adapter supports a TV-Out option, connecting a TV or S-Video
monitor to it is a great way to view your material as it will be seen on TV
screen.
@item @b{Video grabbers} @*
If you have an analog video camera, or want to grab video from a trusty
old VCR, you need some sort of video grabber.  Video grabbers are supported
through Video4Linux in Cinelerra.
@item @b{Firewire} @*
Firewire is the fastest way to download material into your system.  Unless
you will be importing your media from a CD, any other
pre-captured format or use an analog video grabber, you will need firewire
on your system. 
@item @b{DV cameras} @*
There is an large variety of DV cameras that can be used with Cinelerra.
Almost any camera that can connect using firewire will work.
Be sure to set the appropriate parameters on the video grabbing system to match
your particular camera.  @uref{http://www.linux1394.org} has a complete listing
of supported cameras.
@end itemize

@c cincvdoc_node_number_9
@node Software requirements
@section Software requirements
@cindex Software requirements
@cindex Requirements, software

To install Cinelerra you should have a current version of GNU/Linux with XFree86
and some audio management software properly running.  You should also have the
following libraries installed (partial list):
@itemize @bullet
@item a52dec
@item dv
@item faac
@item ffmpeg
@item fftw
@item lame
@item libavc1394
@item libfaad2
@item libraw1394
@item mjpegtools
@item OpenEXR
@item theora
@item x264
@end itemize

@c cincvdoc_node_number_10
@node Installing an RPM
@section Installing an RPM
@cindex Installing an RPM
@cindex RPM, installing an
@cindex Fedora
The easiest way to install Cinelerra is by downloading and installing an RPM
using the following command @*
@command{rpm -U --force --nodeps hvirtual*.rpm} @*
on a Fedora 4 system.

@cindex rpm2cpio
On systems that do not support RPM, look for a utility called @b{rpm2cpio}.
Download a Cinelerra RPM and from the @file{/} directory run @*
@command{rpm2cpio hvirtual*.rpm | cpio -i --make-directories} @*
This does not always work because there are many forks of the C library, each
incompatible with the others.  This is the biggest reason to compile from
scratch.

@c cincvdoc_node_number_11
@node Compiling Cinelerra CV
@section Compiling Cinelerra CV
@cindex Compiling Cinelerra CV
@cindex Cinelerra, compiling

@menu
* Usual compilation process::
* Compiling with debugging symbols::
@end menu

@c cincvdoc_node_number_12
@node Usual compilation process
@subsection Usual compilation process
@cindex Usual compilation process

You can install Cinelerra CV by fetching the sources and compiling them.  That
the method to use if you want to compile the most up-to-date version of
Cinelerra CV@.

@enumerate 1
@item First you have to fetch Cinelerra CV's sources from the SVN repository
(approximately 170Mb).  Run: @*
@command{svn checkout svn://svn.skolelinux.org/cinelerra/trunk/hvirtual} @*
If you already fetched the sources of an out of date revision, you can update
to the latest revision from within the @file{hvirtual} folder by running: @*
@command{svn update} @*
If you wish to fetch an old revision, run: @*
@command{svn checkout -r <revision>
svn://svn.skolelinux.org/cinelerra/trunk/hvirtual}

@item Go in the hvirtual folder: @*
@command{cd hvirtual}

@item Create the @file{./configure} file by running: @*
@command{autoreconf -i --force}

@item Then run the @file{.configure} file: @*
@command{./configure --with-buildinfo=svn/recompile} @*
You can have a look at all the options available by running: @*
@command{./configure --help}

@item And run make: @*
@command{make}

@item Finally, install Cinelerra CV: @*
@command{sudo make install}
@end enumerate

@b{Notes:}
@itemize @bullet
@item @b{SMP machine:} @*
If you compile Cinelerra CV on a multiprocessor machine (SMP), we recommend you
to add the @option{-j 3} option to make in order to benefit from the available
CPUs.
@item @b{For x86 CPUs only}: @*
You probably want to enable MMX support.  To do that, run @command{./configure}
with the @option{--enable-mmx} option.  If you do that, you may have to use the
@option{--without-pic} option too, otherwise, compilation may fail.
@item @b{For Pentium-M:} @*
Here are some useful compiler flags: @*
@command{./configure --prefix=/usr --enable-x86 --enable-mmx --enable-freetype2
--with-buildinfo=svn/recompile CFLAGS='-O3 -pipe -fomit-frame-pointer
-funroll-all-loops -falign-loops=2 -falign-jumps=2 -falign-functions=2
-ffast-math -march=pentium-m -mfpmath=sse,387 -mmmx -msse'}
@item @b{Installing several revisions:} @*
If you wish to install several revisions of Cinelerra CV on your computer,
create a @file{/usr/local_cinelerra} folder, and use the following options with
@command{./configure} (replace @option{xxx} by the number of the revision you
are compiling): @*
@option{--prefix=/usr/local_cinelerra/rxxx
--exec-prefix=/usr/local_cinelerra/rxxx --program-suffix=_rxxx} @*
If you install Cinelerra using this method, the translated @file{.po} files do
not get installed correctly.  Go in the @file{hvirtual} directory where the sources
are and run: @*
@command{./configure prefix=/usr} @*
@command{cd po} @*
@command{sudo make install} @*
You will have to run Cinelerra CV from the directory in which it is installed: @*
@command{cd /usr/local_cinelerra/r960} @*
@command{./cinelerra_r960}
@item @b{Automake version:} @*
You need automake version 1.7 to build.  1.4 will not work.  Autoconf 2.57 is
also required to build.
@end itemize

@c cincvdoc_node_number_13
@node Compiling with debugging symbols
@subsection Compiling with debugging symbols
@cindex Compiling with debugging symbols
@cindex Debugging symbols, compiling with

When Cinelerra CV crashes, one can compile it with debugging symbols and run it
inside gdb.  The information displayed by gdb is far more detailed and
will help CV developers find bugs faster.

First, fetch the SVN sources as usual.  Then, run the following commands: @*
@command{cd hvirtual} @*
@command{nice -19 autoreconf -i --force} @*
@command{mkdir ../hvdbg} @*
@command{cd ../hvdbg} @*
@command{nice -19 ../hvirtual/configure CXXFLAGS='-O0 -g' CFLAGS='-O0 -g'
--with-buildinfo=svn/recompile} @*
@command{cd quicktime/ffmpeg} @*
@command{nice -19 make CFLAGS='-O3'} @*
@command{cd ../..} @*
@command{nice -19 make} @*
@command{nice -19 make install}

@xref{Reporting bugs}, for information about running Cinelerra inside gdb.

@c cincvdoc_node_number_14
@node Running Cinelerra
@section Running Cinelerra
@cindex Cinelerra, running

The simplest way to run Cinelerra is by running @command{/usr/bin/cinelerra} @*
Command line options are also available by typing @command{cinelerra -h}  These
options are described in several sections below.  For rendering from the
command line @xref{Rendering files}.

@c cincvdoc_node_number_15
@node Debian
@section Debian
@cindex Debian

@menu
* Debian binaries::
* Debian prerequisites::
@end menu

@c cincvdoc_node_number_16
@node Debian binaries
@subsection Debian binaries
@cindex Binaries, Debian

Andraz @sc{Tori} maintains build rules for Debian Sid.  He also makes binary
.deb packages for Sid.  They are built from the unofficial SVN releases.
Debian Sid packages can be found here:
@itemize @bullet
@item @b{Apt source for i386:} @*
@command{@w{deb http://www.kiberpipa.org/~minmax/cinelerra/builds/sid/ ./}}
@item @b{Apt source for Pentium4 (optimized):} @*
@command{@w{deb http://www.kiberpipa.org/~minmax/cinelerra/builds/pentium4/ ./}}
@item @b{Apt source for Pentium-M (optimized):} @*
@command{@w{deb http://www.kiberpipa.org/~minmax/cinelerra/builds/pentiumm/ ./}}
@item @b{Apt source for AthlonXP (optimized):} @*
@command{@w{deb http://www.kiberpipa.org/~minmax/cinelerra/builds/athlonxp/ ./}}
@item @b{Apt source for Athlon64 (optimized):} @*
@command{@w{deb http://labbs.net/~vale/debian ./}}
@end itemize

@b{Note:} If Cinelerra produces the following error: @*
@command{cinelerra: relocation error: /usr/lib/libavcodec.so.0.4.8:
undefined symbol: faacDecOpen} @*
You can solve the problem by entering the following command as root: @*
@command{apt-get install --reinstall libfaad2-0=2.0.0-0.5}

@c cincvdoc_node_number_17
@node Debian prerequisites
@subsection Debian prerequisites
@cindex Prerequisites, Debian

You need some prerequisites which are not found in Debian's official
repositories.  You should add in your @file{/etc/apt/sources.list} the following
line, which is Christian Marillat's repository: @*
@command{deb http://www.debian-multimedia.org/ sid main}

In order to use the mirror you need to add in your gpg keyring Marillat's
gpg-key: @*
@command{gpg --keyserver hkp://wwwkeys.eu.pgp.net --recv-keys 1F41B907} @*
@command{gpg --armor --export 1F41B907 | sudo apt-key add -} @*
If you do not use sudo, do the following under root: @*
@command{gpg --armor --export 1F41B907 | apt-key add -}

@c cincvdoc_node_number_325
@node Ubuntu
@section Ubuntu
@cindex Ubuntu

FIXME

@c cincvdoc_node_number_18
@node Gentoo
@section Gentoo
@cindex Gentoo

Installation for Gentoo GNU/Linux is very straight forward.  Simply type: @*
@command{emerge cinelerra} @*
as root and it should install and run with no problems.

@c cincvdoc_node_number_19
@node Knoppix
@section Knoppix
@cindex Knoppix

Knoppix is a bootable CD with a collection of GNU/Linux software, automatic
hardware detection, and support for many graphics cards, sound cards, SCSI and
USB devices and other peripherals.  Knoppix can be used as a GNU/Linux demo,
educational CD, rescue system, or adapted and used as a platform for commercial
software product demos.  It is not necessary to install Knoppix on a hard disk.
(from @uref{http://www.knoppix.org})

Known Knoppix distributions which include Cinelerra:
@itemize @bullet
@item @uref{http://www.dynebolic.org}
@item @uref{http://linux.slo-tech.com}
@item @uref{http://garbure.org/pho}
@end itemize

@c cincvdoc_node_number_326
@node Redhat
@section Redhat
@cindex Redhat

FIXME

@c cincvdoc_node_number_20
@node Mandrake
@section Mandrake
@cindex Mandrake

Perhaps the greatest challenge to compiling Cinelerra on Mandrake is gathering
all the proper lib dependencies.  Once all the required libraries (and their
devel packages) are installed the compile process works smoothly.

@c cincvdoc_node_number_21
@node Slackware
@section Slackware
@cindex Slackware

Rafael @sc{Diniz} build Slackware packages for Cinelerra.

@itemize @bullet
@item @b{For x86:} @*
@uref{http://slack.sarava.org/packages/slackware/slackware-11.0/multimedia/}
@item @b{For slackintosh:} @*
@uref{http://slack.sarava.org/packages/slackintosh/slackintosh-11.0/multimedia/}
@end itemize

@c cincvdoc_node_number_22
@node Suse
@section Suse
@cindex Suse

RPMs for SuSE 9 are built from CVS by Kevin @sc{Brosius}, and available at
@uref{http://cin.kevb.net/files/RPM/}

@c cincvdoc_node_number_23
@node MacOSX
@section MacOSX
@cindex MacOSX

FIXME

@c cincvdoc_node_number_24
@node Configuration
@chapter Configuration
@cindex Configuration

Because of its flexibility, Cinelerra cannot be optimized without
special configuration for your specific needs.  Unfortunately, very few parameters are
adjustable at compile time.  Therefore, runtime configuration is the only option for most
users because of the multitude of parameters available. @*
Below are configuration options as well as the supported API's in GNU/Linux. @*
In Cinelerra, go to @b{settings->preferences} to see the options.

@menu
* Environment variables::  These environment variables are recognized by Cinelerra
* Audio drivers::          Information about the audio drivers
* Video drivers::          Information about the video drivers
* Playback::               Configuring parameters related to playback.
* Recording::              Configuring parameters related to recording.
* Performance::            Configuring parameters related to how fast things go.
* Interface::              Configuring the user interface.
* About::                  Viewing information about the program.
@end menu

@c cincvdoc_node_number_25
@node Environment variables
@section Environment variables
@cindex Environment variables
@cindex Variables, environment
@cindex Ladspa, path

In UNIX derivatives, environment variables are global variables in the shell
which all applications can read.  They are set with a command like @command{set
VARIABLE=value}.  All the environment variables can be viewed with a command
like @command{env}.  Cinelerra recognizes the following environment variables:

@itemize @bullet
@item @b{LADSPA_PATH} @*
If you want to use LADSPA plugins, this must be defined: a colon separated
list of directories to search for LADSPA plugins.  These are not native
Cinelerra plugins.  @xref{Ladspa effects}.

@item @b{GLOBAL_PLUGIN_DIR} @*
The directory in which Cinelerra should look for native plugins.  The default is
@file{/usr/lib/cinelerra} but you may need an alternate directory if you share
the same executable directory among many machines via NFS@.  Plugins of
different binary formats need to be in different directories.

@item @b{LANG} @*
Cinelerra has been translated into many languages.  Cinelerra language settings
are normally read from your GNU/Linux language settings.  To run Cinelerra on a
language different than the one selected on your system just change the LANG
environment variable. @*
For example, open a shell and type: @command{export LANG=es_ES}, then run
cinelerra from the same shell.  It will open set on the Spanish language. @*
Available languages are:
@itemize @bullet
@item es_ES - Espanol
@item sl_SI - Slovenian
@item fr_FR - Francais
@item eu_ES - Euskera
@item de_DE - German
@item pt_BR - Brazilian Portuguese
@end itemize
If you installed Cinelerra CV by compiling the sources, and you specified a
@option{--prefix=} option different from @file{/usr/local}, the translated
files were probably not installed.  @xref{Usual compilation process}, for more
information.
@end itemize

@c cincvdoc_node_number_26
@node Audio drivers
@section Audio drivers
@cindex Audio drivers
@cindex Drivers, audio

The audio drivers are used for both recording and playback.  Their
functionality is described below:

@menu
* Common sound driver attributes:: Attributes used for more than one sound driver.
* OSS:: Notes about the OSS driver
* OSS Envy24:: Notes about the OSS driver for the Envy24 chip
* Alsa:: Notes about the ALSA driver
* Esound:: Notes about the ESound driver
* Raw 1394:: Notes about the Raw1394 driver
* DV 1394:: Notes about the DV1394 driver
* IEC 61883:: Notes about the IEC 61883 driver
@end menu

@c cincvdoc_node_number_27
@node Common sound driver attributes
@subsection Common sound driver attributes
@cindex Common sound driver attributes

@itemize @bullet
@item @b{Device path}  @*
Usually a file in the @file{/dev/} directory which controls the device.

@item @b{Bits} @*
The number of bits of precision Cinelerra should set the device for.  This
sometimes has a figurative meaning.  Some sound drivers need to be set to 32
bits to perform 24 bit playback and will not play anything when set to 24 bits.
Some sound drivers need to be set to 24 bits for 24 bit playback.

@item @b{Channels} @*
The number of channels Cinelerra should set the device for.  Regardless of
the number of channels in the project, the number of channels set here will be
written to the device.  When this is set to 2 and the project has 1 channel you
will hear sound through the left speaker and not centered as expected for a
monaural project.  When this is set to 1 and the project has 2 channels you
will hear the left channel centered and not 2 channels mixed together.
@end itemize

@c cincvdoc_node_number_28
@node OSS
@subsection OSS
@cindex OSS

This was the first GNU/Linux sound driver.  It had an open source
implementation and a commercial implementation with more sound cards supported.
It was the standard sound driver up to GNU/Linux 2.4.  It still is the only
sound driver which an i386 binary can use when running on an x86_64 system.

@c cincvdoc_node_number_29
@node OSS Envy24
@subsection OSS Envy24
@cindex OSS Envy24
@cindex Envy24

The commercial version of OSS had a variant for 24 bit 96 KHz soundcards.  This
variant required significant changes to the way the sound drivers were used,
hence the need for the new driver.

@c cincvdoc_node_number_30
@node Alsa
@subsection Alsa
@cindex Alsa

ALSA is the most common sound driver in GNU/Linux 2.6.  It supports most
of soundcards now.  It takes advantage of low latency features in GNU/Linux 2.6
to get better performance than OSS had in 2.4, but roughly the same performance
that OSS had in 2.0.  Unfortunately ALSA is constantly changing.  A program
which works with it one day may not the next day.  New wrappers are being
developed on top of ALSA.  We plan to support them at regular
intervals, though not at every new release of a new wrapper. @*
ALSA is no longer portable between i386 and x86_64.  If an i386 binary tries to
play back on an x86_64 kernel, it will crash.  For this scenario, use OSS@.

@c cincvdoc_node_number_31
@node Esound
@subsection Esound
@cindex Esound

ESOUND was a sound server that sat on top of OSS@.  It was written for a window
manager called Enlightenment.  It supports a limited number of bits and has
high latency compared to more modern drivers, but it does have the ability to multiplex multiple audio sources.
It is unknown whether it still works.

@c cincvdoc_node_number_32
@node Raw 1394
@subsection Raw 1394
@cindex Raw 1394

The was the first interface between GNU/Linux software and firewire camcorders.
It is the least reliable way to play audio to a camcorder and consists of
a library on top of the kernel commands.

@c cincvdoc_node_number_33
@node DV 1394
@subsection DV 1394
@cindex DV 1394

This is the second rewrite of DV camcorder support in GNU/Linux.  This is the most
reliable way to play audio to a camcorder and consists of direct kernel
commands.

@c cincvdoc_node_number_34
@node IEC 61883
@subsection IEC 61883
@cindex IEC 61883

The third rewrite of DV camcorder support in GNU/Linux.  This is a library on
top of RAW 1394 which is a library on top of the kernel commands.  It is less
reliable than DV 1394 but more reliable than RAW 1394.  The next rewrite ought
to fix that.  Visit @uref{http://www.linux1394.org} for more information and
the latest drivers.

@c cincvdoc_node_number_35
@node Video drivers
@section Video drivers
@cindex Video drivers
@cindex Drivers, video

The video drivers are used for video playback in the compositor and the viewer.

@menu
* Common video driver attributes:: Parameters used by more than one driver.
* X11::
* X11-XV::
* X11-OpenGL::
* Buz::
* Raw 1394 video playback::
* DV 1394 video playback::
* IEC 61883 video playback::
@end menu

@c cincvdoc_node_number_36
@node Common video driver attributes
@subsection Common video driver attributes
@cindex Common video driver attributes
@cindex Video driver, common attributes

@itemize @bullet
@item @b{Display} @*
@cindex Dual monitor displays
The interface is intended for dual monitor displays.  Depending on the value of
Display, the Compositor window will appear on a different monitor from the rest
of the windows.

@item @b{Device path} @*
Usually a file in the @file{/dev/} directory which controls the device.

@item @b{Swap fields} @*
Make the even lines odd and the odd lines even when sending to the device.  On
an NTSC or 1080i monitor the fields may need to be swapped to prevent jittery
motion.

@item @b{Output channel} @*
You may need a specific connector to send video out to devices with multiple outputs.

@item @b{Port} @*
The IEEE1394 standard specifies something known as the @b{port}.  This is
probably the firewire card number.

@item @b{Channel} @*
The IEEE1394 standard specifies something known as the @b{channel}.  For DV
cameras it is always @b{63}.
@end itemize

@c cincvdoc_node_number_37
@node X11
@subsection X11
@cindex X11

This was the first method of graphical display on any UNIX system. 
It just writes the RGB triplet for each pixel directly to the
window.  It is the slowest playback method.  It is still useful as a fallback
when graphics hardware can not handle very large frames.

@c cincvdoc_node_number_38
@node X11-XV
@subsection X11-XV
@cindex X11-XV

This was an enhancement to X11 in 1999.  It
converts YUV to RGB in hardware with scaling.  It is the preferred playback
method but can not handle large frame sizes.  The maximum video size for XV is
usually 1920x1080.

@c cincvdoc_node_number_39
@node X11-OpenGL
@subsection X11-OpenGL
@cindex X11-OpenGL
@cindex OpenGL

The most powerful video playback method is OpenGL@.  With this driver, most
effects are done in hardware.  OpenGL allows video sizes up to the maximum
texture size, which is usually larger than what XV supports, depending on the
graphics driver.  To enable it you will need a binary built with OpenGL
support.  The @command{configure} option to enable OpenGL is
@option{--enable-opengl}.  You need a video card which supports OpenGL 2.0.
Recent Nvidia video cards should work.  You also need to use a video driver
supporting OpenGL 2.0, such as Nvidia's binary driver.  To know if your video
driver supports OpenGL 2.0, type the following command: @command{glxinfo | grep
"OpenGL version"}

@itemize @bullet
@item Video driver supporting hardware OpenGL 2.0 rendering: @*
@command{OpenGL version string: 2.0.2 NVIDIA 87.74}
@item Video driver not supporting hardware OpenGL 2.0 rendering: @*
@command{OpenGL version string: @b{1.4} (2.0.2 NVIDIA 87.74)}
@end itemize

OpenGL relies on PBuffers and shaders to do video rendering.  The graphics
driver must support OpenGL 2.0 and Cinelerra needs to be explicitly compiled with
OpenGL 2.0 support.  This requires compiling it on a system with the OpenGL 2.0
headers.  PBuffers are known to be fickle.  If the graphics card does not have
enough memory or does not have the right visuals, PBuffers will not work.  If
OpenGL does not work, try seeking several frames or restarting Cinelerra.

@b{Limitations:}
@itemize @bullet
@item OpenGL does not affect rendering.  It just accelerates playback.
@item X11-OpenGL processes everything in 8 bit color models, although the
difference between YUV and RGB is retained.
@item OpenGL does not work with frames whose size is larger than 4096x4096. @*
Here is what is written in the console when working with large frames: @*
@code{BC_Texture::create_texture frame size <frame_width>x<frame_height> bigger
than maximum texture 4096x4096.}
@item The scaling equation set in the preferences window is ignored by OpenGL@.
OpenGL always uses linear scaling.
@item Project and track sizes need to be multiples of four for OpenGL to work.
@item To get the most acceleration, OpenGL-enabled effects must be placed after
software-only effects.  All rendering before the last software-only effect is
done in software.  The core Cinelerra operations like camera and projector are
OpenGL@.
@item Not all of the effects support OpenGL acceleration. The following effects
support OpenGL: Brightness, Chromakey, Chromakeyhsv, Color balance,
Deinterlace, Diffkey, Dissolve, Flip, Frames to fields, Freeze frame, Gamma,
Gradient, Histogram, Hue saturation, Interpolate Pixels, Invert video, Linear
blur, Overlay, Perspective, Radial blur, RGB601, Rotate, Scale, Threshold,
Zoomblur.
@end itemize

@c cincvdoc_node_number_40
@node Buz
@subsection Buz
@cindex Buz
@cindex Video4Linux

This is a method for playing motion JPEG-A files directly to a composite analog
signal.  It uses a popular hack of the Video4Linux 1 driver from 2000 to
decompress JPEG in hardware.  Even though analog output is largely
obsolete, newer drivers have replaced BUZ@.

@c cincvdoc_node_number_41
@node Raw 1394 video playback
@subsection Raw 1394 video playback
@cindex Raw 1394

This was the first interface between GNU/Linux software and firewire
camcorders.  It is the least reliable way to play video to a camcorder and it
consists of a library on top of the kernel commands.

@c cincvdoc_node_number_42
@node DV 1394 video playback
@subsection DV 1394 video playback
@cindex DV 1394

The second rewrite of DV camcorder support in GNU/Linux.  This was the most
reliable way to play video to a camcorder and consists of direct kernel
commands.

@c cincvdoc_node_number_43
@node IEC 61883 video playback
@subsection IEC 61883 video playback
@cindex IEC 61883

The third rewrite of DV camcorder support in GNU/Linux.  This is a library on
top of RAW 1394 and is less
reliable than DV 1394 but more reliable than RAW 1394.  The next rewrite ought
to fix that.  Visit @uref{http://www.linux1394.org} for more information and
the latest drivers.

@c cincvdoc_node_number_44
@node Playback
@section Playback
@cindex Playback

@menu
* Audio out::
* Video out::
@end menu

@c cincvdoc_node_number_45
@node Audio out
@subsection Audio out
@cindex Audio out
@cindex Audio samples

These determine what happens when you play sound from the timeline.

@cindex Samples to send to console
@cindex Console, samples to send to
@itemize @bullet
@item @b{Samples to send to console} @*
For playing audio, small fragments of sound are read from disk and processed sequentially in
a virtual console.  A larger value here causes more latency when
you change mixing parameters but yields more reliable playback. @*
Some sound drivers do not allow changing of the console fragment, so latency is
unchanged no matter what the value. @*
Previously, a good way of ensuring high quality playback was to read bigger fragments from
the disk and break them into smaller fragments for the soundcard.  That changed
when the virtual console moved from the push model to the pull model.  Since
different stages of the rendering pipeline can change the rate of the incoming
data, it would be difficult to disconnect the size of the console fragments
from the size of the fragments read from disk.

@cindex Audio offset
@cindex Offset, audio
@item @b{Audio offset} @*
The ability to tell the exact playback position on GNU/Linux sound drivers is
poor, if it is provided at all.  Since this information is required for
proper video synchronization, it has to be accurate.  The @b{audio offset}
allows users to adjust the position returned by the sound driver in order to reflect
reality.  The audio offset does not affect the audio playback or rendering at
all.  It merely changes the synchronization of video playback. @*
The easiest way to set the audio offset is to create a timeline with one video
track and one audio track.  Expand the audio track and center the audio pan.
The frame rate should be larger than 24 fps and the sampling rate should be
greater than 32000.  The frame size should be small enough for your computer to render
it at the full framerate.  Highlight a region of the timeline starting at 10
seconds and ending at 20 seconds.  Drop a @b{gradient} effect on the video
track and configure it to be clearly visible.  Drop a @b{synthesizer} effect on
the audio and configure it to be clearly audible. @*
Play the timeline from 0 and watch to see if the gradient effect starts exactly
when the audio starts.  If it does not, expand the audio track and adjust the
nudge.  If the audio starts ahead of the video, decrease the nudge value.  If
the audio starts after the video, increase the nudge value.  Once the tracks
play back synchronized, copy the nudge value to the @b{audio offset} value in
preferences. @*
@b{Note:} if you change sound drivers or you change the value of @b{Use
software for positioning information}, you will need to change the audio offset
because different sound drivers are unequally inaccurate.

@cindex View follows playback
@cindex Playback, view follows
@item @b{View follows playback} @*
This causes the timeline window to scroll when the playback cursor moves.  This
can bog down the X Server or cause the timeline window to lock up for long
periods of time while drawing the assets.

@cindex Use software for positioning information
@cindex Positioning information, use software for
@item @b{Use software for positioning information} @*
Most soundcards and sound drivers do not give reliable information on the number
of samples the card has played.  You need this information
for synchronization when playing back video.  This option causes the sound driver to be ignored and a
software timer to be used for synchronization.

@cindex Audio playback in realtime
@cindex Realtime, audio playback in
@item @b{Audio playback in realtime} @*
Back in the days when 150 MHz was the maximum speed for a personal computer, this setting allowed uninterrupted
playback during periods of heavy load.  It forces the audio playback to the highest priority
in the kernel.  Today, it is most useful for achieving very low latency between
console tweeks and soundcard output.  You must be root to get real-time
priority.

@cindex Audio driver
@item @b{Audio driver} @*
There are many sound drivers for GNU/Linux.  This allows selecting one sound
driver and setting parameters specific to it.  The sound drivers and their
parameters are described in the sound driver section.  @xref{Audio drivers}.
@end itemize

@c cincvdoc_node_number_46
@node Video out
@subsection Video out
@cindex Video out

These determine how video gets from the timeline to your eyes.

@cindex Play every frame
@cindex Frame, play every
@itemize @bullet
@item @b{Play every frame} @*
This causes every frame of video to be displayed even if it means that the playback of the audio tracks(s) will fall
behind.  This option should always be enabled unless you use uncompressed
codecs.  As of 1/2007, most compressed codecs do not support frame dropping anymore.

@cindex Framerate achieved
@item @b{Framerate achieved} @*
The number of frames per second being displayed during playback.  This is
updated during playback only.

@cindex Decode frames asynchronously
@item @b{Decode frames asynchronously} @*
If you have lots of memory and more than one CPU, this option can improve
playback performance by decoding video on one CPU as fast as possible while
dedicating the other CPU to playing back video.  It assumes all playback
operations are forward and no frames are dropped.  Operations involving reverse
playback or frame dropping are negatively impacted. @*
Since this option requires enormous amounts of memory, Cinelerra may crash if the
input frames are very large.

@cindex Scaling equation
@cindex Equation, scaling
@item @b{Scaling equation} @*
This algorithm is used when video playback involves any kind of scaling or
translation.  This does not affect 1:1 playback.
@itemize @bullet
@item @b{Nearest neighbor enlarge and reduce} @*
Low quality output with fast playback.  Produces jagged edges and uneven motion.
@item @b{Bicubic enlarge and bilinear reduce} @*
High quality output with slow playback.  Bicubic interpolation is used for enlarging,
which blurs slightly but does not show stair step artifacts.  A bilinear
interpolation is used for reduction, which produces very sharp images and reduces noise.
Bilinearly reduced images can be sharpened with a sharpen effect with less noise
side effects than a normal sized image.
@item @b{Bilinear enlarge and bilinear reduce} @*
When slight enlargement is needed, a bilinear enlargement looks better than a
bicubic enlargement.
@end itemize

@cindex Preload buffer for Quicktime
@cindex Quicktime, preload buffer for
@item @b{Preload buffer for Quicktime} @*
The Quicktime/AVI decoder can handle DVD sources better when this is set to around
10000000.  This reduces the amount of seeking required.  When
reading high bitrate sources from a hard drive, this tends to impair performance by slowing down decoding.
For normal use this should be 0.

@cindex DVD subtitles
@cindex Subtitles, DVD
@item @b{DVD subtitle to display} @*
DVD IFO files usually contain subtitle tracks.  These must be decoded with
the MPEG decoder.  Select @b{Enable subtitles} to enable subtitle decoding.
There are usually multiple subtitle tracks indexed by number and starting from
0.  Enter the index number of the subtitle track to be decoded in the "DVD
Subtitle to display" text box or use the tumbler to increase the index value.
Go to the asset corresponding to the MPEG file in the Resources window and
right click.  Click on Info.  The number of subtitle tracks is shown at the
bottom.

@cindex CR2 images
@cindex Images, CR2
@item @b{Interpolate CR2 images} @*
Enables interpolation of CR2 images.  Interpolation is required since the raw image in a
CR2 file is a Bayer pattern.  The interpolation uses dcraw's built-in
interpolation and is very slow.  This operation can be disabled and the
@b{Interpolate Pixels} effect used instead for faster previewing.

@cindex White balance, CR2 images
@item @b{White balance CR2 images} @*
This enables white balancing for CR2 images if interpolation is also enabled.
This is because proper white balancing needs a blending of all 3 primary colors.
White balance uses the camera's matrix which is contained in the CR2 file. @*
Disabling white balancing is useful for operations involving dark frame
subtraction.  The dark frame and the long exposure need to have the same color
matrix. @*
If you disable @b{Interpolate CR2 Images} and use the @b{Interpolate Pixels}
effect, be aware the @b{Interpolate Pixels} effect always does both
interpolation and white balancing using the camera's matrix, regardless of the
settings in Preferences.  Dark frame subtraction needs to be performed before
@b{Interpolate Pixels}.

@cindex Video device driver
@item @b{Video driver} @*
Normally video on the timeline goes to the compositor window during both continuous
playback and when the insertion point is repositioned.  Instead of sending
video to the Compositor window, the video driver can be set to send video to
another output device during continuous playback.  However, this does not affect where
video is routed when the insertion point is repositioned. @*
The video drivers and their parameters are described in the video drivers
section.  @xref{Video drivers}.
@end itemize

@c cincvdoc_node_number_47
@node Recording
@section Recording
@cindex Recording

The parameters here expedite the @b{File->Record...} function by allowing the
user to pre-configure the file format.  The file format is applied to all
recordings.  Also set here is the hardware for recording, since the hardware
determines the supported file format (in most cases).

@menu
* File format::
* Audio in::
* Video in::
@end menu

@c cincvdoc_node_number_48
@node File format
@subsection File format
@cindex File format
@cindex Format, file

This determines the output file format for recordings.  It depends heavily on
the type of driver used.  The menu selections are the same as the rendering interface.  See @xref{Rendering files}.
The @b{Record audio tracks} toggle must be enabled to record audio.  The
@b{Record video tracks} toggle must be enabled to record video.  The wrench
button left of each toggle opens a configuration dialog in order to set the compression scheme (codec)
for each audio and video output stream.  The audio and video is wrapped in a container format
defined by the @b{File Format} menu.  Different wrappers may record audio only,
video only, or both.

Some video drivers can only record to a certain container.  DV, for example, can
only record to Quicktime with DV as the video compression scheme.  If the video driver
is changed, the file format may be updated to give the supported output.  If
you change the file format to an unsupported format, it may not work with the
video driver.

@c cincvdoc_node_number_49
@node Audio in
@subsection Audio in
@cindex Audio in

These determine what happens when you record audio.

@cindex Record driver
@cindex Driver, record
@cindex Device path
@cindex Bits
@itemize @bullet
@item @b{Record driver} @*
This is used for recording audio in the Record window.  It may be configured the same as
the Record Driver for video if the audio and video are wrapped in the same
stream.  Available parameters vary depending on the driver.  Note that the drivers
are the same as those available in Preferences->Playback.
@itemize @bullet
@item @b{Device path} @*
This is usually a file in the @file{/dev/} directory that controls the device.
@item @b{Bits} @*
The number of bits of precision Cinelerra should set the device for.  This
sometimes has a figurative meaning.  Some sound drivers need to be set to 32
bits to perform 24 bit recording and will not record anything when set to 24 bits.
Some sound drivers need to be set to 24 bits for 24 bit recording.
@item @b{Channels} @*
The number of channels Cinelerra should set the device for.  Regardless of
the number of channels in the project, the number of channels set here will be
written to the device.  When this is set to 2 and the project has 1 channel you
will hear sound through the left speaker and not centered as expected for a
monaural project.  When this is set to 1 and the project has 2 channels you
will hear the left channel centered and not 2 channels mixed together.
@end itemize

@cindex Samples to write at a time
@item @b{Samples to write at a time} @*
First, audio is read in small fragments from the device.  Then, many small fragments
are combined into a large fragment before writing to disk.  The disk writing
process is done in a different thread.  The value here determines how large the
combination of fragments is for each disk write.

@cindex Sample rate for recording
@cindex Recording, sample rate for
@item @b{Sample rate for recording} @*
Regardless of what the project settings are, the value set here will be the sample rate used for
recording.  The sample rate should be set to the highest value the audio device supports.
@end itemize

@c cincvdoc_node_number_50
@node Video in
@subsection Video in
@cindex Video in

These determine what happens when you record video.

@cindex Record driver
@cindex Driver, record
@itemize @bullet
@item @b{Record driver} @*
This is used for recording video in the Record window.  It may be configured the same as
the Record Driver for video if the audio and video are wrapped in the same
container.  Available parameters vary depending on the driver.  Note that the drivers available
are the as those available in Preferences->Playback.

@cindex Frames to record to disk at a time
@item @b{Frames to record to disk at a time} @*
Frames are recorded in a pipeline.  First, frames are buffered in the device.
Then, they are read into a larger buffer for writing to disk.  The disk writing
is done in a separate thread from the device reading.  For certain codecs the
disk writing uses multiple processors.  The value set here determines how many frames
are written to disk at a time.

@cindex Frames to buffer in device
@item @b{Frames to buffer in device} @*
This is the number of frames to store in the device before reading and determines
how much latency there can be in the system before frames are dropped.

@cindex Use software for positioning information
@item @b{Use software for positioning information} @*
Video uses audio for synchronization, but most soundcards do not give accurate
position information.  Selecting this options makes Cinelerra calculate an estimation of audio position in
software instead of hardware for synchronization.

@cindex Sync drives automatically
@item @b{Sync drives automatically} @*
For high bitrate recording, the disk drives you use may be fast enough to store the data but
your operating system may wait several minutes and stall as it writes several minutes of
data at a time.  This forces the operating system to flush its buffers every second
instead of every few minutes to produce slightly better real-time behavior.

@cindex Size of captured frame
@cindex Captured frame, size of
@item @b{Size of captured frame} @*
This is the size of the recorded frames in pixels.  It is independent of the project
frame size because most video devices only record a fixed frame size.  If the
frame size given here is not supported by the device, Cinelerra may crash.

@cindex Frame rate for recording
@cindex Recording, frame rate for
@item @b{Frame rate for recording} @*
The frame rate recorded is different from the project settings.  This sets the
recorded frame rate.
@end itemize

@c cincvdoc_node_number_51
@node Performance
@section Performance
@cindex Performance

You will spend most of your time configuring this section.  The main focus of
the performance section is rendering parameters not available in the rendering dialog.

@cindex Cache items
@itemize @bullet
@item @b{Cache items} @*
To speed up rendering, several assets are kept open simultaneously.  This
determines how many are kept open.  A number too large may exhaust your memory
pretty fast and result in a crash.  A number too small may result in slow
playback as assets need to be reopened more frequently.

@cindex Seconds to preroll renders
@item @b{Seconds to preroll renders} @*
Some effects need a certain amount of time to settle in.  Checking this option sets a number of
seconds to render without writing to disk before the selected region is
rendered.  When using the renderfarm, you will sometimes need to preroll to get
seemless transitions between the jobs.  Every job in a renderfarm is prerolled
by this value.  This does not affect background rendering, however.  Background
rendering uses a different preroll value.

@cindex Force single processor use
@cindex SMP, force single processor use
@item @b{Force single processor use} @*
Cinelerra tries to use all processors on the system by default, but sometimes
you will only want to use one processor, like in a renderfarm client.  This
forces only one processor to be used.  In addition, the operating system usually
uses the second processor for disk access.  So this option is really a
1.25 processor mode.  The value of this parameter is used in renderfarm
clients.
@end itemize

@menu
* Background rendering::
* Renderfarm::
@end menu

@c cincvdoc_node_number_52
@node Background rendering
@subsection Background rendering
@cindex Background rendering
@cindex Rendering, background

Background rendering was originally conceived to allow HDTV effects to be
displayed in real-time.  Background rendering causes temporary output to
be rendered constantly while the timeline is being modified.  The temporary
output is displayed during playback whenever possible.  This is useful for
transitions and previewing effects that are too slow to display in
real time.  If a renderfarm is enabled, the renderfarm is used
for background rendering.  This gives you the potential for real-time effects if
enough network bandwidth and CPU nodes exist.

Background rendering is enabled in the @b{Performance} tab of the
@b{Preferences} window.  It has one interactive function @b{Settings menu ->
Set background render}.  This sets the point where background rendering starts
up to the position of the insertion point.  If any video exists, a red bar appears in the time
ruler showing what has been background rendered.

It is often useful to insert an effect or a transition and then select
@b{Settings menu -> Set background render} right before the effect to preview
it in real time and full framerates.

@cindex Frames per background rendering job
@itemize @bullet
@item @b{Frames per background rendering job} @*
This only works if a renderfarm is being used; otherwise, background rendering
creates a single job for the entire timeline.  The number of frames specified
here is scaled to the relative CPU speed of rendering nodes and used in a
single renderfarm job.  The optimum number is 10 - 30 since network bandwidth
is used to initialize each job.

@cindex Frames to preroll background
@item @b{Frames to preroll background} @*
This is the number of frames to render ahead of each background rendering job.
Background rendering is degraded when preroll is used since the jobs are small.
When using background rendering, this number is ideally 0.  Some effects may
require 3 frames of preroll.

@cindex Output for background rendering
@item @b{Output for background rendering} @*
Background rendering generates a sequence of image files in a certain
directory.  This parameter determines the filename prefix of the image files.
It should be on a fast disk, accessible to every node in the renderfarm by the
same path.  Since hundreds of thousands of image files are usually created,
@command{ls} commands will not work in the background rendering directory.  The
@image{manual_images_en/magnify,7mm} browse button for this option normally will not
work either, but the @image{manual_images_en/wrench,4.33mm} configuration button for
this option works.

@cindex File format
@item @b{File format} @*
The file format for background rendering has to be a sequence of images.  The
format of the image sequences determines the quality and speed of playback.
JPEG is good most of the time.
@end itemize

@c cincvdoc_node_number_53
@node Renderfarm
@subsection Renderfarm
@cindex Renderfarm

To use the renderfarm, set these options.  Ignore them for a standalone system

@cindex Use render farm for rendering
@itemize @bullet
@item @b{Use render farm for rendering} @*
When selected, all the @b{file->render} operations use the renderfarm.

@cindex Nodes
@item @b{Nodes} @*
Displays all the nodes on the renderfarm and shows which ones are active.  Nodes are
added by entering the host name of the node, verifying the value of @b{port}
and selecting @b{add node}.  If you have hundreds of nodes, experienced
users may be better off editing the
@file{~/.bcast/.Cinelerra_rc} file rather than using configuration dialog.
Remember that @file{.Cinelerra_rc} is overwritten whenever a copy of Cinelerra
exits. @*
Once nodes are created, select the @b{ON} column to activate and deactivate nodes.
Nodes may be edited by highlighting a row and hitting @b{apply
changes}.

@cindex Hostname, renderfarm
@item @b{Hostname} @*
Edit the hostname of an existing node or enter the hostname of a new node here.

@cindex Port, renderfarm
@item @b{Port} @*
Edit the port number of an existing node or enter the port number of a new node here.

@cindex Node, replace
@item @b{Replace node} @*
When editing an existing node, select this to commit the changes to @b{hostname}
and @b{port}.  The changes will not be committed if you do not click this button.

@cindex Node, add a
@item @b{Add node} @*
Create a new node with the @b{hostname} and @b{port} settings.

@cindex Node, delete a
@item @b{Delete node} @*
Deletes whatever node is highlighted in the @b{nodes} list.

@cindex Sort nodes
@cindex Nodes, sort
@item @b{Sort nodes} @*
Sorts the @b{nodes} list based on the hostname.

@cindex Rates, reset
@item @b{Reset rates} @*
This sets the framerate for all the nodes to 0.  Frame rates are used to scale
job sizes based on CPU speed of the node.  Frame rates are calculated only when
renderfarm is enabled.

@cindex Total jobs to create
@cindex Jobs, total number to create
@item @b{Total jobs to create} @*
Determines the number of jobs to dispatch to the renderfarm.  The more jobs you
create, the more finely balanced the renderfarm becomes. @*
You can determine the total jobs to create by multiplying the number of nodes
including the master node by some number.  Multiply them by 1 to have one job
dispatched for every node.  Multiply them by 3 to have 3 jobs dispatched for
every node.  If you have 10 slave nodes and one master node, specify 33 to have
a well balanced renderfarm.
@end itemize

@c cincvdoc_node_number_54
@node Interface
@section Interface
@cindex Interface

These parameters affect purely how the user interface works.

@cindex Index files, location
@cindex Index files
@itemize @bullet
@item @b{Index files go here} @*
Back in the days when 4 MB/sec was extremely fast for a hard drive, index
files were introduced to speed up drawing the audio tracks.  This option
determines where index files are placed on the hard drive.

@cindex Index file, size of
@item @b{Size of index file} @*
This determines the size of an index file.  Larger index sizes allow smaller files
to be drawn faster, while slowing down the drawing of large files.  Smaller
index sizes allow large files to be drawn faster, while slowing down small
files.

@cindex Index files, number to keep
@item @b{Number of index files to keep} @*
To keep the index directory from becoming unruly, old index files are deleted.
This determines the maximum number of index files to keep in the directory.

@cindex Index files, delete all
@item @b{Delete all indexes} @*
When you change the index size or you want to clean out excess index files,
this deletes all the index files.

@cindex Time representation
@item @b{Use hours:minutes:seconds.xxx} @*
Various representations of time are given.  Select the most convenient one.
The time representation can also be changed by @key{CTRL} clicking on the time
ruler.

@cindex Thumbnails
@item @b{Use thumbnails} @*
The Resource Window displays thumbnails of assets by default.  Drawing asset thumbnails can take a
while.  This option disables thumbnail drawing.

@cindex In/out points, clicking does what
@item @b{Clicking in/out points does what} @*
Cinelerra not only allows you to perform editing by dragging in/out points, but
also defines three separate operations that occur when you drag an in/out
point.  Here you can select the behavior of each mouse button.  The
usage of each editing mode is described in editing.

@cindex Meter, Min dB
@item @b{Min dB for meter} @*
Some sound sources have a lower noise threshold than others.  Everything below
the noise threshold is meaningless.  This option sets the meters to clip below
a certain level.  Consumer soundcards usually bottom out at -65.  Professional
soundcards bottom out at -90.  @xref{Sound level meters window}.

@cindex Meter, Max dB
@item @b{Max dB for meter} @*
This sets the maximum sound level represented by the sound meters.  No matter
what this value is, no soundcard can play sound over 0 dB@.  This value is
presented merely to show how far over the limit a sound wave is.  @xref{Sound
level meters window}.

@cindex Theme
@item @b{Theme} @*
Cinelerra supports variable themes.  Select one here and restart Cinelerra to
see it.
@end itemize

@c cincvdoc_node_number_55
@node About
@section About window

This section gives you information about the copyright, the time of the current
build, the lack of a warranty, and the versions of some of the libraries.  Be
sure to agree to the terms of the lack of the warranty.

@c cincvdoc_node_number_56
@node Project attributes
@chapter Project attributes
@cindex Project attributes
@cindex Attributes of project

@menu
* Set format window::
* Presets::
* Audio attributes::
* Video attributes::
@end menu

@c cincvdoc_node_number_57
@node Set format window
@section Set format window
@cindex Set format window
@cindex Format, window

When you play media files in Cinelerra, the media files have a certain number
of tracks, a certain frame size, a certain sample size, and so on and so forth.
No matter what attributes the media file has, it is played back according
to the project attributes.  So, if an audio file's sample rate is different than the
project attributes, it is resampled.  In like fashion, if a video file's frame size is different
than the project attributes, the video is composited on a black frame, either cropped
or bordered with black.

The project attributes are adjusted in @b{Settings->Set Format} and to a
lesser extent in @b{File->New}.  When you adjust project settings in
@b{file->new}, a new, empty timeline is created.  Every timeline created
from this point on uses the same settings.  When you adjust settings in
@b{settings->format}, media on the timeline is left unchanged. Also, every
timeline created from this point uses the same settings.

@center @image{manual_images_en/format,70mm}
@center @b{Set format window}

In addition to the traditional settings for sample rate, frame rate, frame
size, Cinelerra uses some unusual settings like @b{channel positions, color
model, and aspect ratio.}

@c cincvdoc_node_number_58
@node Presets
@section Presets
@cindex Presets

Select an option from this menu to have all the project settings set to
one of the known standards.

@c cincvdoc_node_number_59
@node Audio attributes
@section Audio attributes
@cindex Audio attributes

@itemize @bullet
@item @b{Tracks} @*
Sets the number of audio tracks for the new project.  Tracks can
be added or deleted later, but options are provided here for convenience.

@item @b{Channels} @*
Sets the number of audio channels for the new project.  The number
of audio channels does not have to be the same as the number of tracks.

@item @b{Samplerate} @*
Sets the samplerate of the audio.  The project samplerate does not have to
be the same as the media sample rate that you load.  Media is resampled to
match the project sample rate.

@item @b{Channels positions} @*
The currently enabled audio channels and their positions in the user interface
boxes are displayed in the channel position widget.
@end itemize

@center @image{manual_images_en/channelpositions,40mm}

The channels are numbered.  When rendered, the output from channel 1 is
rendered to the first output track in the file or the first soundcard channel
of the soundcard.  Later channels are rendered to output tracks numbered
consecutively.

The audio channel positions correspond to where in the panning widgets each of
the audio outputs is located.  The closer the panning position is to one of the audio
outputs, the more signal that speaker gets.  Click on a speaker icon and drag
to change the audio channel location.

The speakers can be in any orientation.  A different speaker arrangement is
stored for every number of audio channels since normally you do not want the
same speaker arrangement for different numbers of channels.

Channel positions is the only setting that does not affect the output
necessarily.  Click on a speaker icon and drag to change the position of a
channel.  It is merely a convenience, so that when more than two channels are used, the
pan controls on the timeline can distinguish between them.  It has nothing to
do with the actual arrangement of speakers.

Different channels can be positioned very close together to make them have
the same output.

@c cincvdoc_node_number_60
@node Video attributes
@section Video attributes
@cindex Video attributes

@itemize @bullet
@item @b{Tracks} @*
Sets the number of video tracks the new project is assigned.  Tracks can
be added or deleted later, but options are provided here for convenience.

@item @b{Framerate} @*
Sets the framerate of the video.  The project framerate does not have to be
the same as an individual media file frame rate that you load.  Media is reframed to match the
project framerate.

@item @b{Canvas size} @*
Sets the size of the video output.  In addition, each track also has its own frame
size.  Initially, the @b{New} dialog creates video tracks whose size match
the video output. The video track sizes can be changed later without
changing the video output.

@item @b{Aspect ratio} @*
Sets the aspect ratio.  The aspect ratio is applied to the video output.
The aspect ratio can be different than the ratio that results from the formula:
h / v (the number of horizontal pixels divided into the number of vertical
pixels).  If the aspect ratio differs from the results of the formula above,
your output will be in non-square pixels.

@item @b{Auto aspect ratio} @*
If this option is checked, the @b{New} dialog always recalculates the @b{Aspect
ratio} setting based upon the given @b{Canvas size}.  This ensures pixels are
always square.

@item @b{Color model} @*
The project will be stored in the color model video intermediates selected in
the dropdown. @*
Color model is very important for video playback because video has the
disadvantage of being very slow.  Although it is not noticeable, audio
intermediates contain much more information than the audio on disk and the
audio which is played.  Audio always uses the highest bandwidth intermediate
because it is fast. @*
Video intermediates must use the least amount of data for the required quality
because it is slow, but video intermediates still use a higher bandwidth color
model than video which is stored and video which is played.  This allows more
processing to be done with less destruction of the original data. @*
The video is stored on disk in one colormodel, usually a YUV
derivative.  When played back, Cinelerra decompresses it from the file format
directly into the format of the output device.  If effects are processed, Cinelerra
decompresses the video into an intermediate colormodel first and then converts it to
the format of the output device.  The selection
of an intermediate colormodel determines how fast and accurate the effects are. @*
Cinelerra colormodels are described using a certain packing order of components
and a certain number of bits for each component.  The packing order is printed
on the left and the bit allocation is printed on the right.
@cindex RGB-888
@itemize @bullet
@item @b{RGB-888} @*
This allocates 8 bits for the R, G, and B channels and no alpha.  This is
normally used for uncompressed media with low dynamic range.
@cindex RGBA-8888
@item @b{RGBA-8888} @*
This allocates an alpha channel to the 8 bit RGB colormodel.  It is used
for overlaying multiple tracks.
@cindex YUV-888
@item @b{YUV-888} @*
This allocates 8 bits for Y, U, and V@.  This is used for low dynamic range
operations in which the media is compressed in the YUV color space.  Most
compressed media is in YUV and this derivate allows video to be processed fast with the
least color degradation.
@cindex YUVA-8888
@item @b{YUVA-8888} @*
This allocates an alpha channel to the 8 bit YUV colormodel for transparency.
@cindex RGB-Float
@item @b{RGB-Float} @*
This allocates a 32 bit float for the R, G, and B channels and no alpha.
This is used for high dynamic range processing with no transparency.
@cindex RGBA-Float
@item @b{RGBA-Float} @*
This adds a 32 bit float for alpha to RGB-Float.  This is used for high
dynamic range processing with transparency.
@end itemize
@b{In order to do effects which involve alpha channels, a colormodel with an
alpha channel must be selected}.  These are RGBA8888, YUVA8888, and RGBA Float.
The 4 channel colormodels are slower than 3 channel colormodels,
with the slowest being RGBA Float.  Some effects, like fade, work around the
need for alpha channels while other effects, like chromakey, require an alpha
channel to do anything, so it is a good idea to try the effect without alpha
channels to see if it works before settling on an alpha channel and slowing it
down. @*
When using compressed footage, YUV colormodels are usually faster than RGB colormodels.
They also destroy fewer colors than RGB colormodels.  If
footage stored as JPEG or MPEG is processed many times in RGB, the colors will
fade whereas they will not fade if processed in YUV@. @*
Years of working with high dynamic range footage have shown floating point RGB
to be the best format for high dynamic range.  16 bit integers were used
in the past and were too lossy and slow for the amount of improvement. @*
RGB float does not destroy information when used with YUV source footage and
also supports brightness above 100%.  Be aware that some effects, like
Histogram, still clip above 100% when in floating point.
@end itemize

@c cincvdoc_node_number_61
@node Loading and saving files
@chapter Loading and saving files
@cindex Loading and saving files
@cindex Files, loading and saving

@menu
* Supported file formats::     What formats Cinelerra can import and export
* Loading files::              Loading all types of files
* Loading the backup::         Recovering the session from before a crash
* Saving files::               Saving edit decision lists
* Merging projects::
@end menu

@c cincvdoc_node_number_62
@node Supported file formats
@section Supported file formats
@cindex Supported file formats
@cindex File format

Here are most of the supported file formats and notes regarding their
compression.  You may be able to load other formats not described here. @*
The format of the file affects what Cinelerra does with it.  Edit decision
lists stored in XML save the project settings.  Formats which contain media but no edit
decisions just add data to the tracks.  If your project sample rate is 48 kHz
and you load a sound file with 96khz, you will still be playing it at 48 kHz.  If
you load an EDL file at 96khz and the current project sample rate is 48 kHz,
you will change it to 96 kHz. @*
Some file formats are very slow to display on the timeline.  These usually have
video which is highly compressed.  Drawing highly compressed video thumbnails on the timeline (picons) can
be very slow.  Disable picon drawing for these files with the @b{draw media}
toggle to speed up operations.

@center @image{manual_images_en/track_attributes}
@center @b{Track attributes}

Supported file formats are currently:
@itemize @bullet
@item WAV
@item PCM
@item AIFF
@item AC3 audio
@end itemize

@menu
* Quicktime::
* MPEG-4 audio::
* Images sequence::
* Still images::
* AVI::
* MPEG files containing video::
* DVD movies::
* MPEG 1 audio::
* Ogg Theora/Vorbis::
* Edit decision list::
@end menu

@c cincvdoc_node_number_63
@node Quicktime
@subsection Quicktime
@cindex Quicktime

Quicktime is not the standard for UNIX but we use it because it is well
documented.  All of the Quicktime movies on the internet are compressed.
Cinelerra supports some compressed Quicktime movies.  If Cinelerra
crashes when loading a Quicktime movie, it is most likely because the format
was not supported. @*
Quicktime is a container for two streams, a video stream and an audio stream.  These
streams are compressed using separate encoding schemes.  The preferred encoding for
Quicktime output is MPEG-4 Video and MPEG-4 Audio.  This format is compatible in the
commercial players for Windows, has good compression quality and good output quality.  For better
compression, use H-264 video.  Unfortunately H-264 decoding is so slow it can
not play very large frame sizes. @*
Cinelerra supports two non-standard codecs: Dual MPEG-4 video and Dual H.264
video.  These will not play in anything but Cinelerra and XMovie.  They are
designed for movies where the frames have been divided into two fields, each
field displayed sequentially.  The dual codecs interleave two video streams to
improve efficiency without requiring major changes to the player.

@c cincvdoc_node_number_64
@node MPEG-4 audio
@subsection MPEG-4 audio
@cindex MPEG-4 audio

This is the same as Quicktime with MPEG-4 Audio as the audio codec.

@c cincvdoc_node_number_65
@node Images sequence
@subsection Images sequence
@cindex Images sequence

Rendering an images sequence is not the same as rendering a single image.  When
rendering an images sequence Cinelerra generates a table of contents file for
the images sequence and creates a different image file for every timeline
position.  To get better performance, the table of contents can be loaded instead of the individual images.
To learn more about the different image formats
supported in an images sequence, read about still images.

@c cincvdoc_node_number_66
@node Still images
@subsection Still images
@cindex Still images
@cindex Images, still

@menu
* Loading still images::
* Still images size::
* Open EXR images::
* Raw digital camera images::
@end menu

@c cincvdoc_node_number_67
@node Loading still images
@subsubsection Loading still images
@cindex Loading still images
@cindex Still images, loading

Rendering a single image causes the image file to be overwritten for every
timeline position.  No table of contents is created.  When loaded, the image
takes up one frame in length.  To view the image, zoom in time on the timeline so you can see the single frame.  To extend the
length of the image, drag it just as you would do with regular video media.  You can drag a still
image as much as you want.  Images in Cinelerra have the ability to be dragged to an infinite length. @*
Cinelerra lets you define the initial size of the loaded images.  The parameter
is set in the Images section of the @b{settings preferences recording window}. @*
Unless your original material comes from a digital source (like a digital photo
camera), the first thing you have to do before you can use it is to somehow
capture the assets into a usable digital medium. @*
For old photos, paper maps, drawings or diagrams, you might have to use a scan
them into a file format like PNG, TIF, TGA or JPG files by using a digital scanner.  You might
want to use Gimp to post-process the images, clean damaged areas or color
correct the assets. @*
If your assets come from a digital source like a digital camera or a screen
capture, be sure to capture the material using the best resolution possible.
This will allow you to get the best quality output from your Cinelerra project.

@c cincvdoc_node_number_68
@node Still images size
@subsubsection Still images size
@cindex Still images size

@b{Important:} Imported images always stay at their original size.  Therefore,
you have to take into account the aspect ratio of your video in Cinelerra and
rescale your pictures before importing them in Cinelerra. @*
For example, PAL images aspect ratio is 4/3, but 720x576 is 5/4. For your
imported images to be displayed correctly, you have to rescale their horizontal
size: @*
new horizontal size=@math{(5 / 4) / (4 / 3)} x original horizontal size @*
For PAL videos, you have to multiply the horizontal size of the pictures you
want to import by a factor of 0.9375. @*
Here is a small shell script which, when ran from a folder containing jpg
images, resize those images and put the new images in a @file{resized} folder:
@verbatim
#/bin/sh
mkdir resized
for element in `ls . | grep jpg`;
do
    size=`identify ${element}`
    width=`echo ${size} | sed '+s+.*JPEG ++' | sed '+s+x.*++'`
    height=`echo ${size} | sed '+s+.*JPEG [0-9]*x++' | sed '+s+DirectClass.*++'`
    let new_width=${width}*9375/10000
    convert -resize ${new_width}x${height} -quality 100 ${element} resized/${element}
done
@end verbatim

@c cincvdoc_node_number_69
@node Open EXR images
@subsubsection Open EXR images
@cindex EXR images
@cindex Images, EXR

You may not know about Open EXR@.  This format stores floating point RGB images.
It also supports a small amount of compression.  Projects which render to EXR
should be in a floating point color model to take advantage of the benefits of EXR @xref{Project
attributes}.  Several compression options are available for EXR@.

@cindex Compression
@cindex PIZ compression
@itemize @bullet
@item @b{PIZ:} Lossless wavelet compression.  This is the best compression.
@cindex ZIP compression
@item @b{ZIP:} Lossless gzip algorithm.
@cindex RLE compression
@item @b{RLE:} Lossless run length encoding.  This is the fastest, but worst
compression.
@cindex PXR24 compression
@item @b{PXR24:} Lossy compression where the floating point numbers are
converted to 24 bits and compressed with gzip.
@end itemize

Select @b{Use Alpha} if the project colormodel has an alpha channel and you
want to retain it in the file.  Otherwise the primary colors are multiplied by
the alpha channel.

@c cincvdoc_node_number_70
@node Raw digital camera images
@subsubsection Raw digital camera images
@cindex Raw digital camera images
@cindex Camera images, raw digital

RAW digital camera images are a special kind of image file that Cinelerra only
imports.  Once they are on the timeline, these must be processed in a floating point color space.
Raw images from Canon cameras are the only ones tested.  They
need to have the @b{Gamma} effect applied to correct gamma.  Because raw images
take a long time to interpolate, they are usually viewed first in a proxy file
and then touched up.

First apply the Gamma effect to a track of raw images and set it to
@b{automatic} with @b{0.6} gamma.  Then render the timeline to a Quicktime JPEG
file.  Append the Quicktime JPEG file in a new track and disable playback of
the old track.  Now the gamma corrected copy of each raw image can be previewed
relatively fast in the same timeline position as the original image.

@c cincvdoc_node_number_71
@node AVI
@subsection AVI
@cindex AVI

Because AVI (Audio-Video Interleave) is so fragmented with varied audio and
video codecs, you may not be able to play all AVI formatted files.

@c cincvdoc_node_number_72
@node MPEG files containing video
@subsection MPEG files containing video
@cindex MPEG files containing video
@cindex mpeg3toc

MPEG files containing video can be loaded directly into Cinelerra.  If the file
is supported, a table of contents is built.  If the file is unsupported, it
usually crashes or shows very short tracks.  Unfortunately, this method of
loading MPEG files is not good enough if you intend to use the files in a
renderfarm. @*
To use MPEG files in a renderfarm, you need to run mpeg3toc in order to generate a table
of contents for the file and then load the table of contents.  mpeg3toc needs the
absolute path of the MPEG file.  If you do not use an absolute path, it assumes that
the MPEG file is in the same directory that Cinelerra is run from. @*
MPEG streams are structured into multiple tracks.  Each track can be video or
audio.  Each audio track can have 1-6 channels.  Cinelerra converts each
channel of audio into a track.

@cindex mpeg2enc
@b{Notes on mpeg video encoding:} @*
MPEG video encoding is done separately from MPEG audio encoding.  In MPEG video
there are two colormodels.  The YUV 4:2:0 colormodel is encoded by a highly
optimized version of mpeg2enc with presets for standard consumer electronics.
In the process of optimizing mpeg2enc, they got rid of YUV 4:2:2 encoding.  The
YUV 4:2:2 colormodel is encoded by a less optimized version of mpeg2enc. @*
YUV 4:2:2 encoding was kept around because the NTSC version of DV video loses
too much quality when transferred to YUV 4:2:0.  This DV video must be
transferred to YUV 4:2:2. @*
When encoding YUV 4:2:0, the bitrate parameter changes meaning depending on
whether the bitrate or quantization is fixed.  If the bitrate is fixed, it is
the target bitrate.  If the quantization is fixed, it is the maximum bitrate
allowed.  This is a quirk of the mpeg2enc version.

@c cincvdoc_node_number_73
@node DVD movies
@subsection DVD movies
@cindex DVD movies
@cindex IFO file

DVD are split into a number of programs, each identified by a unique
@file{IFO} file.  If you want to load a DVD, find the corresponding @file{IFO}
file for the program of interest.  Load the IFO file directly and a table of
contents will be built.  Alternatively for renderfarm usage, a table of
contents can be created separately. @*
@cindex mpeg3toc
Run: @command{mpeg3toc -v /cdrom/video_ts/vts_01_0.ifo dvd.toc} @*
or something similar.  Then load @file{dvd.toc}.

@c cincvdoc_node_number_74
@node MPEG 1 audio
@subsection MPEG 1 audio
@cindex MPEG 1 audio

These are .mp2 and .mp3 files.  If the files are encoded using a fixed bitrate, they can be loaded directly
with no table of contents.  Variable bitrate streams need to have a table of
contents created with mpeg3toc.

@c cincvdoc_node_number_75
@node Ogg Theora/Vorbis
@subsection Ogg Theora/Vorbis
@cindex Ogg Theora/Vorbis

The OGG format is an antiquated but supposedly unpatented way of compressing
audio and video.  The quality is not as good as H.264 or MPEG-4 Audio.  In
reality, anyone with enough money and desire can find a patent violation so the
justification for OGG is questionable.

@c cincvdoc_node_number_76
@node Edit decision list
@subsection Edit decision list
@cindex EDL

Edit decision lists are generated by Cinelerra for storing projects.  EDL files end
in .xml.  When loaded, they change the attributes of the current project.  Because edit decision
lists consist of text, they can be edited in a text editor.

@c cincvdoc_node_number_77
@node Loading files
@section Loading files
@cindex Loading files
@cindex Files, loading

All data that you work with in Cinelerra is acquired either by @b{recording
from a device} or by @b{loading from disk}.  This section describes loading. @*
The loading and playing of files is just as you would expect.  Just go to
@b{file->Load}, select a file for loading, and click @b{ok}.  Click the forward
play button and it should start playing, regardless of whether a progress bar
has appeared.

@center @image{manual_images_en/load, 80mm}
@center @b{The load window}

If the file is a still image, the project's attributes are not changed and the
first frame of the track becomes the image.  If the file has audio, Cinelerra
may build an index file in order to speed up drawing.  You can edit and play the
file while the index file is being built.

@menu
* Insertion strategy::
* Loading multiple files::
* Loading files from the command prompt::
* Filtering files by extension::
* Loading other formats::
@end menu

@c cincvdoc_node_number_78
@node Insertion strategy
@subsection Insertion strategy
@cindex Insertion strategy
@cindex Strategy, insertion

Usually, three things happen when you load a file:
@enumerate 1
@item the existing project is cleared from the screen
@item the project's attributes are changed to match the file's attributes
@item the new file's tracks are created in the timeline
@end enumerate
However, Cinelerra lets you change what happens when you load a file. @*
In the file selection box go to the @b{Insertion strategy} box and select it.
Each of these options loads the file a different way.
@itemize @bullet
@item @b{Replace current project} @*
All tracks in the current project are deleted and new tracks are created to
match the source.  Project attributes are only changed when loading XML@.  If
multiple files are selected, Cinelerra adds new tracks for each file.

@item @b{Replace current project and concatenate tracks} @*
Same as replace current project, except that if multiple files are selected, Cinelerra
will concatenates the tracks of each file, one after another, in alphanumeric order.

@item @b{Append in new tracks} @*
The current project is not deleted and new tracks are created for the
source.

@item @b{Concatenate to existing tracks} @*
The current project is not deleted and new files are concatenated to the
existing tracks.

@item @b{Paste at insertion point} @*
The file is pasted into the timeline at the insertion point.

@item @b{Create new resources only} @*
The timeline is unchanged and new resources are created in the Resource
Window.
@end itemize

The insertion strategy is a recurring option in many of Cinelerra's functions.
In each place the options do the same thing.  Using these options, you can almost
do all your editing by loading files.  If you load files by passing command
line arguments to Cinelerra, the files are loaded with @b{Replace current
project} rules.

@c cincvdoc_node_number_79
@node Loading multiple files
@subsection Loading multiple files
@cindex Loading multiple files
@cindex Files, loading multiple

In the file selection box go to the list of files.  Select a file.  Go to
another file and select it while holding down @key{CTRL}.  This selects one
additional file.  Go to another file and select it while holding down
@key{SHIFT}.  This selects every intervening file.  This behavior is available
in most every list box. @*
Use this method and the @b{Concatenate to existing tracks} insertion strategy
to create an images slideshow or a song playlist.

@c cincvdoc_node_number_80
@node Loading files from the command prompt
@subsection Loading files from the command prompt
@cindex Loading files from the command prompt
@cindex Command prompt, loading files from the

Another way to load files is to pass the filenames as arguments on the command
line. @*
@command{cinelerra myvideo.mov myothervideo.mov} @*
This creates new tracks for every file and starts the program with all the
arguments loaded.

@c cincvdoc_node_number_81
@node Filtering files by extension
@subsection Filtering files by extension
@cindex Filtering files by extension
@cindex Files, extension
@cindex Extension, filtering files by

If there are too many files in your media directory, it can be difficult to
find the file you want.  For this purpose, the @b{load
window} allows you to filter which files are displayed in the list box by extension
name. @*
Click the dropdown box (right below the file name text box)
and select the file extension of your media (i.e.  mpg,
mov, mp3, avi, etc).  The file list now shows only files with the selected
extension.

@c cincvdoc_node_number_82
@node Loading other formats
@subsection Loading other formats
@cindex Loading other formats
@cindex Other formats, loading

If you can not load a particular kind of video clip and do not have the
original source file, you will have to convert it to a format supported by Cinelerra.

@itemize @bullet
@item @b{ffmpeg} @*
Syntaxis: @*
@command{ffmpeg -sameq -i ORIGINAL.avi new_video.mpeg} @*
The @option{-sameq} option maintains the original quality.

@item @b{mjpegtools} @*
Syntaxis: @*
Video: @command{lav2yuv original.avi | mpeg2enc -o output.m1v} @*
Audio: @command{lav2wav original.avi | mp2enc -o output.mp2} @*
Mixing audio & video: @command{mplex output.mp2 output.m1v -o new_video.m1v}

@item @b{mencoder} @*
First option: @*
@command{mencoder INPUT_FILE.avi -ovc lavc -lavcopts vcodec=mpeg4:
vhq:vbitrate=300 -oac mp3lame -lameopts br=64:vol=1 -o OUTPUT_FILE.avi} @*
Second option: @*
@command{mencoder input.avi -ovc rawyuv -oac copy -o output.avi}

@item @b{dv-utils} @*
Syntaxis: @*
@command{dv2dv -o qt input_file.avi output_file.qt}

@item @b{Transcode} @*
First option: @*
@command{transcode -i test.avi -o test_divx4mp3.avi -y divx4} @*
Second option: @*
@command{transcode -i test.mov -x mplayer -o salida.mpg -y yuv4mpeg} @*
More information about the available transcode options can be found at: @*
@uref{http://www.theorie.physik.uni-goettingen.de/~ostreich/transcode}
@end itemize

@c cincvdoc_node_number_83
@node Loading the backup
@section Loading the backup
@cindex Loading the backup
@cindex Backup, loading the

There is one special XML file on disk at all times.  After every editing
operation, Cinelerra saves the current project to a backup in
@file{$HOME/.bcast/backup.xml}.  In the event of a crash, the first thing you should do
after restarting Cinelerra is select @b{file->load backup} in order to load the backup.
This will start Cinelerra at the point in your editing operations directly before the program
crashed.  It is important after a crash to restart Cinelerra
without performing any editing operations as you will overwrite the backup.

@c cincvdoc_node_number_84
@node Saving files
@section Saving files
@cindex Saving files
@cindex Files, saving
@cindex Files, XML
@cindex XML files

When Cinelerra saves a file, it saves an edit decision list of the current
project but does not save any media.  Go to @b{File->save as...}.  Select a file
to overwrite or enter a new file.  Cinelerra automatically concatenates
@samp{.xml} to the filename if no @samp{.xml} extension is given.

The saved file contains all the project settings and locations of every edit.
Instead of media, the file contains pointers to the original media files on disk.

For each media file, the XML file stores either an absolute path or just the
relative path.  If the media is in the same directory as the XML file, a
relative path is saved.  If it is in a different directory, an absolute path is
saved.

In order to move XML files around without breaking the media linkages, you
need either to keep the media in the same directory as XML file forever or save
the XML file in a different directory than the media and not move the media
ever again.

If you want to create an audio playlist and burn it on a CD-ROM, save the XML
file in the same directory as the audio files and burn the entire directory.
This keeps the media paths relative.

XML files are useful in saving the current state of Cinelerra before retiring from an editing session.
XML files are specific to
Cinelerra only.  You can not play XML files in a dedicated movie player.  Real-time
effects in an XML file have to be re-synthesized every time you play it back.
The XML file also requires you to maintain copies of all the source assets on
hard disk, which can take up space and cost a lot of electricity to spin.
Render your projects to a final format for more persistent storage of the output.

@c cincvdoc_node_number_85
@node Merging projects
@section Merging projects
@cindex Merging projects
@cindex Projects, merging

To merge several separate projects into one big one :
@enumerate 1
@item Open cinelerra
@item Load project A
@item Open a second copy of cinelerra
@item Load project B
@item Cut and paste from A to B
@end enumerate

@c cincvdoc_node_number_86
@node Program window
@chapter Program window
@cindex Program window

@menu
* Navigating the program window::
* Editing::                           Moving the media in time.
@end menu

This contains the timeline and the entry point for all menu driven operations.
The timeline consists of a vertical stack of tracks with a horizontal
representation of time.  This defines the output of rendering operations and
what is saved when you save files.  To the left of the timeline is the patchbay which
contains options affecting each track.

@center @image{manual_images_en/program_insertion_point,120mm}
@center @b{The timeline}

Under the @b{Window} menu, you will find options that affect the main windows.
@b{default positions} repositions all the windows to a 4 screen editing
configuration.  On dual headed displays, the @b{default positions} operation
fills only one monitor with windows.

@c cincvdoc_node_number_87
@node Navigating the program window
@section Navigating the program window
@cindex Navigating the program window
@cindex Program window, navigating the

The program window contains many features for navigation and displays the
timeline as it is structured in memory: tracks stacked vertically and extending
across time horizontal.  The horizontal scroll bar allows you to scan across
time.  The vertical scroll bar allows you to scan across tracks.

@menu
* Video tracks::
* Audio tracks::
* Track navigation::
* The track popup menu::
* The insertion point::
* The in/out points::
* Using labels in the program window::
@end menu

@c cincvdoc_node_number_88
@node Video tracks
@subsection Video tracks
@cindex Video tracks
@cindex Tracks, video

@center @image{manual_images_en/track_video,120mm}
@center @b{A video track}

Video tracks represent the duration of your videos and clips, just as if you placed real
photographic film stock end-to-end on a table.  The individual images you see on the
track are samples of what is located at that particular instant on the timeline.

@c cincvdoc_node_number_89
@node Audio tracks
@subsection Audio tracks
@cindex Audio tracks
@cindex Tracks, audio

@center @image{manual_images_en/track_audio,120mm}
@center @b{An audio track}

Audio tracks represent your sound media as an audio waveform.  Following the film
analogy, it would be as if you "viewed" magnetic tape horizontally on your
table. @*
You can adjust the horizontal and vertical magnification of the tracks, using
the @b{zoom panel bar}. @*
You can adjust the magnification of the audio "waveform" display using the
@b{zoom panel bar}. @*
The controls to the left of the tracks are called the @b{patch bay}.  The patch bay is
used to control the behavior of the tracks.

@c cincvdoc_node_number_90
@node Track navigation
@subsection Track navigation
@cindex Track navigation
@cindex Navigation, track

Track Navigation involves both selecting a specific audio or video track and
moving to a certain time in the track.  The program window contains many
features for navigation and displays the timeline as it is structured in
memory.

The horizontal scroll bar allows you to scan across time.

The vertical scroll bar allows you to scan across tracks.

In addition to the graphical tools, you may also use the keyboard
to navigate.  As a general rule, keyboard navigation is faster than navigation
with a mouse.
Use @kbd{PAGE UP} and @kbd{PAGE DOWN} to scroll up and down the
tracks.

You will often need to scroll beyond the end of the timeline, but the scrollbars
will not let you do it.  Instead, use the RIGHT arrow to scroll past the end of
timeline.

Use the @key{HOME} and @key{END} keys to instantly go to the beginning or end
of the timeline.  In @b{I-beam} mode, hold down @key{SHIFT} while pressing
@key{HOME} or @key{END} in order to select the region of the timeline between the
insertion point and the key pressed.

Below the timeline, you will find the zoom panel.  The zoom panel contains values
for @b{sample zoom}, @b{amplitude}, @b{track zoom}, and @b{curve zoom}.  In addition
to the scrollbars, these values are the main tools for positioning the
timeline.

@center @image{manual_images_en/zoompanel,100mm}

Changing the @b{sample zoom} causes the unit of time displayed in the timeline to change size.  It
allows you to view your media all the way from individual frames to the
entire length of your project.  The higher the setting, the more frames you can see per
screen.  @b{If your mouse has a wheel and it works in X11, mouseover the tumblers
and use the wheel to zoom in and out.}

@cindex Amplitude
The @b{amplitude} only affects audio.  It determines how large the waveform appears if
the waveform is drawn.

@cindex Track zoom
The @b{track zoom} affects all tracks.  It determines the height of each track.
If you change the track zoom, the amplitude zoom compensates so that the audio waveforms
look proportional.

@cindex Curve zoom
The @b{curve zoom} affects the curves in all the tracks.  It determines the
amplitude and offset of the curves.  The tumbler changes curve amplitude, but
the only way to curve offset is to use the @b{fit curves} button.
@image{manual_images_en/fit_curves}

Use the @kbd{LEFT} and @kbd{RIGHT} arrows to move across time in small
increments.  You will often need to scroll beyond the end of the timeline, but
scrollbars will not let you do this.  Instead, use the @kbd{RIGHT} arrow to scroll
past the end of timeline.

Use the @kbd{UP} and @kbd{DOWN} arrows to change the sample zoom by a power of
two.

@kbd{CTRL-UP} and @kbd{CTRL-DOWN} cause the amplitude zoom to change.

@kbd{CTRL-PGUP} and @kbd{CTRL-PGDOWN} cause the track zoom to change.

@kbd{ALT-UP} and @kbd{ALT-DOWN} cause the curve amplitude to change.

@c cincvdoc_node_number_91
@node The track popup menu
@subsection The track popup menu
@cindex The track popup menu

Each Track has a popup menu.  To activate the @b{track popup menu}, RIGHT-click
on the track.  The popup menu affects the track whether the track is armed on the
@b{patch bay} or not.  The Track Menu contains a number of options:
@itemize @bullet
@item Attach Effect - opens a dialog box of effects applicable to the type of track
@item Move up - moves the selected track up in the stack.
@item Move down - moves the selected track down in the stack.
@item Delete track - removes the track from the program
@item Add Track - adds a track of the same media type (audio/video) as the one
selected.
@item Resize Track - resizes the track
@item Match Output - Size resizes the track to match the current output size
@end itemize

@c cincvdoc_node_number_92
@node The insertion point
@subsection The insertion point
@cindex Insertion point

The first time you start Cinelerra, you will see a flashing insertion point in the program window.
Analogous to the cursor on your word processor, the
insertion point marks the place on the timeline where the next activity on the
program will begin.  It is also the starting point of all playback operations.
When rendering, it defines the region of the timeline to be rendered.

@center @image{manual_images_en/program_insertion_point,100mm}
@center @b{The insertion point on the main program,}
@center @b{marked by the vertical hair-line at the 00:00.500 point}

Normally, the insertion point is moved by clicking inside the timebar.  Any
region of the timebar not obscured by labels and in or out points is a hotspot
for repositioning the insertion point.

@center @image{manual_images_en/main_timebar,160mm}
@center @b{The main timebar}

Depending on the mode of operation, the insertion point can be moved by clicking in the timeline itself.
The insertion point has two modes of operation:
@itemize @bullet
@item drag and drop mode
@item cut and paste mode
@end itemize

The mode of operation is determined by selecting the arrow or the i-beam in the
buttonbar.

@center @image{manual_images_en/editing_mode,15mm}
@center @b{The editing mode buttons}

@cindex Drag and drop mode
@cindex Mode, drag and drop
If the arrow is highlighted, it enables @b{drag and drop} mode.  In drag and
drop mode, clicking in the timeline does not reposition the insertion point.
Instead, it selects an entire edit.  Dragging in the timeline repositions the
edit, snapping it to other edit boundaries.  This is useful for
reordering audio playlists and moving effects around.

@cindex Cut and paste mode
@cindex Mode, cut and paste
If the i-beam is highlighted it enables @b{cut and paste mode}.  In cut and
paste mode, clicking in the timeline repositions the insertion point.  Dragging
in the timeline highlights a region.  The highlighted region becomes the
playback range during the next playback operation, the rendered range during
the next render operation, and the region affected by cut and paste operations.

@center @image{manual_images_en/program_highlight,100mm}
@center @b{Tracks with highlighted area, shown inside the green area}

@b{SHIFT-clicking} in the timeline extends the highlighted region.

@b{Double-clicking} in the timeline selects the entire edit the
cursor is over.

When moving the insertion point and selecting regions,
the positions are either aligned to frames or aligned to samples.  When editing
video, you will want to align to frames.  When editing audio you will want to
align to samples.  Select your preference by using @b{settings->align cursor on frames}.

If the highlighted region is the region affected by cut and paste operations,
how do I cut and paste in @b{drag and drop} mode?  In this case, you need to set
@b{in/out points} to define an affected region.

@c cincvdoc_node_number_93
@node The in/out points
@subsection The in/out points
@cindex In/out points

In both editing modes, you can set in/out points.  The in/out points define the
affected region.  In drag and drop mode, they are the only way to define an
affected region.  In both cut and paste mode and drag and drop mode, the
highlighted area overrides the in/out points.  If a highlighted area and in/out
points are set, the highlighted area is affected by editing operations and the
in/out points are ignored.  If no region is highlighted, the in/out points are
used.

Normally, in/out points do not affect the playback region. The in/out points
determine the playback region only if you hold down @key{CTRL} while issuing a
playback command.

To set in/out points, go to the timebar and position the insertion point
somewhere.  Select the @image{manual_images_en/in_point_button,5mm} @b{in point button}.
Go to a position after the in point and click the
@image{manual_images_en/out_point_button,5mm} @b{out point button}.

@center @image{manual_images_en/inout_points,160mm}
@center @b{Timebar with in/out points set}.

If you select either the in point or the out point, the insertion point will jump to
that location.  After selecting an in point, if you click the @b{in point button}
the in point will be deleted.  After selecting an out point, if you click the
@b{out point button} the out point will be deleted.

If you select a region somewhere else while in/out points already exist, the
existing points will be repositioned when you click the in/out buttons.

@b{SHIFT-clicking} on an in/out point highlights the region between the
insertion point and that in/out point.

Instead of using the button bar, you can use the @kbd{[} and @kbd{]} keys to
toggle in/out points.

In both cut and paste editing mode and drag and drop editing mode, in/out
points override the highlighted area.  If a highlighted area and in/out points
are set, the highlighted area affects playback while the in/out points affect
editing operations.  To avoid confusion, it is better to use either highlighting
or in/out points but not both simultaneously.

The insertion point and the in/out points allow you to define an affected
region, but they do not let you jump to exact points on the timeline very easily.
For this purpose there are labels.

@c cincvdoc_node_number_94
@node Using labels in the program window
@subsection Using labels in the program window
@cindex Labels, using in the program window

Labels are an easy way to set exact locations on the timeline that you want to jump
to.  When you position the insertion point somewhere and click the
@image{manual_images_en/label_button,5mm} @b{label button}, a new label appears on the
timeline.

@center @image{manual_images_en/timebar_label,160mm}
@center @b{Timebar with a label on it}

No matter what the zoom settings are, clicking on the label positions the
insertion point exactly where you set it.  Hitting the label button again when
a label is selected deletes it.

@b{SHIFT-clicking} on a label extends the highlighted region.

@b{Double-clicking} between two labels highlights the region between the
labels.

Hitting the @key{l} key has the same effect as the label button.

If you hit the label button when a region is highlighted, labels are
created at each end of the highlighted region.  However, if one end already has a label,
then the existing label is deleted.

Labels can reposition the insertion point when they are selected but they can
also be traversed with the @image{manual_images_en/label_traversal,15mm} @b{label
traversal} buttons.  When a label is out of view, the label traversal buttons
reposition the timeline so the label is visible.  There are keyboard shortcuts
for label traversal, too.

@kbd{CTRL-LEFT} repositions the insertion point on the previous label.

@kbd{CTRL-RIGHT} repositions the insertion point on the next label.

With label traversal you can quickly seek back and forth on the
timeline but you can also select regions.

@kbd{SHIFT-CTRL-LEFT} highlights the region between the insertion point and the
previous label.

@kbd{SHIFT-CTRL-RIGHT} highlights the region between the insertion point and
the next label.

Manually hitting the label button or @key{l} key over and over again to delete
a series of labels can get tedious.  To delete a set of labels, first
highlight a region. Second, use the @b{Edit->Clear labels} function.  If
in/out points exist, the labels between the in/out points are cleared and the
highlighted region is ignored.

@b{Editing or locking labels from moving:} @*
In @b{Cut and Paste} editing mode only, by enabling "Edit labels" in the
settings menu, or by disabling the @image{manual_images_en/locklabels_unlocked,5mm}
@b{"Lock labels from moving"} button on the program toolbar labels will be cut,
copied or pasted along with the selected region of the first armed track. @*
Similarly, if a selected area of a resource is spliced from the viewer to the
timeline in a position before labels, these labels will be pushed to the right
on the timebar for the length of the selected area. @*
To prevent labels from moving on the timebar, just disable the
"Edit labels" option or enable the @image{manual_images_en/locklabels_unlocked,5mm}
@b{"Lock labels from moving"} button. @*
In @b{Drag and Drop} editing mode labels will be always locked to the timebar,
even with the "Edit labels" option enabled.

@c cincvdoc_node_number_95
@node Editing
@section Editing
@cindex Editing

Editing comprises both the time domain and the track domain.  Since the
timeline consists of a stack of tracks, you need to worry about how to sort and
create tracks in addition to what time certain media appears on a track.

In the time domain, Cinelerra offers many ways to approach the editing process.
The three main methods are two screen editing, drag and drop editing, and cut
and paste editing.

There are several concepts Cinelerra uses when editing which apply to all the
methods.  The @b{timeline} is where all editing decisions are represented.
This is a stack of tracks in the center of the main window.  It can be scrolled
up, down, left and right with the scrollbars on the right and bottom of it.  It
can also be scrolled up and down with a mouse wheel.

The @b{active region} is the range of time which is affected by editing
commands on the timeline.  The active region is determined first by the
presence of in/out points in the timeline.  If those do not exist the
highlighted region is used.  If no highlighted region exists the insertion
point is used as the start of the active region.  Some commands treat all the
space to the right of the insertion point as active, like @b{Render}, while
others treat the active length as 0 if no end point for the active region is
defined.

Finally, editing decisions never affect source material.  This is @b{non
destructive editing} and it became popular with audio because it was much
faster than if you had to copy all the media affected by an edit.  Editing only
affects pointers to source material, so if you want to have a media file at the
end of your editing session which represents the editing decisions, you need to
@b{render} it.  @xref{Rendering files}.

Every track on the timeline has a set of attributes on the left, the most
important of which is the @b{arm track} attribute.

@menu
* The patchbay::           Enabling different features on different tracks
* Nudging tracks::         Move entire tracks horizontally
* Panning tracks::         Changing the audio output channels
* Automatic track panning::  Panning tracks to common speaker arrangements
* Standard audio mappings::  Making audio panning that works on other players.
* Manipulating tracks::    Moving whole tracks around
* Two screen editing::     Using two video windows to edit
* Drag and drop editing::  Dragging objects to edit
* Cut and paste editing::  Editing media like text
* Trimming::               Changing in and out points
@end menu

@c cincvdoc_node_number_96
@node The patchbay
@subsection The patchbay
@cindex Patchbay

On the left of the timeline is a region affectionately known as the patchbay.
The patchbay enables features specific to each track.  All tracks have a text
area for naming the track.

All tracks have an @b{expander} @image{manual_images_en/expandpatch_checked,5mm} for
viewing more options and for viewing the effects on the track.  Click on the
expander to expand or collapse the track.  If it is pointing sideways, the
track is collapsed.  If it is pointing down, the track is expanded.  Existing
effects appear below the media for the track.

All tracks have the following row of toggles for several features.

@center @image{manual_images_en/track_attributes}
@center @b{Track attributes}

If the toggle is colored, it is enabled.  If the toggle is the background color
of most of the windows, it is disabled.  Click on the toggle to enable or
disable the feature.  Several mouse operations speed up the configuration of
several tracks at a time.

Click on an attribute and drag across adjacent tracks to copy the same
attribute to those tracks.

Hold down @key{SHIFT} while clicking a track's attribute to enable the
attribute in the current track and toggle the attribute in all the other
tracks.

Hold down @key{SHIFT} while clicking an attribute.  Click until all the tracks
except the selected one are disabled.  Then drag the cursor over the adjacent
track to enable the attribute in the adjacent track.

The other attributes affect the output of the track:
@cindex Play track
@cindex Track, play
@itemize @bullet
@item @b{Play track} @*
Determines whether the track is rendered or not.  If it is off,
the track is not rendered.  However, if the track is chained to any other
tracks, the other tracks perform all the effects in the chained track,
regardless of play status.

@cindex Arm track
@cindex Track, arm
@item @b{Arm track} @*
Determines whether the track is armed or not.  Only the @b{armed
tracks} are affected by editing operations.  Make sure you have enough armed
destination tracks when you paste or splice material or some tracks in the
material will get left out. @*
In addition to restricting editing operations, the armed tracks in combination
with the active region determine where material is inserted when loading files.
If the files are loaded with one of the insertion strategies which do not delete
the existing project, the armed tracks will be used as destination tracks. @*
Press @key{TAB} while the cursor is anywhere over a track to toggle the track
arming status. @*
Press @kbd{SHIFT-TAB} while the cursor is over a track to toggle the arming
status of every other track.

@cindex Gang fader
@cindex Fader, gang
@item @b{Gang fader} @*
Causes the fader to track the movement of whatever other fader
you are adjusting.  A fader is only ganged if the @b{arm track} is also on.
This is normally used to adjust audio levels on all the tracks simultaneously.
Gang also causes @b{Nudge} parameters to synchronize across all the ganged
tracks.

@cindex Draw media
@cindex Media, draw
@item @b{Draw media} @*
Determines if picons or waveforms are drawn on the track.  By
default, some file formats load with this off while other file formats load
with it on.  This depends on whether the file format takes a long time to draw
on the timeline.  Merely set it to on if you want to see picons for any file
format.

@cindex Mute track
@cindex Track, mute
@item @b{Mute track} @*
Causes the output to be thrown away once the track is completely
rendered.  This happens whether or not @b{play track} is on.  If the track is
part of an effect chain, the output of the effect chain track is overlaid on
the final output even though it is routed back to another track.  Mute track is
used to keep the effect chain track from overlapping the output of the source
track.

@cindex Fader
@item @b{Fader} @*
All tracks have a fader, but the units of each fader depend on whether it is
audio or video.  Click and drag the fader to fade the track in and out.  If it
is ganged to other tracks of the same media type, with the @b{arm} option
enabled, the other faders should follow.  Hold down @key{SHIFT} and drag a
fader to center it on 0.
@end itemize

@c cincvdoc_node_number_97
@node Nudging tracks
@subsection Nudging tracks
@cindex Nudging tracks
@cindex Tracks, nudging

Each track has a nudge textbox in its patchbay.  You may have to expand the
track to see it.  These are views of the patchbays when expanded.

@center @image{manual_images_en/apatches}
@center @b{Pan and nudge for an audio track}

@center @image{manual_images_en/vpatches}
@center @b{Overlay mode and nudge for a video track}

The nudge is the amount the track is shifted left or right during playback.
The track is not displayed shifted on the timeline, but it is shifted when it
is played back.  This is useful for synchronizing audio with video, creating
fake stereo, or compensating for an effect which shifts time, all without
tampering with any edits.

Merely enter the amount of time to shift to instantly shift the track.
Negative numbers make the track play later.  Positive numbers make the track
play sooner.  The nudge units are either @b{seconds} or the native units for
the track.  Select the units by @b{right clicking} on the nudge textbox and
using the context sensitive menu.

Nudge settings are ganged with the @b{Gang faders} toggle and the @b{Arm track}
toggle.

Use the mouse wheel over the nudge textbox to increment and decrement it.

@c cincvdoc_node_number_98
@node Panning tracks
@subsection Panning tracks
@cindex Panning tracks
@cindex Tracks, panning

Audio tracks have a panning box in their patchbays.  A patchbay may have to be expanded
to see the panning box.  The panning box is shown here.

@center @image{manual_images_en/apatches}
@center @b{Pan and nudge for an audio track}

Position the pointer in the panning box and click/drag to reposition the audio
output among the speaker arrangement.  The loudness of each speaker is printed
during the dragging operation.  The panning box uses a special algorithm to try
to allow audio to be focused through one speaker or branched between the
nearest speakers when more than 2 speakers are used.

@c cincvdoc_node_number_99
@node Automatic track panning
@subsection Automatic track panning
@cindex Track panning, automatic

Several convenience functions are provided for automatically setting the
panning to several common standards.  They are listed in the @b{Audio} menu.
These functions only affect audio tracks with @b{recording} enabled.

@itemize @bullet
@item @b{Audio->Map 1:1} @*
This maps every track to its own channel and wraps around when all the
channels are allocated.  It is most useful for making 2 tracks with 2 channels
map to stereo and for making 6 tracks with 6 channels map to a 6 channel
soundcard.

@item @b{Audio->Map 5.1:2} @*
This maps 6 tracks to 2 channels.  The project should have 2 channels when
using this function.  Go to @b{Settings->format} to set the output channels to
2.  This is most useful for down-mixing 5.1 audio to stereo.
@end itemize

@c cincvdoc_node_number_100
@node Standard audio mappings
@subsection Standard audio mappings
@cindex Standard audio mappings
@cindex Audio mappings, standard

Although Cinelerra lets you map any audio track to any speaker, there are
standard mappings you should use to ensure the media can be played back
elsewhere.  Also, most audio encoders require the audio tracks to be mapped to
standard speaker numbers or they will not work.

@cindex Channel position
@cindex Position, channel
In the @b{channel position} widget @xref{Project attributes}, the
channels are numbered to correspond to the output tracks they are rendered to.
For stereo, the source of channel 1 needs to be the left track and the source
of channel 2 needs to be the right track.

For 5.1 surround sound, the sources of the 6 channels need to be in the order
of center, front left, front right, back left, back right, low frequency
effects.  If the right tracks are not mapped to the right speakers, most audio
encoders will not encode the right information if they encode anything at all.
The low frequency effects track specifically can not store high frequencies in
most cases.

@c cincvdoc_node_number_101
@node Manipulating tracks
@subsection Manipulating tracks
@cindex Manipulating tracks
@cindex Tracks, manipulating

Tracks in Cinelerra either contain audio or video.  There is no special
designation for tracks other than the type of media they contain.  When you
create a new project, it contains a number of default tracks.  You can
still add or delete tracks from the menus.  The @b{Tracks} menu
contains a number of options for dealing with multiple tracks simultaneously.
Each track itself has a popup menu which affects one track.

Bring up the popup menu by moving over a track and right clicking.  The popup
menu affects the track whether it is armed or not. @*
@cindex Move tracks
@cindex Tracks, move
@b{Move up} and @b{move down} moves the track up or down in the stack.
@b{Delete track} deletes the track.

Operations in the @b{Tracks} menu affect only tracks which are armed:
@itemize @bullet
@item @b{Move tracks up} and @b{Move tracks down} shift all the armed tracks up
or down the stack.
@cindex Delete tracks
@cindex Tracks, delete
@item @b{Delete tracks} deletes the armed tracks.
@item @b{Delete last track} deletes the last track, whether it is armed or not.
Holding down the @b{d} key quickly deletes all the tracks.
@cindex Concatenate tracks
@cindex Tracks, concatenate
@item @b{Concatenate tracks} is more complicated.  It takes every @b{playable}
track and concatenates it to the end of the first @b{armed tracks}.  If there
are two armed tracks followed by two playable tracks, the concatenate operation
puts the two playable tracks after the two armed tracks.  If there are three
playable tracks instead, two tracks are put after the armed tracks and a third
track is put on the end of the first armed track.  The destination track wraps
around until all the playable tracks are concatenated.
@end itemize

Finally, you will want to create new tracks.  The @b{Audio} and @b{Video} menus
each contain an option to add a track of their specific type.  In the case of
audio, the new track is put on the bottom of the timeline and the output
channel of the audio track is incremented by one.  In the case of video, the
new track is put on the top of the timeline.  This way, video has a natural
compositing order.  New video tracks are overlaid on top of old tracks.

@c cincvdoc_node_number_102
@node Two screen editing
@subsection Two screen editing
@cindex Two screen editing

This is the fastest way to construct a program out of movie files.  The idea
consists of viewing a movie file in one window and viewing the program in
another window.  subsections of the movie file are defined in one window and
transferred to the end of the program in the other window.

The way to begin a two screen editing session is to load some resources.  In
@b{file->load} load some movies with the insertion mode @b{create new
resources}.  You want the timeline to stay unchanged while new resources are
brought in.  Go to the Resource Window and select the @b{media} folder.  The
newly loaded resources should appear.  Drag a resource from the media side of
the window over the Viewer window.

There should be enough armed tracks on the timeline to put the subsections of
source material that you want.  If there are not, create new tracks or arm more
tracks.

In the viewer window seek to the starting point of a clip you want to use.  Use
either the @b{slider} or the @b{transport controls}.  Use the @b{preview
region} to narrow down the search.  Set the starting point with the
@image{manual_images_en/in_point_button,5mm} @b{in point button}.

Seek to the ending point of the clip you want to use.  Set the ending point
with the @image{manual_images_en/out_point_button,5mm} @b{out point button}.  The two
points should now appear on the timebar and define a clip.

There are several things you can do with the clip now.

@itemize @bullet
@item @b{Splice} @*
@image{manual_images_en/splice_button,5mm} Inserts the clip in the timeline, pushing
everything back.  If an @b{in point} or @b{out point} exists on the timeline it
is inserted there, otherwise it is inserted after the insertion point.  After
that, the insertion point moves to the end of the clip.  If there is no in/out
point, the insertion point will be used as the next splice location.  This way
you can continuously build up the program by splicing.
@item @b{Overwrite} @*
@image{manual_images_en/overwrite_button,5mm} Overwrites the region of the timeline with
the clip.  If an @b{in point} or @b{out point} exists on the timeline it is
overwritten there, otherwise it is overwritten after the insertion point.  If a
region is highlighted or both in and out points exist the difference between
the active region and the clip length is deleted.
@item @b{Create a clip} @*
@image{manual_images_en/toclip_button,5mm} Generates a new clip for the resource window
containing the affected region but does not change the timeline.  Every clip
has a title and a description.  These are optional.
@item Copy behaves the same as in cut and paste editing.
@end itemize
Two screen editing can be done purely by keyboard shortcuts.  When you move the
pointer over any button a tooltip should appear, showing what key is bound to
that button.  In the Viewer window, the number pad keys control the transport
and the @kbd{[} @kbd{]} @kbd{v} keys perform in/out points and splicing.

@c cincvdoc_node_number_103
@node Drag and drop editing
@subsection Drag and drop editing
@cindex Drag and drop editing
@cindex Editing, drag and drop

@b{Drag and drop editing} is a quick and simple way of working in Cinelerra,
using only the mouse.  The basic idea is to create a bunch of clips, then drag
them in order into the timeline building a prototype film that you can watch on
the compositor.  If after watching it, you wish to re-arrange your clips, set
effects, add transition or insert/delete material, just drag and drop them on
the timeline.

@enumerate 1
@item Load some files using @b{file->load}.
@item Set the insertion mode to @b{Create new resources}.  This loads the files
into the Resource Window.
@item Create some audio and video tracks on the timeline using the video and
audio menus.
@item Open the @b{Media} folder in the resource window.
@item Make sure the necessary tracks are armed and drag a media file from the
resource window to the timeline.  If the media has video, drag it onto a video
track.  If the media is pure audio, drag it onto an audio track.

@center @image{manual_images_en/drag_to_program,70mm}

@end enumerate
Cinelerra fills out the audio and video tracks below the dragging cursor with
data from the file.  This affects what tracks you should create initially and
which track to drag the media onto.  If the media has one video track and two
audio tracks, you will need one video track and two audio tracks on the timeline
and the media should be dragged over the first video track.  If the media has
audio only you will need one audio track on the timeline for every audio track in
the media and the media should be dragged over the first audio track.

When dragging, the media snaps to the start of track if the track is empty.  If
there are edits on the track, the media snaps to the nearest edit boundary.

You can also drag multiple files from the resource window.  Either draw a box
around the files, use @key{SHIFT}, or use @key{CTRL} when selecting files.
When you drop the files in the timeline, they are concatenated.  The behavior
of @key{SHIFT} and @key{CTRL} changes depending on if the resources are in text
or icons.

To display the resources as text or icons, right click inside the media list.
Select either @b{display icons} or @b{display text} to change the list format.

When displaying text in the resource window @b{SHIFT-clicking} on media files
extends the number of highlighted selections.  @b{CTRL-clicking} on media files
in text mode selects additional files one at a time.

When displaying icons in the resource window @b{SHIFT-clicking} or
@b{CTRL-clicking} selects media files one at a time.

In addition to dragging media files, if you create clips and open the @b{clip}
folder you can drag clips on the timeline.

In the timeline there is further dragging functionality.  Dragging edits around
the timeline allows you to sort music playlists, sort movie scenes, and give
better NAB demos but not much else.  To enable the dragging functionality of
the timeline, select the arrow toggle @image{manual_images_en/arrow,2.67mm}.
Move over an edit and drag it.  During a dragging operation the edit snaps to
the nearest boundary.

@center Select a track with various scenes.

@center @image{manual_images_en/drop_before}

@center Original track with three scenes.

@center Go to scene #3, click and drag it to the middle.

@center @image{manual_images_en/drag_track}

@center When you drop scene #3

@center @image{manual_images_en/drop_concept}

@center scene #2 shifts to the right

@center @image{manual_images_en/drop_after}

@center This is how the finished sequence looks.

If more than one track is armed, Cinelerra will drag any edits which start on
the same position as the edit the cursor is currently over.  In other words,
you can drag and drop a group of edits.  Cinelerra recognises as a group the
edits of different armed tracks that have aligned beginnings, regardless of whether they have
the same source or aligned ends.

In Drag and Drop editing mode you can't drag and drop labels. They will be
always locked to the timebar, even with the "Edit labels" option enabled.
Still, with the "Edit labels" option enabled, if a selected area of a resource
is spliced from the Viewer to the timeline in a position before labels, these
labels will be pushed to the right for the length of the selected area.

@c cincvdoc_node_number_104
@node Cut and paste editing
@subsection Cut and paste editing
@cindex Cut and paste editing
@cindex Editing, cut and paste

This is the traditional method of editing in audio editors.  In the case of
Cinelerra, you either need to start a second instance of Cinelerra and copy from
one instance to the other, copy from different tracks in the same instance, or load a
media file into the Viewer and copy from there.

Load some files onto the timeline.  To perform cut and paste editing select the
@image{manual_images_en/ibeam,1.67mm} i-beam toggle.  Select a region of the timeline
and select the @image{manual_images_en/cut,5.67mm} cut button to cut it.  Move the
insertion point to another point in the timeline and select the
@image{manual_images_en/paste,5mm} paste button.  Assuming no in/out points are
defined on the timeline this performs a cut and paste operation.

If in/out points are defined, the insertion point and highlighted region are
overridden by the in/out points for clipboard operations.  Thus, with in/out
points you can perform cut and paste in drag and drop mode as well as cut and
paste mode.

When editing audio, it is customary to cut from one part of a waveform into the
same part of another waveform.  The start and stop points of the cut are
identical in each waveform and might be offset slightly, while the wave data is
different.  It would be very hard to highlight one waveform to cut it and
highlight the second waveform to paste it without changing the relative start
and stop positions.

One option for simplifying this is to open a second copy of Cinelerra, cutting
and pasting to transport media between the two copies.  This way two
highlighted regions can exist simultaneously.

Another option is to set in/out points for the source region of the source
waveform and set labels for the destination region of the destination waveform.
Perform a cut, clear the in/out points, select the region between the labels,
and perform a paste.

A final operation in cut and paste editing is the @b{edit->clear} operation.
If a region is highlighted or in/out points exist, the affected region is
cleared by @b{edit->clear}.  But if the insertion point is over an edit
boundary and the edits on each side of the edit boundary are the same resource,
the edits are combined into one edit comprised by the resource.  The start of
this one edit is the start of the first edit and the end of this one edit is
the end of the second edit.  This either results in the edit expanding or
shrinking.

@c cincvdoc_node_number_105
@node Trimming
@subsection Trimming
@cindex Trimming

With some edits on the timeline it is possible to do trimming.  By trimming you
shrink or grow the edit boundaries by dragging them.  In drag and drop mode or
cut and paste mode, move the cursor over an edit boundary until it changes
shape.  The cursor will either be an expand left or an expand right.  If the
cursor is an expand left, the dragging operation affects the beginning of the
edit.  If the cursor is an expand right, the dragging operation affects the end
of the edit.

When you click on an edit boundary to start dragging, the mouse button number
determines which dragging behavior is going to be followed.  3 possible
behaviors are bound to mouse buttons in the interface preferences.
@xref{Interface}.

The effect of each drag operation not only depends on the behavior button but
whether the beginning or end of the edit is being dragged.  When you release
the mouse button, the trimming operation is performed.

In a @b{Drag all following edits} operation, the beginning of the edit either
cuts data from the edit if you move it forward or pastes new data from before
the edit if you move it backward.  The end of the edit pastes data into the
edit if you move it forward or cuts data from the end of the edit if you move
it backward.  All the edits thereafter shift.  Finally, if you drag the end of
the edit past the start of the edit, the edit is deleted.

In a @b{Drag only one edit} operation, the behavior is the same when you drag
the beginning or end of an edit.  The only difference is none of the other
edits in the track shift.  Instead, anything adjacent to the current edit
expands or shrinks to fill gaps left by the drag operation.

In a @b{Drag source only} operation, nothing is cut or pasted.  If you move the
beginning or end of the edit forward, the source reference in the edit shifts
forward.  If you move the beginning or end of the edit backward, the source
reference shifts backward.  The edit remains in the same spot in the timeline
but the source shifts.

For all file formats besides still images, the extent of the trimming operation
is clamped to the source file length.  Attempting to drag the start of the edit
beyond the start of the source clamps it to the source start.

In all trimming operations, all edits which start on the same position as the
cursor when the drag operation begins are affected.  Unarm tracks to prevent
edits from being affected.

Most effects in Cinelerra can be figured out just by using them and tweeking.
Here are brief descriptions of effects which you might not utilize fully by
mere experimentation.

@c cincvdoc_node_number_106
@node Compositor window
@chapter Compositor window
@cindex Compositor window

This window displays the output of the timeline.  It is the interface for most
compositing operations or operations that affect the appearance of the timeline
output.  Operations done in the Compositor affect the timeline but do not affect
clips.

@menu
* Compositor controls::
* Compositing::
@end menu

@c cincvdoc_node_number_107
@node Compositor controls
@section Compositor controls
@cindex Compositor controls

The video output has several navigation functions.  The video output size is
either locked to the window size or unlocked with scrollbars for navigation.
The video output can be zoomed in and out and panned.  Navigating the video
output this way does not affect the rendered output; it just changes the point
of view in the compositor window.

If it is unlocked from the window size, middle clicking and dragging anywhere
in the video pans the point of view.

Hitting the @kbd{+} and @kbd{-} keys zooms in and out of the video output.

Underneath the video output are copies of many of the functions available in
the main window.  In addition there is a @image{manual_images_en/cwindow_zoom,30mm}
zoom menu and a @image{manual_images_en/cwindow_light,8mm} tally light.

The zoom menu jumps to all the possible zoom settings and, through the @b{Auto}
option, locks the video to the window size.  The zoom menu does not affect the
window size.

The tally light turns red when rendering is happening.  This is useful for
knowing if the output is current.

Right clicking anywhere in the video output brings up a menu with all the zoom
levels and some other options.  In this particular case the zoom levels resize
the entire window and not just the video.

The @b{reset camera} and @b{reset projector} options center the camera and
projector @xref{Compositing}.

@cindex Hide controls
@cindex Controls, hide
The @b{Hide controls} option hides everything except the video.

On the left of the video output is a toolbar specific to the compositor window.
Here are the functions in the toolbar:

@menu
* Protect video::
* Magnifying glass::
* Masks tool::
* Camera::
* Projector::
* Crop tool::
* Eyedropper::
* Tool info::
* Safe regions tool::
@end menu

@c cincvdoc_node_number_108
@node Protect video
@subsection Protect video
@cindex Protect video
@cindex Video, protect

This disables changes to the compositor output from clicks in it.  It is an
extra layer on top of the track arming toggle to prevent unwanted changes.

@c cincvdoc_node_number_109
@node Magnifying glass
@subsection Magnifying glass
@cindex Magnifying glass

This tool @image{manual_images_en/magnify,7mm} zooms in and out of the compositor
output without resizing the window.  If the video output is currently locked to
the size of the window, clicking in it with the magnifying glass unlocks it and
creates scrollbars for navigation.

Left clicking in the video zooms in. @*
Ctrl clicking in the video zooms out. @*
Rotating the wheel on a wheel mouse zooms in and out.

@c cincvdoc_node_number_110
@node Masks tool
@subsection Masks tool
@cindex Masks tool

This tool @image{manual_images_en/mask} brings up the mask editing tool @xref{Masks}.
Enable the @image{manual_images_en/toolwindow,2.67mm} tool window to see options
for this tool.

@c cincvdoc_node_number_111
@node Camera
@subsection Camera
@cindex Camera

This tool @image{manual_images_en/camera,4.67mm} brings up the camera editing tool
@xref{The camera and projector}.  Enable the
@image{manual_images_en/toolwindow,2.67mm} tool window to see options for this
tool.

@c cincvdoc_node_number_112
@node Projector
@subsection Projector
@cindex Projector

This tool @image{manual_images_en/projector,4.67mm} brings up the projector
editing tool @xref{The camera and projector}.  Enable the
@image{manual_images_en/toolwindow,2.67mm} tool window to see options for this
tool.

@c cincvdoc_node_number_113
@node Crop tool
@subsection Crop tool
@cindex Crop tool

This tool @image{manual_images_en/crop,4.33mm} brings up the cropping tool
@xref{Cropping}.  The @image{manual_images_en/toolwindow,2.67mm} tool window must
be enabled to use this tool.

@c cincvdoc_node_number_114
@node Eyedropper
@subsection Eyedropper
@cindex Eyedropper

This brings up the eyedropper.  The eyedropper detects whatever color is under
it and stores it in a temporary area.  Enabling the
@image{manual_images_en/toolwindow,2.67mm} tool info shows the currently selected color.
Click anywhere in the video output to select the color at that point. @*
The eyedropper not only lets you see areas which are clipped, but its value can
be applied to many effects.  Different effects handle the eyedropper
differently.

@c cincvdoc_node_number_115
@node Tool info
@subsection Tool info
@cindex Tool info

This tool @image{manual_images_en/toolwindow,2.67mm} button works only in
conjunction with the other controls on the compositor.  Based on what
compositing control is active the toggle button will activate/deactivate the
appropriate control dialog box.

Controls with dialog boxes are:
@itemize @bullet
@item Edit mask
@item Camera automation
@item Projector automation
@item Crop control
@end itemize

@c cincvdoc_node_number_116
@node Safe regions tool
@subsection Safe regions tool
@cindex Safe regions tool

This tool @image{manual_images_en/titlesafe} draws the safe regions in the video
output.  This does not affect the rendered output @xref{Safe regions}.

@c cincvdoc_node_number_117
@node Compositing
@section Compositing
@cindex Compositing

A large amount of Cinelerra's binary size is directed towards compositing.
When you remove the letterboxing from a widescreen show, you are compositing.
Changing the resolution of a show, making a split screen, and fading in and out
among other things are all compositing operations in Cinelerra.  Cinelerra
detects when it is in a compositing operation and plays back through the
compositing engine only then.  Otherwise, it uses the fastest decoder available
in the hardware.

Compositing operations are done on the timeline and in the Compositor window.
Shortcuts exist in the Resource window for changing some compositing
attributes.  Once some video files are on the timeline, the compositor window
is a good place to try compositing.

@menu
* The camera and projector::
* Masks::
* Cropping::
* Safe regions::
* Overlay modes::
* Track and output sizes::
@end menu

@c cincvdoc_node_number_118
@node The camera and projector
@subsection The camera and projector
@cindex Camera
@cindex Projector

@menu
* The temporary::
* Compositing projector controls::
* Compositing camera controls::
* Popup menu of options::
* The camera and projector tool window::
@end menu

@c cincvdoc_node_number_119
@node The temporary
@subsubsection The temporary
@cindex Temporary

In the compositor window, the most important functions are the
@image{manual_images_en/camera,4.67mm} camera button and the
@image{manual_images_en/projector,4.67mm} projector button.  These control
operation of the camera and projector.  Cinelerra's compositing routines use a
"temporary", a frame of video in memory where all graphics processing is
performed. Inside Cinelerra's compositing pipeline, the camera determines where
in the source video the "temporary" is copied from.  The projector determines
where in the output the "temporary" is copied to.

@center @image{manual_images_en/temporary_explained,140mm}

The process is pretty much as if we scanned in a roll of film one frame at a
time, then (using Gimp, for example) digitally altered the scanned image with
various filters.  Once the image has been transformed by the filters (color
correction, for example) we then project the finished image back into a new
roll of film, thus creating a new "modified" version of the original.

Each track has a different "temporary" which is defined by the track size.  By
resizing the tracks you can create split screens, pans, and zooms.

@center @image{manual_images_en/compositing_pipeline,140mm}
@center @b{Visual representation of the compositing pipeline}

When editing the camera and projector in the compositing window, the first
track with @b{record} enabled is the track affected.  Even if the track is
completely transparent, it is still the affected track.  If multiple video
tracks exist, the easiest way to select one track for editing is to
@b{SHIFT-click} on the record icon of the track.  This solos the track.

@center @image{manual_images_en/projector_concept,120mm}

The purpose of the projector is to place the contents of the "temporary" into the
project's output.  The intent of the projector is to composite several
sources from the various tracks into one final output track.

The projector alignment frame is identical to the camera's viewport, except
that it guides where on the output canvas to put the contents of each
temporary.

@center @image{manual_images_en/projector_screen}

@c cincvdoc_node_number_120
@node Compositing projector controls
@subsubsection Compositing projector controls
@cindex Compositing projector controls
@cindex Projector controls, compositing

When the @b{projector} button is enabled in the compositor window, you are in
projector editing mode.  A guide box appears in the video window.  Dragging
anywhere in the video window causes the guide box to move, hopefully along with
the video.  @b{SHIFT-dragging} anywhere in the video window causes the guide
box to shrink and grow along with the video.  Once you have positioned the
video with the projector, you are ready to master the camera.

@c cincvdoc_node_number_121
@node Compositing camera controls
@subsubsection Compositing camera controls
@cindex Compositing camera controls
@cindex Camera controls, compositing

Select the @image{manual_images_en/camera,4.67mm} camera button to enable camera
editing mode.  In this mode, the guide box shows where the camera position is
in relation to past and future camera positions but not where it is in relation
to the source video.  Dragging the camera box in the compositor window does not
move the box but instead moves the location of the video inside the box.

@cindex Viewport
The @b{viewport} is a window on the camera that frames the area of source video
to be scanned.  The viewport is represented as a red frame with diagonal cross
bars.

@center @image{manual_images_en/camera_concept,100mm}
@center @b{The viewport}

@center @image{manual_images_en/viewport_sizes,150mm}
@center @b{Viewport sizes}

The size of the viewport is defined by the size of the current track.  A
smaller viewport (640x400) captures a smaller area.  A larger viewport
(800x200) captures an area larger than the source video and fills the empty
spaces with blanks.

Once we have our viewport defined, we still need to place the camera right
above the area of source video we are interested on.  To control the location
of the camera:
@enumerate 1
@item Open the compositor window with a track selected.
@item Select the camera button to enable camera editing mode.
@item Drag over the display window.
@end enumerate

When we drag over the viewport in the compositor window (although initially
counter-intuitive), the viewport does not moves but the area of video that sits
under the camera's location does, like when watching the output of a moving camera.

@center @image{manual_images_en/viewport_drag,100mm}
@center @b{In the compositor window, the viewport is always}
@center @b{shown centered, what moves is the video under it}

For example, when you drag the camera down, the viewport in effect is moving
downwards on the video, showing its path towards the bottom of the video, but
from our perspective on the compositor screen, we see the video moving up.
When you drag the camera right, the video seems to move left, and so on.

@b{Note:} The guide box shows where the camera position is in relation to past
and future camera positions, not where it is in relation to the source video.

@c cincvdoc_node_number_122
@node Popup menu of options
@subsubsection Popup menu of options
@cindex Popup menu of options
@cindex Options, popup menu of

In the compositing window, there is a popup menu of options for the camera and
projector.  Right click over the video portion of the compositing window to
bring up the menu.

@itemize @bullet
@item Reset Camera causes the camera to return to the center position.
@item Reset Projector causes the projector to return to the center.
@end itemize

@c cincvdoc_node_number_123
@node The camera and projector tool window
@subsubsection The camera and projector tool window
@cindex The camera and projector tool window

The camera and projector have shortcut operations that do not appear in the popup menu and are not
represented in video overlays.  These are accessed in the @b{Tool window}.
Most operations in the Compositor window have a tool window which is enabled by
activating the @image{manual_images_en/toolwindow,2.67mm} question mark.

@center @image{manual_images_en/compositor_campro_tool,40mm}
@center @b{The camera and projector tool window}

In the case of the camera and projector, the tool window shows x, y, and z
coordinates.  By either tumbling or entering text directly, the camera and
projector can be precisely positioned.  9 justification types are also defined
for easy access.  A popular justification operation is upper left projection
after image reduction.  This is used when reducing the size of video with
aspect ratio adjustment.

@itemize @bullet
@item @image{manual_images_en/button_justify_left} Left
@item @image{manual_images_en/button_justify_centerH} Center Horizontal
@item @image{manual_images_en/button_justify_right} Right
@item @image{manual_images_en/button_justify_top} Top
@item @image{manual_images_en/button_justify_centerV} Center Vertical
@item @image{manual_images_en/button_justify_bottom} Bottom
@end itemize

The translation effect allows simultaneous aspect ratio conversion and
reduction but is easier to use if the reduced video is put in the upper left of
the temporary instead of in the center.  The track size is set to the original
size of the video and the camera is centered.  The output size is set to the
reduced size of the video.  Without any effects, this produces just the cropped
center portion of the video in the output.

The translation effect is dropped onto the video track.  The input dimensions
of the translation effect are set to the original size and the output
dimensions are set to the reduced size.  To put the reduced video in the center
subsection that the projector shows would require offsetting @b{out x and out y}
by a complicated calculation.  Instead, we leave @b{out x and out y} at 0 and
use the projector's tool window.

Merely by selecting @image{manual_images_en/left_justify} left justify and
@image{manual_images_en/top_justify} top justify, the projector displays the reduced
image from the top left corner of the temporary in the center of the output.

@c cincvdoc_node_number_124
@node Masks
@subsection Masks
@cindex Masks

Masks select a region of the video for either displaying or hiding.  Masks are
also used in conjunction with another effect to isolate the effect to a certain
region of the frame.  A copy of one video track may be delayed slightly and
unmasked in locations where the one copy has interference but the other copy
does not.  Color correction may be needed in one subsection of a frame but not
another.  A mask can be applied to just a subsection of the color corrected track
while the vanilla track shows through.  Removal of boom microphones, airplanes,
and housewives are other mask uses.

The order of the compositing pipeline affects what can be done with masks.
Mainly, masks are performed on the temporary after effects and before the
projector.  This means multiple tracks can be bounced to a masked track and
projected with the same mask.

Our compositing pipeline graph now has a masking stage.  There are 8 possible
masks per track.  Each mask is defined separately, although they each perform
the same operation, whether it is addition or subtraction.

@center @image{manual_images_en/compositing_pipeline2,140mm}
@center @b{Compositing pipeline with masks}

To define a mask, go into the Compositor window and enable the
@image{manual_images_en/mask} @b{mask} toggle.  Now go over the video and click-drag.

@b{IMPORTANT:} You have to select @b{automatic keyframes} (@xref{Automatic
keyframes},) if you wish to move a mask over time.  If you do not select
@b{automatic keyframes}, the mask position will be the same even if you edit at
different places on the timeline.

@center @image{manual_images_en/compositor_mask1,60mm}

Click-drag again in another part of the image to create each new point of the
mask.  While it is not the conventional Bezier curve behavior, this masking
interface performs in realtime what the effect of the mask is going to be.
Creating each point of the mask expands a rubber band curve.

Once points are defined, they can be moved by @b{CTRL-dragging} in the vicinity
of the corner.

@center @image{manual_images_en/compositor_mask_drag}
@center @b{CTRL-drag allows you to move existing points to}
@center @b{new locations, thus altering the shape of the mask}

This, however, does not smooth out the curve.  The in-out points of the Bezier
curve are accessed by @b{SHIFT-dragging} in the vicinity of the corner.  Then
@b{SHIFT-dragging} near the in or out point causes the point to move.

@center @image{manual_images_en/compositor_mask_bezier}
@center @b{SHIFT-drag activates belzier handles}

@center @b{to create curves between mask points}

Finally, once you have a mask, the mask can be translated in one piece by
@b{ALT-dragging} the mask.  Mask editing in Cinelerra is identical to how The
Gimp edits masks except in this case the effect of the mask is always on.

@center @image{manual_images_en/compositor_mask_translate}
@center @b{CTRL-ALT-drag translates an entire mask}
@center @b{to a new location on the screen}

The masks have many more parameters which could not be represented with video
overlays.  These are represented in the tool window for masks.  Selecting the
@image{manual_images_en/toolwindow,2.67mm} question mark when the
@image{manual_images_en/mask} mask toggle is highlighted brings up the mask options.

@center @image{manual_images_en/mask_dialog,70mm}
@center @b{Mask options window}

The @b{mode} of the mask determines if the mask removes data or makes data
visible.  If the mode is subtractive, the mask causes video to disappear.  If
the mode is additive, the mask causes video to appear and everything outside
the mask to disappear.

@center @image{manual_images_en/compositor_mask_mode,160mm}
@center @b{Mask mode}

The @b{value} of the mask determines how extreme the addition or subtraction
is.  In the subtractive mode, higher values subtract more alpha.  In the
additive mode, higher values make the region in the mask brighter while the
region outside the mask is always hidden.

@center @image{manual_images_en/composite_mask_value}
@center @b{Mask value}

The mask number determines which one of the 8 possible masks we are editing.
Each track has 8 possible masks.  When you click-drag in the compositor window,
you are only editing one of the masks.  Change the value of @b{mask number} to
cause another mask to be edited.  The previous mask is still active but only
the curve overlay for the currently selected mask is visible.

When multiple masks are used, their effects are ORed together.  Every mask in a
single track uses the same value and mode.

@cindex Feather mask
@cindex Mask, feather
The edges of a mask are hard by default but this rarely is desired.  The
@b{feather} parameter determines how many pixels to feather the mask.  This
creates softer edges but takes longer to render.

@center @image{manual_images_en/compositor_feather,120mm}
@center @b{Feather parameter}

Finally, there are parameters which affect one point on the current mask
instead of the whole mask.  These are @b{Delete, x, y}.  The active point is
defined as the last point dragged in the compositor window.  Any point can be
activated merely by @b{CTRL-clicking} near it without moving the pointer.  Once
a point is activated, @b{Delete} deletes it and @b{x, y} allow repositioning by
numeric entry.

@c cincvdoc_node_number_125
@node Cropping
@subsection Cropping
@cindex Cropping

Cropping changes the value of the output dimensions and the projector to reduce
the visible picture area.  Enable the @image{manual_images_en/crop,4.33mm} crop toggle
and the @image{manual_images_en/toolwindow,2.67mm} tool window in the @b{compositing
window} to display the @b{Crop control dialog box}.

@center @image{manual_images_en/compositor_crop_tool,60mm}
@center @b{Crop control dialog box}

Click-drag anywhere in the video to define the crop area.  This draws a
rectangle over the video.  Click-drag anywhere in the video to start a new
rectangle.  Click-drag over any corner of the rectangle to reposition the
corner.

@center @image{manual_images_en/compositor_crop}
@center @b{Crop area defined}

@b{ALT-click} in the cropping rectangle to translate the rectangle to any
position without resizing it.

The tool window allows text entry of the coordinates and executes the cropping
operation.  When the rectangle is positioned, hit the @b{do it} button in the
tool window to execute the cropping operation.

@b{Note:} The X1,Y1 & X2,Y2 coordinates on the crop control dialog allows text
entry of the top left and bottom right coordinates that define the crop
rectangle.

@c cincvdoc_node_number_126
@node Safe regions
@subsection Safe regions
@cindex Safe regions
@cindex Regions, safe

On consumer displays the borders of the image are cut off and within the
cut-off point is a region which is not always square like it is in the
compositor window.  The borders are intended for scratch room and vertical
blanking data.  You can show where these borders are by enabling the
@image{manual_images_en/titlesafe} safe regions toggle.  Keep titles inside the inner
rectangle and keep action inside the outer rectangle.

@c cincvdoc_node_number_127
@node Overlay modes
@subsection Overlay modes
@cindex Overlay modes

Every video track has an overlay mode, accessible by expanding the track.  The
overlay mode is a pull-down menu on the left under the fader.  When collapsed,
it displays an icon representing the current overlay mode.

Select the @image{manual_images_en/expandpatch_checked} expand track toggle to view all
the options for a video track if you can not see the overlay mode.  The overlay
mode of video tracks is @b{normal} by default.  Select other modes by clicking
the overlay button and selecting an item from the popup menu.

Overlay modes are processed inside the projector stage of compositing.  The
different modes are summarized below.

@cindex Normal overlay mode
@itemize @bullet
@item @b{Normal} @*
This mode uses a traditional Porter-Diff equation to blend tracks with alpha.
When no alpha exists in the project color model, the new track always replaces
the output.

@cindex Addition overlay mode
@item @b{Addition} @*
In this mode, whatever is in the output is added to the current track.
The result is blended based on the current track's alpha onto the output.

@cindex Subtraction overlay mode
@item @b{Subtraction} @*
In this mode, the current track is subtracted from the output and the
result is alpha blended onto the output.

@cindex Multiply overlay mode
@item @b{Multiply} @*
This is the most useful operation.  The current track is multiplied by the
output and the result blended onto the output.  Usually a black and white image
with no alpha channel or a white title on a black image is used as the current
track.  With the multiply operation, only the output portions under the white
area show.

@cindex Divide overlay mode
@item @b{Divide} @*
This mode divides the current track by the output and the result is
blended into the output.  It usually results in overloaded levels.

@cindex Replace overlay mode
@item @b{Replace} @*
This mode does no blending and overwrites the output with the current
track.
@end itemize

@c cincvdoc_node_number_128
@node Track and output sizes
@subsection Track and output sizes
@cindex Track and output sizes
@cindex Sizes, track and output

@menu
* Track size::
* Output size::
@end menu

The size of the temporary and the size of the output in our compositing
pipeline are independent and variable.  This fits into everything covered so
far.  The camera's viewport is the temporary size.  Effects are processed in
the temporary and are affected by the temporary size.  Projectors are rendered
to the output and are affected by the output size.  If the temporary is smaller
than the output, the temporary is bordered by blank regions in the output.  If
the temporary is bigger than the output, the temporary is cropped.

@c cincvdoc_node_number_149
@node Track size
@subsubsection Track size
@cindex Track size
@cindex Size, track

The temporary size is defined as the track size.  Each track has a different
size.  Right click on a track to bring up the track's menu.  Select @b{Resize
Track} to resize the track to any arbitrary size.  Alternatively you can select
@b{Match output size} to make the track the same size as the output.

@center @image{manual_images_en/resize_track,70mm}
@center @b{The resize track window}

For example, the next image shows how a video track and a project output of
equal sizes look when displayed on the compositor.

@center @image{manual_images_en/compositor_output_equal,70mm}
@center @b{Project output size and video}
@center @b{track with equal dimensions (720x480)}

If you resize a track, then its appearance on the compositor changes
accordingly.

Reducing the track (to 640 x 400) and leaving the project's output size
untouched makes the track show on the compositor smaller and framed by a blank
area.

@center @image{manual_images_en/compositor_output_small,70mm}
@center @b{New track (640x400), smaller}
@center @b{than the project's output (720x480)}

Enlarging the track (to 800 x 560) and leaving the project's output size
untouched makes the track show on the compositor larger and cropped to the
output's dimension.

@center @image{manual_images_en/compositor_output_large,70mm}
@center @b{New track (800x560), cropped to}
@center @b{the project's output size (720x480)}

Using this relationship between the track and the project's output size you can
effectively reduce or magnify the size of a particular track with regards to
the final output and therefore create visual "effects" like split screens,
pans, and zooms on the compositor.

@c cincvdoc_node_number_129
@node Output size
@subsubsection Output size
@cindex Output size
@cindex Size, output

The output size is set in either @b{New} when creating a new project or
@b{Settings->Format}.  In the Resource window there is another way to change
the output size.  Right click on a video asset and select @b{Match project
size} to conform the output to the asset.  When new tracks are created, the
track size always conforms to the output size specified by these methods.

When rendering, the project's output size is the final video track size where
the temporary pipeline is rendered into.

If the output size is larger than the temporary then the image transferred from
the temporary will fit inside the Output Track.  Any space left on the Output
is left blank.

@center @image{manual_images_en/output_small}
@center @b{Output size (shown in green) larger than the temporary}

If the output size is smaller than the temporary then some of the temporary
video will be cropped out.

@center @image{manual_images_en/output_large}
@center @b{Output size too small for the temporary}

@c cincvdoc_node_number_130
@node Viewer window
@chapter Viewer window
@cindex Viewer window

The viewer window is a place to load and preview your source media and clips.
Here you can quickly browse through an asset using the @b{slider control},
focus on an area of work with the @b{preview region} or you use @b{editing
controls} to cut & paste segments into the project or create a clip for later
use.

@center @image{manual_images_en/viewer,80mm}
@center @b{The viewer window}

To open the viewer window, go to @b{Window->Show Viewer}

The display is the area on the viewer where you actually see media playing.
Before you can play any media, you first must load it on the viewer.

To load media into the viewer:
@enumerate 1
@item Open the @b{resources manager} window and select the @b{asset manager} or
the @b{clip manager} folder.
@item Drag a file from the @b{asset manager} or the @b{clip manager} to the
viewer

@center @image{manual_images_en/drag_to_viewer,101.75mm}
@end enumerate

You can also load media onto the viewer by right clicking on a file in the
@b{asset manager} and selecting @b{View} from the popup menu or by double
clicking on the icon.

Once your media loads you will see it appear on the display. To play, rewind or
forward through it use the @b{slider control} or the @b{transport controls}.

You can change the media display size by right clicking on the screen to
activate the display zoom menu.  Select zoom levels of 50%, 100% or 200% of the
original media size.

When displaying media, the viewer uses the project's defined output size format
settings, not the original assets format.  You can change the project's output
to match the asset's format using the @b{match project size} menu option in the
@b{asset manager}.

In here you will scrub around source media and clips, selecting regions to paste
into the project.  Operations done in the viewer affect a temporary EDL or a
clip but not the timeline.

@c cincvdoc_node_number_131
@node Resources window
@chapter Resources window
@cindex Resources window

@menu
* Navigating the resources::
@end menu

Effects, transitions, clips, and assets are accessed here.  Most of the
resources are inserted into the project by dragging them out of the resource
window.  Management of resource allocation is also performed here.

@c cincvdoc_node_number_132
@node Navigating the resources
@section Navigating the resources
@cindex Navigating the resources
@cindex Resources, navigating the

The resource window is divided into two areas.  One area lists folders and
another area lists folder contents.  Going into the folder list and clicking on
a folder updates the contents area with the contents of that folder.

@center @image{manual_images_en/resources_audio_effects,60mm}
@center @b{The resources window}

The folder and contents can be displayed as icons or text.

@b{Right clicking} in the folder or contents area brings up a menu containing
formatting options.  Select @b{Display text} to display a text listing.  Select
@b{Sort items} to sort the contents of the folder alphabetically.

The @b{asset info window} displays detailed information about the selected
media file.  To access it, go to the asset manager folder and right click on
the label or icon of the file you are interested on.  An asset menu will
appear, then click on Info.

@center @image{manual_images_en/asset_info,50mm}
@center @b{The asset info window}

@c cincvdoc_node_number_133
@node Sound level meters window
@chapter Sound level meters window
@cindex Sound level meters window

An additional window, the @b{levels window} can be brought up from the
@b{Window} menu.  The @b{levels} window displays the output audio levels after
all mixing is done.

@center @image{manual_images_en/sound_level_meters_window,,80mm}

@center @b{The sound level meters window}

Sound level meters appear in many locations.  They can be toggled in the viewer
and compositor windows with the level toggle.  They appear in the patchbay when
a track is expanded (@xref{The patchbay}.)  They appear in the recording
monitor when audio is being recorded.

The sound levels in the @b{levels window, compositor, and viewer} correspond to
the final output levels before they are clipped to the soundcard range.  In the
@b{record monitor} they are the input values from the sound card.  In the
@b{patchbay} they are the sound levels for each track after all effects are
processed and before down-mixing for the output.

Most of the time, audio levels have numerical markings in dB but in the
patchbay there is not enough room.

The sound level is color coded as an extra means of determining the sound
level.  Even without numerical markings, the sound level color can distinguish
between several ranges and overload.  Look at the color codings in a meter with
numerical markings to see what colors correspond to what sound level.  Then for
meters in the patchbay in expanded audio tracks, use the color codings to see
if it is overloading.

Be aware that sound levels in Cinelerra can go above 0 dB@.  This allows not
only seeing if a track is overloading but how much information is being lost by
the overloading.  Overloading by less than 3 dB is usually acceptable.  While
overloading is treated as positive numbers in Cinelerra, it is clipped to 0
when sent to a sound card or file.

The visible range of the sound level meters is configurable in
@b{settings->preferences->interface} (@xref{Interface}.)

@c cincvdoc_node_number_134
@node Transport controls
@chapter Transport controls
@cindex Transport controls

Transport controls are just as useful in navigation as they are in playing back
footage, hence they are described here in the navigation section.  Each of the
Viewer, Compositor, and Program windows has a transport panel.

@center @image{manual_images_en/transport_panel,130mm}
@center @b{The transport panel}.

The transport panel is controlled by the keyboard as well as the graphical
interface.  For each of the operations it performs, the starting position is
the position of the insertion point in the Program window and the slider in the
Compositor window.  The ending position is either the end or start of the
timeline or the end or start of the selected region if there is one.

The orientation of the end or start depends on the direction of playback.  If
it is forward the end position is the end of the selected region.  If it is
backward the end position is the start of the selected region.

The insertion point moves to track playback.  When playback stops, the
insertion point stays where playback stopped.  Thus, by playing back you change
the position of the insertion point.

The keyboard interface is usually the fastest and has more speeds.  The
transport keys are arranged in a sideways @b{T} on the number pad.

@multitable @columnfractions .08 .17 .08 .17 .08 .17 .08 .17
@item @kbd{4}
@tab Frame back
@tab @kbd{5}
@tab Reverse Slow
@tab @kbd{6}
@tab Reverse
@tab @kbd{+}
@tab Reverse Fast
@item @kbd{1}
@tab Frame Forward
@tab @kbd{2}
@tab Forward Slow
@tab @kbd{3}
@tab Play
@tab @kbd{Enter}
@tab Fast Forward
@item @kbd{0}
@tab Stop
@tab
@tab
@tab
@tab
@tab
@tab
@end multitable

Hitting any key on the keyboard twice pauses it.

When using frame advance functions the behavior may seem odd.  If you frame
advance forward and then frame advance backward, the displayed frame does not
change.  This is because the playback position is not the frame but the time
between two frames.  The rendered frame is the area that the playback position
crosses.  When you increment the time between two frames by one and decrement
it by one, you cross the same frame both times and so the same frame is
displayed.

The transport behavior changes if you hold down @key{CTRL} when issuing any of
the transport commands.  This causes the starting point to be the in point if
playing forward and the out point if playing backward.  If playing forward, the
out point becomes the ending point and if playing backward, the in point
becomes the ending point.  If no in/out points are specified, the behavior
falls back to using the insertion point and track boundaries as the starting
and ending points.

It is possible to use a hardware JogShuttle@footnote{Refer to David Arendt's
message posted on the Cinelerra CV mailing-list on the 2003-11-11 for more
information}

@c cincvdoc_node_number_135
@node Timebar
@chapter Timebar
@cindex Timebar

The navigation features of the Viewer and Compositor behave very similarly.
Each has a timebar and slider below the video output.  The timebar and slider
are critical for navigation.

@center @image{manual_images_en/timebarslider,160mm}

@cindex Preview region
The timebar represents the entire time covered by the program.  When you define
labels and in/out points it defines those, too.  Finally the timebar defines a
region known as the @b{preview region}.

The @b{preview region} is the region of the timeline which the slider affects.
The slider only covers the time covered by the preview region.  By using a
preview region inside the entire program and using the slider inside the
preview region you can quickly and precisely seek in the compositor and viewer.

When you replace the current project with a file the preview region
automatically resizes to cover the entire file.  When you append data or change
the size of the current project, the preview region stays the same size and
shrinks.  Therefore, you need to resize the preview region.

Load a file and then slide around it using the compositor slider.  The
insertion point in the main window follows the compositor.  Move the pointer
over the compositor's timebar until it turns into a left resize pointer.  The
click and drag right.  The preview region should have changed and the slider
resized proportionally.

Go to the right of the timebar until a right resize pointer appears.  Drag left
so the preview region shrinks.

Go to the center of the preview region in the timebar and drag it around to
convince yourself if can be moved.

@b{Note:} When you append data or change the size of the current project, the
preview region stays the same size and shrinks.  Therefore, you need to resize
the preview region.

@center @image{manual_images_en/previewregion,160mm}
@center @b{Preview region in compositor}

If you go to the slider and slide it around with the preview region shrunk,
you will see the slider only affects the preview region.  The timebar and slider
in the viewer window work exactly the same.

Labels and in/out points are fully supported in the viewer and compositor.  The
only difference between the viewer and compositor is the compositor reflects
the state of the program while the viewer reflects the state of a clip but not
the program.

When you hit the @b{label button} in the compositor, the label appears both in
the compositor timebar and the program timebar.

When you select a label or in/out point in the compositor, the program window
jumps to that position.

@center @image{manual_images_en/viewer_labels,160mm}
@center @b{Labels and in/out points in the viewer}.

In the viewer and compositor, labels and in/out points are displayed in the
timebar.  Instead of displaying just a region of the program, the timebar
displays the entire program here.

Like the program window, the compositor has a zoom capability.  First, the
pull-down menu on the bottom of the compositor window has a number of zoom
options.  When set to @b{Auto} the video is zoomed to match the compositor
window size as closely as possible.  When set to any other percentage, the
video is zoomed a power of 2 and scrollbars can be used to scroll around the
output.  When the video is zoomed bigger than the window size, not only do
scrollbars scan around it but @b{middle mouse button} dragging in the video
output scans around it.  This is exactly when The Gimp does.

Furthermore, the zoom @image{manual_images_en/magnify,7mm} toggle causes the
Compositor window to enter zoom mode.  In zoom mode, clicking in the video
output zooms in while @b{ctrl-clicking} in the video output zooms out.  If you
have a wheel mouse, rotating the wheel zooms in or out too.

Zooming in or out with the zoom tool does not change the rendered output, mind
you.  It is merely for scrutinizing video or fitting it in the desktop.

Playing video on the compositor when zoomed to any size other that 100%, the
original size, requires Cinelerra to do extra processing steps.  This could
affect performance on slower systems.

@c cincvdoc_node_number_136
@node Realtime effects
@chapter Realtime effects
@cindex Realtime effects
@cindex Effects, realtime

These are layered under the track they apply to.  They process the track when
the track is played back, with no permanent storage of the output except when
the project is rendered.

All the realtime effects are listed in the resource window, divided into two
groups: audio effects and video effects.  Audio effects should be dragged from
the resource window onto audio tracks.  Video effects should be dragged onto
video tracks.

If there is data on the destination track, the effect is applied to the entire
track.  If there is no data on the track the effect is deleted.  Finally, if a
region of the track is selected the effect is pasted into the region,
regardless of whether there is data.

Some of the effects do not process data but synthesize data.  In the case of a
synthesis effect, you will want to select a region of the track so the dragging
operation pastes it without deleting it.

When dragging more than one effect onto a track, you will see the effects
layering from top to bottom, on the bottom of the track.  When the track is
played back, effects are processed from top to bottom.  The output of the top
effect becomes the input of the bottom effect and so on and so forth.

In addition to dragging from the resource window, effects may be applied to a
track by a popup menu.  Right click on a track and select @b{Attach effect}
from the popup.  The attach effect dialog gives you more control than pure
dragging and dropping.  For one thing, the attach effect dialog lets you attach
two more types of effects: shared effects and shared tracks.  Select a plugin
from the @b{Plugins} column and hit @b{Attach} under the plugins column to
attach it.  The effect is the same as if the effect was dragged from the
resource window.

When an effect exists under a track, it often needs to be configured.  Go
to the effect and right click on it to bring up the effect popup.  In the
effect popup is a @b{show} option.  The show option causes the GUI for the
effect to appear under the cursor.  Most effects have GUI's but some do not.  If
the effect does not have a GUI, nothing pops up when the @b{show} option is
selected.  When you tweek parameters in the effect GUI, the parameters normally
affect the entire duration of the effect.

@menu
* Realtime effect types::
* Editing realtime effects::
* Realtime audio effects::     Realtime audio effects
* Realtime video effects::     Realtime video effects
@end menu

@c cincvdoc_node_number_137
@node Realtime effect types
@section Realtime effect types
@cindex Realtime effect types

The two other effect types supported by the Attach Effect dialog are recycled
effects.  In order to use a recycled effect, three requirements must be met:
@itemize @bullet
@item There must be other effects in the timeline.
@item The other effects must be of the same type as the track you are attaching
an effect to.  If the track is an audio track, the effects must be audio
effects.  If the track is a video track, the effects must be video effects.
@item The insertion point or selected region must start inside the other
effects.
@end itemize

In the case of a shared effect, these conditions must be true.  In the case of
a shared track, there merely must be another track on the timeline of the same
type as the track you are applying an effect to.  If you right clicked on a
video track to attach an effect, there will not be anything in the @b{shared
tracks} column if no other video track exists.  If you right clicked on an
audio track there will not be anything in the shared track column if no other
audio track exists.

If shared effects or shared tracks are available, they appear in the @b{shared
effects} and @b{shared tracks} columns.  The @b{attach} button under each
column causes anything highlighted in the column to be attached under the
current track.

Shared effects and shared tracks allow very unique things to be done.  In the
case of a shared effect, the shared effect is treated like a copy of the
original effect except in the shared effect the GUI can not be brought up.  All
configuration of the shared effect is determined by the GUI of the original
effect and only the GUI of the original effect can be brought up.

When a shared effect is played back, it is processed just like a normal effect
except the configuration is copied from the original effect.  Some effects
detect when they are being shared, like the reverb effects and the compressor.
These effects determine what tracks are sharing them and either mix the two
tracks together or use one track to stage some value.  The reverb mixes tracks
together to simulate ambience.  The compressor uses one of the sharing tracks
as the trigger.

@cindex Bouncing tracks
@cindex Tracks, bouncing
When an original track has a @b{shared track} as one of its effects, the shared
track itself is used as a realtime effect.  This is more commonly known as
@b{bouncing tracks} but Cinelerra achieves the same operation by attaching
shared tracks.  The fade and any effects in the shared track are applied to the
original track.  Once the shared track has processed the data, the original
track performs any effects which come below the shared track and then
composites it on the output.

In addition, once the shared track has processed the output of the original
track like a realtime effect, the shared track mixes itself into the output
with it is settings for pan, mode, and projector.  Thus, two tracks are mixing
the same data on the output.  Most of the times you do not want the shared track
to mix the same data as the original track on the output.  You want it to stop
right before the mixing stage and give the data back to the original track.  Do
this by enabling the @image{manual_images_en/mutepatch_up} mute toggle next to each
track for whom you do not want to mix on the output.

Suppose you were making video and you did want the shared track to composite
the original track's data on the output a second time.  In the case of video,
the video from the shared track would always appear under the video from the
original track, regardless of whether it was on top of the original track.
This is because shared tracks are composited in order of their attachment.
Since it is part of the original track it has to be composited before the
original track is composited.

@c cincvdoc_node_number_138
@node Editing realtime effects
@section Editing realtime effects
@cindex Realtime effects, editing
Many operations exist for manipulating effects once they are in the timeline.
Because mixing effects and media is such complex business, the methods used in
editing effects are not as concise as cutting and pasting.  Some of the editing
happens by dragging in/out points, some of the editing happens through popup
menus, and some of it happens by dragging effects.

Normally when you edit tracks, the effects follow the editing decisions.  If
you cut from a track, the effect shrinks.  If you drag edit in/out points, the
effect changes length.  This behavior can be disabled by selecting
@b{Settings->edit effects} in the project window.  This decouples effects from
editing operations, but what if you just want to edit the effects?

Move the timeline cursor over the effect borders until it changes to a resize
left or resize right icon.  In this state, if you drag the end of the effect,
it performs an edit just like dragging the end of a track does.

The three editing behaviors of track trimming apply to effect trimming and they
are bound to the mouse buttons that you set in @b{interface preferences}.
@xref{Interface}.  When you perform a trim edit on an effect, the effect
boundary is moved by dragging on it.  Unlike track editing, the effect has no
source length.  You can extend the end of an effect as much as desired without
being limited.

Also unlike track editing, the starting position of the drag operation does not
bind the edit decision to media.  The media the effect is bound to does not
follow effect edits.  Other effects, however, do follow editing decisions made
on an effect.  If you drag the end of an effect which is lined up to effects on
other tracks, the effects on the other tracks will be edited while the media
stays the same.

What happens if you trim the end of an effect in, leaving a lot of unaffected
time near the end of the track?  When you drag an effect in from the Resource
Window you can insert the effect in the portion of the row unoccupied by the
trimming operation.  Realtime effects are organized into rows under the track.
Each row can have multiple effects.

In some cases you will want a trimming operation to change only one row of
effects.  This can be achieved by first positioning the insertion point on the
start or end of the effect.  Then press @key{SHIFT} while beginning the
trimming operation.  This causes the operation to change only one row of
effects.

In addition to trimming, you can move effects up or down.  Every track can have
a stack of effects under it.  By moving an effect up or down you change the
order in which effects are processed in the stack.  Go to an effect and right
click to bring up the effect menu.  The @b{Move up} and @b{Move down} options
move the effect up or down.

@cindex Shared effects
@cindex Effects, shared
When you are moving effects up or down, be aware that if they are shared as
@b{shared effects}, any references will be pointing to a different effect after
the move operation.

Finally, there is dragging of effects.  Dragging effects works just like
dragging edits.  You must select the @image{manual_images_en/arrow,2.67mm} arrow
to enter drag and drop mode before dragging effects.  The effects snap to media
boundaries, effect boundaries, and tracks.  Be aware if you drag a reference to
a shared effect, the reference will usually point to the wrong effect
afterwards.

Right click on an effect to bring up a menu for the effect.  Select
@b{attach...} to change the effect or change the reference if it is a shared
effect.

@c cincvdoc_node_number_139
@node Realtime audio effects
@section Realtime audio effects
@cindex Realtime audio effects

@menu
* Compressor::        How to reduce the dynamic range of audio.
* Live audio::        Pass audio from the soundcard directly to the timeline.
* Pitch shift::
* Reverse audio::     How to play regions in reverse.
* Delay audio::
* Denoise::
* DenoiseFFT::
* Despike::
* EQ Parametric::
* Freeverb::
* Gain::
* Heroine College::
* Interpolate::
* Invert Audio::
* Loop audio::
* Overlay::
* SoundLevel::
* Spectrogram::
* Synthesizer::
* Time stretch::
@end menu

@c cincvdoc_node_number_140
@node Compressor
@subsection Compressor
@cindex Compressor

@image{manual_images_en/compressor,12.5mm}

Contrary to computer science experience, the audio compressor does not reduce
the amount of data required to store the audio.  The audio compressor reduces
the dynamic range of the audio.  In Cinelerra the compressor actually performs
the function of an expander and compressor.

The compressor works by calculating the maximum sound level within a certain
time period of the current position.  The maximum sound level is taken as the
input sound level.  For every input sound level there is an output sound level
specified by the user.  The gain at the current position is adjusted so the
maximum sound level in the time range is the user specified value.

The compressor has a graph which correlates every input sound level to an
output level.  The horizontal direction is the input sound level in dB@.  The
vertical direction is the output sound level in dB@.  The user specifies output
sound levels by creating points on the graph.  Click in the graph to create a
point.  If 2 points exist, drag one point across another point to delete it.
The most recent point selected has its vales displayed in textboxes for more
precise adjustment.

To make the compressor reduce the dynamic range of the audio, make all the
output values greater than the input values except 0 dB@.  To make the
compressor expand the dynamic range of the audio, make all the output values
except 0 dB less than the input values.  The algorithm currently limits all
sound levels above 0 dB to 0 dB so to get an overloaded effect put a gain
effect before the compressor to reduce all the levels and follow it with
another gain effect to amplify all the levels back over 0 dB@.

@b{Reaction secs:} This determines where in relation to the current position
the maximum sound level is taken and how fast the gain is adjusted to reach
that peak.  It is notated in seconds.  If it is negative the compressor reads
ahead of the current position to get the future peak.  The gain is ramped to
that peak over one reaction time.  This allows it to hit the desired output
level exactly when the input peak occurs at the current position.

If the reaction time is positive the compressor scans only the current position
for the gain and ramps gain over one reaction time to hit the desired output
level.  It hits the output level exactly one reaction time after detecting the
input peak.

@b{Decay secs:} If the peak is higher than the current level, the compressor
ramps the gain up to the peak value.  Then if a future peak is less than the
current peak it ramps the gain down.  The time taken to ramp the gain down can
be greater than the time taken to ramp the gain up.  This ramping down time is
the decay seconds.

@b{Trigger type:}  The compressor is a multi-channel effect.  Several tracks can
share one compressor.  How the signal from many tracks is interpreted is
determined by the trigger type.

The @b{Trigger type} uses the value supplied in the @b{Trigger} textbox as the
number of the track to use as input for the compressor.  This allows a track
which is not even heard to determine the loudness of the other tracks.

The @b{Maximum} trigger takes the loudest track and uses it as the input for
the compressor.

The @b{Total} trigger type adds the signals from all the tracks and uses the
total as the input for the compressor.  This is the most natural sounding
compression and is ideal when multiple tracks are averaged into single
speakers.

@b{Trigger:} The compressor is a multichannel effect.  Several tracks can share
one compressor.  Normally only one track is scanned for the input peak.  This
track is specified by the @b{Trigger}.  By sharing several tracks and playing
with the trigger value, you can make a sine wave on one track follow the
amplitude of a drum on another track for example.

@b{Smooth only:} For visualizing what the compressor is doing to the
sound-level, this option causes it to replace the sound wave with just the
current peak value.  It makes it very easy to see how @b{reaction secs} affects
the detected peak values.

@c cincvdoc_node_number_141
@node Delay audio
@subsection Delay audio
@cindex Delay audio

@image{manual_images_en/delayaudio,12.5mm}

Just tell how much seconds you want to delay the video track.

@c cincvdoc_node_number_142
@node Denoise
@subsection Denoise
@cindex Denoise

@image{manual_images_en/denoise,13mm}

FIXME

@c cincvdoc_node_number_143
@node DenoiseFFT
@subsection DenoiseFFT
@cindex DenoiseFFT

@image{manual_images_en/denoisefft,13mm}

FIXME

@c cincvdoc_node_number_144
@node Despike
@subsection Despike
@cindex Despike

@image{manual_images_en/despike,12.5mm}

FIXME

@c cincvdoc_node_number_145
@node EQ Parametric
@subsection EQ Parametric
@cindex EQ Parametric

@image{manual_images_en/parametric,12.5mm}

FIXME

@c cincvdoc_node_number_146
@node Freeverb
@subsection Freeverb
@cindex Freeverb

@image{manual_images_en/freeverb,12.5mm}

FIXME

@c cincvdoc_node_number_147
@node Gain
@subsection Gain
@cindex Gain

@image{manual_images_en/gain,12.5mm}

FIXME

@c cincvdoc_node_number_148
@node Heroine College
@subsection Heroine College
@cindex Heroine College

@image{manual_images_en/reverb,12.5mm}

FIXME

@c cincvdoc_node_number_150
@node Interpolate
@subsection Interpolate
@cindex Interpolate

@image{manual_images_en/interpolateaudio,13.25mm}

FIXME

@c cincvdoc_node_number_151
@node Invert Audio
@subsection Invert Audio
@cindex Invert Audio

@image{manual_images_en/invertaudio,12.5mm}

FIXME

@c cincvdoc_node_number_152
@node Live audio
@subsection Live audio
@cindex Live audio effect

@image{manual_images_en/liveaudio,12.5mm}

This effect reads audio directly from the soundcard input.  It replaces any
audio on the track so it is normally applied to an empty track.

To use Live Audio, highlight a horizontal region of an audio track or define in
and out points.  Then drop the Live Audio effect into it.  Create extra tracks
and attach shared copies of the first Live Audio effect to the other tracks to
have extra channels recorded.

Live Audio uses the sound driver selected in
@b{Settings->Preferences->Playback->Audio Out} for recording, but unlike
recording it uses the @b{playback buffer size} as the recording buffer size and
it uses the @b{project sample rate} as the sampling rate.

These settings are critical since some sound drivers can not record in the same
sized buffer they play back in.  Live audio has been most reliable when ALSA is
the recording driver and the playback fragment size is 2048.

Drop other effects after Live Audio to process soundcard input in realtime.

Now the bad news.  With live audio there is no read-ahead, so effects like
compressor will either delay if they have read-ahead enabled or playback will
under-run.

Another problem is sometimes the recording clock on the soundcard is slightly
slower than the playback clock.  The recording eventually falls behind and
playback sounds choppy.

Finally, live audio does not work in reverse.

@c cincvdoc_node_number_153
@node Loop audio
@subsection Loop audio
@cindex Loop audio

@image{manual_images_en/loopaudio,12.5mm}

FIXME

@c cincvdoc_node_number_154
@node Overlay
@subsection Overlay
@cindex Overlay

@image{manual_images_en/overlay,13.25mm}

FIXME

@c cincvdoc_node_number_155
@node Pitch shift
@subsection Pitch shift
@cindex Pitch shift
@cindex Audio, pitch shift

@image{manual_images_en/pitch,12.5mm}

Like the time stretching methods, there are three pitch shifting methods:
@b{Pitch shift}, @b{Resample}, and @b{Asset info dialog}.  Pitch shift is a
realtime effect which can be dragged and dropped onto recordable audio tracks.
Pitch shift uses a fast Fourier transform to try to change the pitch without
changing the duration, but this introduces windowing artifacts.

Because the windowing artifacts are less obtrusive in audio which is obviously
pitch shifted, Pitch shift is mainly useful for extreme pitch changes.  For
mild pitch changes, use @b{Resample} from the @b{Audio->Render Effect}
interface.  Resample can change the pitch within 5% without a noticeable change
in duration.

Another way to change pitch slightly is to go to the @b{Resources} window,
highlight the @b{media} folder, right click on an audio file, click on
@b{Info}.  Adjust the sample rate in the @b{Info} dialog to adjust the pitch.
This method also requires left clicking on the right boundary of the audio
tracks and dragging left or right to correspond to the length changes.

@c cincvdoc_node_number_156
@node Reverse audio
@subsection Reverse audio
@cindex Reverse audio effect

@image{manual_images_en/reverseaudio,12.5mm}

Apply @b{reverse audio} to an audio track and play it backwards.  The sound
plays forward.

Be aware when reversing audio that the waveform on the timeline does not
reflect the actual reversed output.

@c cincvdoc_node_number_157
@node SoundLevel
@subsection SoundLevel
@cindex SoundLevel

@image{manual_images_en/level,12.5mm}

FIXME

@c cincvdoc_node_number_158
@node Spectrogram
@subsection Spectrogram
@cindex Spectrogram

@image{manual_images_en/spectrogram,12.5mm}

FIXME

@c cincvdoc_node_number_159
@node Synthesizer
@subsection Synthesizer
@cindex Synthesizer

@image{manual_images_en/synthesizer,12.5mm}

FIXME

@c cincvdoc_node_number_160
@node Time stretch
@subsection Time stretch
@cindex Time stretch

@image{manual_images_en/timestretch,12.5mm}

FIXME

@c cincvdoc_node_number_161
@node Realtime video effects
@section Realtime video effects
@cindex Realtime video effects

@menu
* 1080 to 480::       How to convert HDTV into SD
* Aging TV::    How to achieve an old movie look
* Blur::
* Brightness/contrast::    How to adjust brightness and contrast
* Burning TV::        How to make your video "burn"
* Chroma key::        Create transparency based on color similarities.
* Chroma key (HSV)::
* Color balance::
* Delay video::
* Denoise video::
* Denoise video2::
* Decimate::          How to reduce frame rates by eliminating similar frames.
* Deinterlace::       How to convert interlaced video to progressive video.
* Difference key::    Create transparency based on color differences.
* DotTV::             How to give a "DotTV" look to your video
* Downsample::        How to reduce the size of an image by throwing out data
* Fields to frames::  How to recover interlaced video from bobbed video
* Flip::              How to flip a video track
* Frames to fields::
* Freeze frame::
* Gamma::
* Gradient::
* Histogram::         How to change the mapping of different brightness values.
* HolographicTV::
* Hue saturation::    How to adjust hue and saturation
* Interpolate video::
* Interpolate pixels:: How to create the illusion of higher framerates.
* Inverse telecine::  How to convert pulled down frames to progressive frames.
* Invert video::
* Linear blur::
* Live video::        Pass video from the capture card directly to the timeline.
* Loop video::        How to loop regions of the timeline.
* Motion::            The art of motion tracking.
* Motion blur::
* Oil painting::      How to make your video look like a painting
* Overlay video::
* Perspective::       How to modify the perspective of a video track
* Polar::             How to bend and wrap your video
* RGB-601::
* Radial blur::
* ReframeRT::         Changing the number of frames in a sequence.
* Reverse video::     How to play regions in reverse.
* Rotate::            How to rotate your video
* SVG via Inkscape::
* Scale::
* Selective temporal averaging::
* Sharpen::
* ShiftInterlace::
* Swap channels::
* Threshold::         How to get monochrome out of a region of the image.
* Time average::      How to add trail patterns or increase still image quality.
* TimeFront::
* Title::             How to add text to a track from inside Cinelerra.
* Translate::
* Unsharp::           How to unsharp your video
* Videoscope::        How to view the dynamic range of intensity and hue.
* Wave::
* Whirl::
* YUV::
* Zoom blur::
@end menu

@c cincvdoc_node_number_162
@node 1080 to 480
@subsection 1080 to 480
@cindex 1080 to 480 effect

@image{manual_images_en/1080to540,12.5mm}

Most TV broadcasts are received with a 1920x1080 resolution but originate from
a 720x480 source at the studio.  It is a waste of space to compress the entire
1920x1080 if the only resolvable details are 720x480.  Unfortunately resizing
1920x1080 video to 720x480 is not as simple as shrinking it.

At the TV station the original 720x480 footage was first converted to fields of
720x240.  Each field was then scaled up to 1920x540.  The two 1920x540 fields
were finally combined with interlacing to form the 1920x1080 image.  This
technique allows a consumer TV to display the re-sampled image without extra
circuitry to handle 720x480 interlacing in a 1920x1080 image.

If you merely deinterlace the 1920x1080 images, you would end up with
resolution of 720x240.  The @b{1080 to 480} effect properly extracts two
1920x540 size fields from the image, resizes them separately, and combines them
again to restore a 1920x480 interlaced image.  The @b{scale} effect must then
be applied to reduce the horizontal size to 960 or 720 depending on the
original aspect ratio.

The tracks to which @b{1080 to 480} is applied need to be at 1920x1080
resolution.  The project settings in @b{settings->format} should be at least
720x480 resolution.

The effect does not know if the first row in the 1920x1080 image belongs to the
first row of the 720x480 original.  You have to specify what the first row is
in the effect configuration.

The output of this effect is a small image in the middle of the original
1920x1080 frame.  Use the projector to center the output image in the playback.

Finally, once you have 720x480 interlaced video you can either apply @b{frames
to fields} of @b{inverse telecine} to further recover original progressive
frames.

@c cincvdoc_node_number_163
@node Aging TV
@subsection Aging TV
@cindex Aging TV

@image{manual_images_en/aging,17mm}

This effect is the one to use if you want to achieve an "old movie" or TV show
look.  It will put moving lines up and down the movie as well as putting "snow"
on the video.  Use is along with Brightness/Contrast and Color Balance to make
your movie look like a really old black and white movie.

@c cincvdoc_node_number_164
@node Blur
@subsection Blur
@cindex Blur

@image{manual_images_en/blur,12.5mm}

This effect blurs a video track.  The parameters are:
@itemize @bullet
@item @b{Horizontal and vertical} @*
Those parameters are used to tell which one of field blurring affects.  It
can be be both fields.
@item @b{Radius} @*
Use this slider to define the amount of blur to apply.
@item @b{Blur alpha, red, green, blue} @*
Specifies which color channels has to be blurred.
@end itemize

@c cincvdoc_node_number_165
@node Brightness/contrast
@subsection Brightness/contrast
@cindex Brightness/contrast
@cindex Contrast

@image{manual_images_en/brightness,12.5mm}

If you want to brighten a dark shot, or add light, this is the tool to use.  Do
not overuse the effect or you risk degrading your video quality.  Use the effect along
with Keyframing to brighten a long shot that is dark at the beginning but bright
at the end.  Generally you will want to change the brightness and contrast
about the same amount (eg darkness 28 contrast 26) so that your original colors
are kept intact.

@c cincvdoc_node_number_166
@node Burning TV
@subsection Burning TV
@cindex Burning TV
@cindex Video burning

@image{manual_images_en/burn,16mm}

The video burning effect makes your video "burn" where there are small light
colored patches of video, on the edge of a white T-shirt for example.  It can
be a great asset to a music video and just a great outlet to help free your
imagination in your video.

@c cincvdoc_node_number_167
@node Chroma key
@subsection Chroma key
@cindex Chroma key effect

@image{manual_images_en/chromakey,12.5mm}

This effect erases pixels which match the selected color.  They are replaced
to black if there is no alpha channel and transparency if there is an alpha
channel.  The selection of color model is important to determine the behavior.

@cindex Color picker
@cindex Picker, color
Chroma key uses either the lightness or the hue to determine what is erased.
@b{Use value} singles out only the lightness to determine transparency.  Select
a center color to erase using the @b{Color} button.  Alternatively a color can
be picked directly from the output frame by first using the @b{color picker} in
the compositor window and then selecting the @b{Use color picker} button.  This
sets the chroma key color to the current color picker color.

Be aware that the output of the chroma key is fed back to the compositor, so
selecting a color again from the compositor will use the output of the chroma
key effect.  The chroma key should be disabled when selecting colors with the
color picker.

If the lightness or hue is within a certain threshold it is erased.  Increasing
the threshold determines the range of colors to be erased.  It is not a simple
on/off switch, however.  As the color approaches the edge of the threshold, it
gradually gets erased if the slope is high or is rapidly erased if the slope is
low.  The slope as defined here is the number of extra values flanking the
threshold required to go from opaque to transparent.

Normally threshold is very low when using a high slope.  The two parameters
tend to be exclusive because slope fills in extra threshold.

The slope tries to soften the edges of the chroma key but it does not work well
for compressed sources.  A popular softening technique is to use a maximum
slope and chain a blur effect below the chroma key effect to blur just the
alpha.

@c cincvdoc_node_number_168
@node Chroma key (HSV)
@subsection Chroma key (HSV)
@cindex Chroma key effect (HSV)

@image{manual_images_en/chromakeyhsv,12.5mm}

FIXME

@c cincvdoc_node_number_169
@node Color balance
@subsection Color balance
@cindex Color balance
@cindex Balance, color

@image{manual_images_en/colorbalance,12.5mm}

Video Color Balance is a great effect to use along with Brightness/contrast and
Hue/Saturation to try and compensate for possible errors in filming (low
lighting, etc).  It can only do so much without greatly lowering the quality of
the video, however.  It is just like the color balance effect on a picture
editing program, such as GIMP@.  With it you can change the colors being sent to
output CMY (Cyan, Magenta, Yellow) or RGB (Red, Green, Blue).

@c cincvdoc_node_number_170
@node Decimate
@subsection Decimate
@cindex Decimate effect

@image{manual_images_en/decimate,13.25mm}

This effect drops frames from a track which are most similar in order to reduce
the frame rate.  This is usually applied to a DVD to convert the 29.97 fps
video to the 23.97 fps film rate but this decimate effect can take any input
rate and convert it to any lower output rate.

The output rate of @b{decimate} is the project frame rate.  The input rate is
set in the @b{decimate} user interface.  To convert 29.97 fps progressive video
to 23.97 fps film, apply a decimate effect to the track.  Set the decimate
input rate to 29.97 and the project rate to 23.97.

Be aware every effect layered before decimate processes video at the decimate
input rate and every effect layered after decimate processes video at the
project frame rate.  Computationally intensive effects should come below
decimate.

@c cincvdoc_node_number_171
@node Deinterlace
@subsection Deinterlace
@cindex Deinterlace effect

@image{manual_images_en/deinterlace,12.5mm}

The deinterlace effect has evolved over the years to deinterlacing and a whole
lot more.  In fact two of the deinterlacing methods, @b{Inverse Telecine} and
@b{Frames to Fields}, are separate effects.  The deinterlace effect offers
several variations of line replication to eliminate comb artifacts in
interlaced video.  It also has some line swapping tools to fix improperly
captured video or make the result of a reverse effect display fields in the
right order.

@c cincvdoc_node_number_172
@node Delay video
@subsection Delay video
@cindex Delay video

@image{manual_images_en/delayvideo,12.5mm}

FIXME

@c cincvdoc_node_number_173
@node Denoise video
@subsection Denoise video
@cindex Denoise video

@image{manual_images_en/denoisevideo,12.5mm}

FIXME

@c cincvdoc_node_number_174
@node Denoise video2
@subsection Denoise video2
@cindex Denoise video

@image{manual_images_en/denoisemjpeg,12.5mm}

FIXME

@c cincvdoc_node_number_175
@node Difference key
@subsection Difference key
@cindex Difference key effect

@image{manual_images_en/diffkey,12.5mm}

The difference key creates transparency in areas which are similar between 2
frames.  The Difference key effect must be applied to 2 tracks.  One track
contains the action in front of a constant background and another track
contains the background with nothing in front of it.  Apply the difference key
to the track with the action and apply a shared copy of it to the track with
the background.  The track with the background should be muted and underneath
the track with the action and the colormodel should have an alpha channel.

Pixels which are different between the background and action track are treated
as opaque.  Pixels which are similar are treated as transparent.  Change
@b{threshold} in the difference key window to make more pixels which are not the
same color transparent.  Change @b{slope} to change the rate at which the
transparency tapers off as pixels get more different.

The slope as defined here is the number of extra values flanking the threshold
required to go from opaque to transparent.  A high slope is more useful with a
low threshold because slope fills in extra threshold.

@b{Use value} causes the intensity of pixels to be compared instead of the
color.

Applying a blur to the top track with just the alpha channel blurred can soften
the transparency border.

@b{Note:} Currently this effect is known to crash when using YUV modes.

@c cincvdoc_node_number_176
@node DotTV
@subsection DotTV
@cindex DotTV

@image{manual_images_en/dot,16.5mm}

Another effect by Kentaro (effectTV).

@c cincvdoc_node_number_177
@node Downsample
@subsection Downsample
@cindex Downsample

@image{manual_images_en/downsample,12.5mm}

Downsampling is the process of reducing the size of an
image by throwing out data, reducing sampling rate.

Parameters refers to: @*
Horizontal @*
Horizontal offset @*
Vertical @*
Vertical offset @*
Channels

@c cincvdoc_node_number_178
@node Fields to frames
@subsection Fields to frames
@cindex Fields to frames effect

@image{manual_images_en/fieldframe,12.5mm}

This effect reads frames at twice the project framerate, combining 2 input
frames into a single interlaced output frame.  Effects preceding @b{fields to
frames} process frames at twice the project frame rate.  Each input frame is
called a field.

@b{Fields to frames} needs to know what field corresponds to what lines in the
output frame.  The easiest way to figure it out is to try both options in the
window.  If the input fields are the result of a line doubling process like
@b{frames to fields}, the wrong setting results in blurrier output.  If the
input fields are the result of a standards conversion process like @b{1080 to
480}, the wrong setting will not make any difference.

The debobber which converts 720x480 interlaced into 1920x1080 interlaced or
1280x720 progressive seems to degrade the vertical resolution to the point that
it can not be recovered.

@c cincvdoc_node_number_179
@node Flip
@subsection Flip
@cindex Flip

@image{manual_images_en/flip,12.5mm}

This effect permits to flip a video track (or a portion of) from left to right,
right to left, up to down, down to up.

The dialog window is simple, since only the vertical and horizontal parameters
are needed.

@c cincvdoc_node_number_180
@node Frames to fields
@subsection Frames to fields
@cindex Frames to fields

@image{manual_images_en/framefield,13.25mm}

FIXME

@c cincvdoc_node_number_181
@node Freeze frame
@subsection Freeze frame
@cindex Freeze frame effect

@image{manual_images_en/freezeframe,12.5mm}

In its simplest form, highlight a region of the track to freeze, drop the
freeze frame effect on the highlighted region, and the lowest numbered frame in
the affected area will play throughout the entire region.

Freezeframe has an @b{enabled} option which can be keyframed.  Regions of a
freeze frame effect which are enabled repeat the lowest numbered frame since
the last keyframe.  This has unique possibilities.

@itemize @bullet
@item If a freeze frame effect has a keyframe in the middle of it set to
@b{enabled}, the frame in the middle is repeated in the entire effect.
@item If a freeze frame effect has several keyframes, each set to @b{enabled},
every time a keyframe is encountered the frame under it becomes the frozen one.
@item If a freeze frame effect alternates between @b{enabled} and @b{disabled},
each time an @b{enabled} keyframe is encountered the frame under it is
replicated until the next @b{disabled} keyframe.  The disabled regions play
through.
@end itemize

@c cincvdoc_node_number_182
@node Gamma
@subsection Gamma
@cindex Gamma effect
@cindex Raw camera images

@image{manual_images_en/gamma,12.5mm}

Raw camera images store colors in a logarithmic scale.  The blacks in these
images are nearly 0 and the whites are supposed to be infinity.  The graphics
card and most video codecs store colors in a linear scale but Cinelerra keeps
raw camera images in their original logarithmic scale when it renders them.
This is necessary because the raw image parser can not always decode the proper
gamma values for the images.  It also does its processing in 16 bit integers,
which takes away a lot of information.

The gamma effect converts the logarithmic colors to linear colors through a
gamma value and a maximum value.  The gamma value determines how steep the
output curve is and the maximum value is where 1.0 in the output corresponds to
maximum brightness in the input.

The gamma effect has 2 more parameters to simplify gamma correction.  The
@b{automatic} option causes it to calculate @b{max} from the histogram of the
image.  Use this when making a preview of a long list of images since it
changes for every image.

The @b{use color picker} option uses the value currently in the color picker to
set the @b{max} value.  Note that every time you pick a color from the
compositor window, you need to hit @b{use color picker} to apply the new value.

@c cincvdoc_node_number_183
@node Gradient
@subsection Gradient
@cindex Gradient

@image{manual_images_en/gradient,12.5mm}

FIXME

@c cincvdoc_node_number_184
@node Histogram
@subsection Histogram
@cindex Histogram effect

@image{manual_images_en/histogram,12.5mm}

This shows the number of occurrences of each color on a histogram plot.

It is always performed in floating point RGB regardless of the project
color-space.  The histogram has two sets of transfer parameters: the input
transfer and the output transfer.

4 histograms are possible in the histogram viewer.  The red, green, blue
histograms show the input histograms for red, green, blue and multiply them by
an input transfer to get the output red, green, blue.  Then the output red,
green, blue is scaled by an output transfer.  The scaled red, green, blue is
converted into a value and plotted on the value histogram.  The value histogram
thus changes depending on the settings for red, green, blue.  The value
transfers are applied uniformly to R, G, B after their color transfers are
applied.

Select which transfer to view by selecting one of the channels on the top of
the histogram.

The input transfer is defined by a graph overlaid on the histogram.  The
horizontal direction corresponds to every possible input color.  The vertical
direction corresponds to the output color for every input color.  Video
entering the histogram is first plotted on the histogram plot, then it is
translated so output values now equal the output values for each input value on
the input graph.

The input graph is edited by adding and removing any number of points.  Click
and drag anywhere in the input graph to create a point and move it.  Click on
an existing point to make it active and move it.  The active point is always
indicated by being filled in.  The active point's input and output color are
given in text boxes on top of the window.  The input and output color of the
point can be changed through these text boxes.

Points can be deleted by first selecting a point and then dragging it to the
other side of an adjacent point.  They can also be deleted by selecting them
and hitting @b{delete}.

After the input transfer, the image is processed by the output transfer.  The
output transfer is simply a minimum and maximum to scale the input colors to.
Input values of 100% are scaled down to the output's maximum.  Input values of
0% are scaled up to the output minimum.

Input values below 0 are always clamped to 0 and input values above 100% are
always clamped to 100%.  Click and drag on the output gradient's triangles to
change it.  It also has textboxes to enter values into.

Enable the @b{automatic} toggle to have the histogram calculate an automatic
input transfer for the red, green, blue but not the value.  It does this by
scaling the middle 99% of the pixels to take 100% of the histogram width.  The
number of pixels permitted to pass through is set by the @b{Threshold} textbox.
A threshold of 0.99 scales the input so 99% of the pixels pass through.
Smaller thresholds permit fewer pixels to pass through and make the output look
more contrasty.

Automatic input transfer is calculated for the R, G, and B channels but not the
value. @*
@b{Plot histogram} @*
@b{Split output}

@c cincvdoc_node_number_185
@node HolographicTV
@subsection HolographicTV
@cindex HolographicTV

@image{manual_images_en/holo,16.25mm}

By Kentarou effectTV

@c cincvdoc_node_number_186
@node Hue saturation
@subsection Hue saturation
@cindex Hue saturation
@cindex Saturation

@image{manual_images_en/huesaturation,12.5mm}

With that effect you can change hue, saturation and value.  The parameters are
modified using 3 simple sliders.

@c cincvdoc_node_number_187
@node Interpolate video
@subsection Interpolate video
@cindex Interpolate video

@image{manual_images_en/interpolatevideo,13.25mm}

FIXME

@c cincvdoc_node_number_188
@node Interpolate pixels
@subsection Interpolate pixels
@cindex Interpolate pixels

@image{manual_images_en/interpolate,12.5mm}

The interpolate pixels effect tries to create the illusion of a higher frame
rate from source footage of very low framerates by averaging frames over time.
It averages two input frames for each output frame.  The input frames are at
different times, resulting in a dissolve for all output frames between the
input frames.  There are two ways of specifying the input frames.  You can
specify an input frame rate which is lower than the project frame rate.  This
causes input frames to be taken at even intervals,

You can also specify keyframe locations as the positions of the input frames.
In this mode the output frame rate is used as the input frame rate and you just
create keyframes wherever you want to specify an input frame.

@c cincvdoc_node_number_189
@node Inverse telecine
@subsection Inverse telecine
@cindex Inverse telecine effect

@image{manual_images_en/ivtc,12.5mm}

This is the most effective deinterlacing tool when the footage is a video
transfer of a film.  Here the film was converted from 24 fps to 60 fps.  Then
the 60 fps was down-sampled to 30 fps by extracting odd and even lines and
interlacing the lines.  The IVTC effect is primarily a way to convert
interlaced video to progressive video.  It undoes three patterns of
interlacing. @*
@code{A AB BC CD D} @*
@code{AB CD CD DE EF} @*
@code{Automatic}

The first two options are fixed patterns and affected by the @b{pattern offset}
and @b{odd field first} parameters.  The last option creates several
combinations of lines for each frame and picks the most progressive
combination.  It is a brute force algorithm.

This technique does not rely on a pattern like other techniques and is less
destructive but the timing is going to be jittery because of the lack of a
frame rate reduction.  In order to smooth out the timing, you need to follow
inverse telecine with a decimate effect.

@c cincvdoc_node_number_190
@node Invert video
@subsection Invert video
@cindex Invert video

@image{manual_images_en/invertvideo,12.5mm}

Also called invert video, it is a method of reversing the colors of a video
track.

The three parameters refer to channels (Red, Blue, Green, Alpha)

@c cincvdoc_node_number_191
@node Linear blur
@subsection Linear blur
@cindex Linear blur

@image{manual_images_en/linearblur,12.5mm}

Blur has three styles: Linear, Radial, and Zoom

Parameters refer to:
@itemize @bullet
@item @b{Length} @*
Distance between original image and final blur step
@item @b{Angle} @*
Angle of motion, for linear blur
@item @b{Steps} @*
Number of blur steps
@item @b{Channels} @*
Which channel(s) to blur.
@end itemize

@c cincvdoc_node_number_192
@node Live video
@subsection Live video
@cindex Live video effect

@image{manual_images_en/livevideo,12.5mm}

This effect reads video directly from the capture card input.  It replaces any
video on the track so it is normally applied to an empty track.  The
configuration for the capture card is taken from the recording preferences.  Go
to @b{Settings->Preferences->Recording} to set up the capture card.

Go to the @b{Video In} section where it says @b{Record driver}.  It must be set
to either @b{Video4Linux2} or @b{IEC 61883}.  Other video drivers have not been
tested with Live Video and probably will not work.

For live video, the selection for @b{File Format} and @b{Video} needs to be set
to a format the timeline can use.  The file format must be @b{Quicktime for
Linux} and video recording must be enabled for it.  Click on the wrench
@image{manual_images_en/wrench,4.33mm} to set the video compression.

The video compression depends on the recording driver.  For the
@b{Video4Linux2} recording driver, the compression must be @b{Motion JPEG A}.
For the @b{IEC 61883} driver, the compression must be @b{DV}.  This gets the
driver to generate output in a colormodel that the timeline can use.

Some cards provide color and channel settings.  Live video takes the color
settings from the values set in the @b{Video In} window.  Go to
@b{File->Record} to bring up the recording interface and the Video In window.
Values set in the @b{Video in} window are used by @b{Live Video}.  Any channels
the capture card supports need to be configured in the @b{Video in} interface
since the same channels are used by the @b{Live Video} effect.

With the video recording configured, highlight a horizontal region of a video
track or define in and out points.  Then drop the Live Video effect into it.
Drop other effects after Live Video to process the live video in realtime.  For
best results, you should use OpenGL and a video card which supports GL shading
language.  Go to @b{Settings->Preferences->Playback->Video Out} to enable the
OpenGL driver.

Only one Live Video effect can exist at any time on the timeline.  It can not be
shared by more than one track.

@c cincvdoc_node_number_193
@node Loop video
@subsection Loop video
@cindex Loop video effect

@image{manual_images_en/loopvideo,12.5mm}

Sections of video can be looped by dropping a @b{loop} effect on them.
Contrary to the @b{settings->loop playback} option, the loop effects can be
rendered where the @b{settings->loop playback} option can not be.  The loop
effects are also convenient for short regions.

The loop effects have one option: the number of @b{frames} or @b{samples} to
loop.  This specifies the length of the region to loop starting from either the
beginning of the effect or the latest keyframe.  The region is replicated for
the entire effect.

Every time a keyframe is set in a loop effect, the keyframe becomes the
beginning of the region to loop.  Setting several keyframes in succession
causes several regions to loop.  Setting a single keyframe causes the region
after the keyframe to be looped throughout the effect, no matter where the
keyframe is.  The end of an effect can be looped from the beginning by setting
the keyframe near the end.

@c cincvdoc_node_number_194
@node Motion
@subsection Motion
@cindex Motion video effect

@image{manual_images_en/motion,12.5mm}

The motion tracker is almost a complete application in itself.  The motion
tracker tracks two types of motion: translation and rotation.  It can track
both simultaneously or one only.  It can do 1/4 pixel tracking or single pixel
tracking.  It can stabilize motion or cause one track to follow the motion of
another track.

Although the motion tracker is applied as a realtime effect, it usually must be
rendered to see useful results.  The effect takes a long time to precisely
detect motion.

The motion tracker works by using one region of the frame as the region to
track.  It compares this region between 2 frames to calculate the motion.  This
region can be defined anywhere on the screen.  Once the motion between 2 frames
has been calculated, a number of things can be done with that motion vector.
It can be scaled by a user value and clamped to a maximum range.  It can be
thrown away or accumulated with all the motion vectors leading up to the
current position.

To save time the motion result can be saved for later reuse, recalled from a
previous calculation, or discarded.

The motion tracker has a notion of 2 tracks, the master layer and the target
layer.  The master layer is where the comparison between 2 frames takes place.
The target layer is where motion is applied either to track or compensate for
the motion in the master layer.

The intricacies of motion tracking are enough to sustain entire companies and
build careers around.  The motion tracker in Cinelerra is not as sophisticated
as some world class motion trackers but it is enough to sweeten some camcorder
footage.

Here is a brief description of the motion tracking parameters:

@itemize @bullet

@item @b{Track translation} @*
Enables translation operations.  The motion tracker tracks X and Y motion in
the master layer and adjusts X and Y motion in the target layer.

@cindex Motion effect, translation block
@cindex Translation block
@item @b{Translation block size} @*
For the translation operations, a block is compared to a number of neighboring
blocks to find the one with the least difference.  The size of the block to
search for is given by this parameter.

@item @b{Translation search radius} @*
The size of the area to scan for the translation block.

@item @b{Translation search steps} @*
Ideally the search operation would compare the translation block with every
other pixel in the translation search radius.  To speed this operation up, a
subset of the total positions is searched.  Then the search area is narrowed
and rescanned by the same number of search steps until the motion is known to
1/4 pixel accuracy.

@item @b{Block X, Y} @*
These coordinates determine the center of the translation block based on
percentages of the width and height of the image.  The center of the block
should be part of the image which is visible at all times.

@item @b{Maximum absolute offset} @*
The amount of motion detected by the motion tracker is unlimited if this is
100.  If it is under 100 the amount of motion is limited by that percentage of
the image size.

@item @b{Settling speed} @*
The motion detected between every frame can be accumulated to form an absolute
motion vector.  If the settling speed is 100 the absolute vector is added to
the next frame.  If the settling speed is less than 100 the absolute vector is
downscaled by the settling amount before being added to the next frame.

@item @b{Track rotation} @*
Enables rotation operations.  The motion tracker tracks rotation in the master
layer and adjusts rotation in the target layer.

@item @b{Rotation block size} @*
For rotation operations a single block is compared to equally sized blocks,
each rotated by a different amount.  This is the size of the rotation block.

@item @b{Rotation search radius} @*
This is the maximum angle of rotation from the starting frame the rotation
scanner can detect.  The rotation scan is from this angle counterclockwise to
this angle clockwise.  Thus the rotation search radius is half the total range
scanned.

@item @b{Rotation search steps} @*
Ideally every possible angle would be tested to get the rotation.  To speed up
the rotation search, the rotation search radius is divided into a finite number
of angles and only those angles compared to the starting frame.  Then the
search radius is narrowed and an equal number of angles is compared in the
smaller radius until the highest possible accuracy is achieved. @*
Normally you need one search step for every degree scanned.  Since the rotation
scanner scans the rotation search radius in two directions, you need two steps
for every degree in the search radius to search the complete range.

@item @b{Draw vectors} @*
When translation is enabled, 2 boxes are drawn on the frame.  One box
represents the translation block.  Another box outside the translation block
represents the extent of the translation search radius.  In the center of these
boxes is an arrow showing the translation between the 2 master frames. @*
When rotation is enabled a single box the size of the rotation block is drawn
rotated by the amount of rotation detected.

@item @b{Track single frame} @*
When this option is used the motion between a single starting frame and the
frame currently under the insertion point is calculated.  The starting frame is
specified in the @b{Frame number} blank.  The motion calculated this way is
taken as the absolute motion vector.  The absolute motion vector for each frame
replaces the absolute motion vector for the previous frame.  Settling speed has
no effect on it since it does not contain any previous motion vectors.
Playback can start anywhere on the timeline since there is no dependence on
previous results.

@item @b{Track previous frame} @*
Causes only the motion between the previous frame and the current frame to be
calculated.  This is added to an absolute motion vector to get the new motion
from the start of the sequence to the current position.  After every frame
processed this way, the block position is shifted to always cover the same
region of the image.  Playback must be started from the start of the motion
effect in order to accumulate all the necessary motion vectors.

@item @b{Previous frame same block} @*
This is useful for stabilizing jerky camcorder footage.  In this mode the
motion between the previous frame and the current frame is calculated.  Instead
of adjusting the block position to reflect the new location of the image, like
Track Previous Frame does, the block position is unchanged between each frame.
Thus a new region is compared for each frame.

@item @b{Master layer} @*
This determines the track which supplies the starting frame and ending frame
for the motion calculation.  If it is @b{Bottom} the bottom track of all the
tracks sharing this effect is the master layer.  The top track of all the
tracks is the target layer.

@item @b{Calculation} @*
This determines whether to calculate the motion at all and whether to save it
to disk.  If it is @b{Don't Calculate} the motion calculation is skipped.  If
it is @b{Recalculate} the motion calculation is performed every time each frame
is rendered.  If it is @b{Save} the motion calculation is always performed but
a copy is also saved.  If it is @b{Load}, the motion calculation is loaded from
a previous save calculation.  If there is no previous save calculation on disk,
a new motion calculation is performed.

@item @b{Action} @*
Once the motion vector is known this determines whether to move the target
layer opposing the motion vector or following the motion vector.  If it is
@b{Do Nothing} the target layer is untouched.  If it is @b{Track...} the target
layer is moved by the same amount as the master layer.  This is useful for
matching titles to objects in the frame.  If it is @b{Stabilize...} the target
layer is moved opposite to the motion vector.  This is useful for stabilizing
an object in the frame.  The motion operations can be accurate to single pixels
or subpixels by changing the action setting.
@end itemize

@menu
* Secrets of motion tracking::
* 2 pass motion tracking::
* Using blur to improve motion tracking::
* Using histogram to improve motion tracking::
* Motion tracking in action::
* Tracking stabilization in action::
@end menu

@c cincvdoc_node_number_195
@node Secrets of motion tracking
@subsubsection Secrets of motion tracking
@cindex Secrets of motion tracking
@cindex Motion tracking, secrets of

Since it is a very slow effect, there is a method to applying the motion
tracker to get the most out of it.  First disable playback for the track to do
motion tracking on.  Then drop the effect on a region of video with some motion
to track.  Then rewind the insertion point to the start of the region.  Set
@b{Action} -> @b{Do Nothing}.  Set @b{Calculation} -> @b{Don't calculate}.
Enable @b{Draw vectors}.  Then enable playback of the track to see the motion
tracking areas.

Enable which of @b{translation motion} or @b{rotation motion} vectors you want
to track.  By watching the compositor window and adjusting the @b{Block x,y}
settings, center the block on the part of the image you want to track.  Then
set search radius, block size, and block coordinates for translation and
rotation.

Once this is configured, set the calculation to @b{Save coords} and do test
runs through the sequence to see if the motion tracker works and to save the
motion vectors.  Once this is done, disable playback for the track, disable
@b{Draw vectors}, set the motion action to perform on the target layer and
change the calculation to @b{Load coords}.  Finally enable playback for the
track.

When using a single starting frame to calculate the motion of a sequence, the
starting frame should be a single frame with the least motion to any of the
other frames.  This is rarely frame 0.  Usually it is a frame near the middle
of the sequence.  This way the search radius need only reach halfway to the
full extent of the motion in the sequence.

If the motion tracker is used on a render farm, @b{Save coords} and @b{previous
frame} mode will not work.  The results of the save coords operation are saved to
the hard drives on the render nodes, not the master node.  Future rendering
operations on these nodes will process different frames and read the wrong
coordinates from the node filesystems.  The fact that render nodes only
visualize a portion of the timeline also prevents @b{previous frame} from
working since it depends on calculating an absolute motion vector starting on
frame 0.

@c cincvdoc_node_number_196
@node 2 pass motion tracking
@subsubsection 2 pass motion tracking
@cindex 2 pass motion tracking
@cindex Motion tracking, 2 pass

The method described above is 2 pass motion tracking.  One pass is used just to
calculate the motion vectors.  A second pass is used to apply the motion
vectors to the footage.  This is faster than a single pass because errors in
the motion vector calculation can be discovered quickly.

This also allows the motion tracking to use a less demanding colormodel like
RGB888 in the scanning step and a more demanding colormodel like RGB Float in
the action step.  The scanning step takes much longer than action.

This suffers the disadvantage of not being practical for extremely long
sequences where some error is acceptable and the picture quality is lousy to
begin with, like stabilizing camcorder footage.

The slower method is to calculate the motion vectors and apply them
simultaneously.  This method can use one track as the motion vector calculation
track and another track as the target track for motion vector actions.  This is
useful for long sequences where some error is acceptable.

@c cincvdoc_node_number_197
@node Using blur to improve motion tracking
@subsubsection Using blur to improve motion tracking
@cindex Blur, motion tracking
@cindex Motion tracking, improving using blur

With extremely noisy or interlaced footage, applying a blur effect before the
motion tracking can improve accuracy.  Either save the motion vectors in a
@b{tracking pass} and disable the blur for the @b{action pass} or apply the
blur just to the @b{master layer}.

@c cincvdoc_node_number_198
@node Using histogram to improve motion tracking
@subsubsection Using histogram to improve motion tracking
@cindex Motion tracking, using histogram

A histogram is almost always applied before motion tracking to clamp out noise
in the darker pixels.  Either save the motion vectors in a @b{tracking pass}
and disable the histogram for the @b{action pass} or apply the histogram just
to the @b{master layer}.

@c cincvdoc_node_number_199
@node Motion tracking in action
@subsubsection Motion tracking in action
@cindex Motion tracking in action

First, add a motion effect to the track.  Drag it from the resource window and
drop it directly over the video in Cinelerra's main window.  You should see
something similar to this:

@center @image{manual_images_en/cin_timeline_eff_clip,90mm}

Then right-click on the motion effect marker in the timeline and select show
to see the motion tracker dialog:

@center @image{manual_images_en/cin_motion_win,90mm}

Start by looking at your Compositor.  You will see some new boxes
overlaid on the video.  These are important to control the motion tracker.
Here is a quick shot of what it will look like when working:

@center @image{manual_images_en/cin_comp_action_small,90mm}

The image above shows the motion tracker losing track of the object because of
a search window that is too small.  We will talk about this more later, but
quickly: @*
@itemize @bullet
@item The middle small box is the target of the tracker.
@item The middle larger box is the search range for the tracker.  It should
contain the full range of motion for the tracking target.
@item In this example, we are trying to track the hanging handle.  We have
failed in this video frame, because the handle is far right of the center of
frame.
@item The left pointing vector indicates the motion tracker attempting to find
the target.  More on this later.
@end itemize

Move to the beginning of your video clip

Make sure the motion tracker dialog is open

Look at the Compositor

Start adjusting these four knobs:

@center @image{manual_images_en/cin_motion_track,90mm}

Make sure you check Track Translation

Uncheck Track Rotation

Start with knob two - Translation block size - and spin it to get an idea
for what is changing.  Notice that both boxes resize.  Look at the small inside
box.  Adjust it to the size of the target (the object you want to track).  Do not
worry if it does not cover the object yet.

Go on to knobs three and four - Block X and Block Y@.  Use these to put the
target designator over the target itself.

Finally, use the top knob - Translation search radius.  Expand it to
include the full range of travel you expect for the target.  If you look back
at my original action shot, the search radius was to small and the target moved
outside the range.  You can test this by playing the timeline and viewing the
results (if your machine is fast enough for realtime) or by rendering and
viewing the stabilized handle in the output.

Make the first video frame look similar to:

@center @image{manual_images_en/cin_comp_first_setup_small,90mm}

This image shows a lot of detail.  Notice that the small frame is centered over
the handle and sized to just include it.  Those settings are control by knobs
two through four.  Finally, the outer frame is larger than the back and forth
movement of the handle in the entire video clip.

Finally, here are other settings needed to see the effect:

@center @image{manual_images_en/cin_motion_set_output1,90mm}

@itemize @bullet
@item @b{Draw vectors} Uncheck this to prevent rendering of the target boxes and
motion vectors in your rendered video.  If checked, the vectors and boxes are
rendered in output video.
@item @b{Track Single Frame} For this example it is set with a Frame number of 0
(first frame)
@item @b{Master Layer} If the effect is shared between two tracks it specifies
which of those tracks will be the one where motion is tracked (Master Layer)
and which layer will be affected by the resulting translation vectors (Target
Layer).  If there is no second track sharing motion tracker then the
Master=Target.
@item @b{Action} Select the stabilize options to have the rendered video follow
the motion of the target.  Select a Track option to run motion tracking without
adjusting the video.
@item @b{Calculation}
@itemize @bullet
@item @b{Don't Calculate} select this option to turn off adjustment of video.
@item @b{Recalculate} Perform motion tracking and update video per Action
setting.
@item @b{Save and Load} Saves/Loads the translation/rotation vectors (absolute
or relative) to/from files.  Each frame gets an separate file in /tmp directory
which contains its vector.
@end itemize
@end itemize

@c cincvdoc_node_number_200
@node Tracking stabilization in action
@subsubsection Tracking stabilization in action
@cindex Tracking stabilization in action

In this section, we will explain how to stabilize a video.  Such a need can
arise when the video was taken from a vehicle for example.

First select on the timeline the part of the footage you want to stabilize,
using the in and out points.  Then apply the motion effect on that part of the
video.

Select the "Previous frame same block" option.  That option is recommended for
stabilizing jerky camcorder footage.  Its goal is not to "follow" an object.
The block stays exactly at the same place during all the effect length.

Enlarge the block and select almost half the size of the video.  Select the
"Stabilize subpixel" option: it will give a finer stabilization.  Reduce the
"Maximum absolute offset" value to limit the stabilization amplitude.  You
probably prefer to get a non-perfect stabilization on some places on the video,
than having a very large black border on one side of the picture during big
shakes.  Set the "Translation search steps" value to 128.  Increasing that
value will not give a better result, but will considerably increase the
rendering time.  Make sure the "Draw vectors" option is selected, and render
the part of the video where the motion effect is applied.

If the result is good, deselect the "Draw vectors" option.  The block and
vectors were not drawn anymore on the video.  Then, render your video to a
@file{.dv} file, and import it into your project.

You will notice the video is stabilized but there are black borders which
appear on sides of the frame.  You have to zoom in and define projector
keyframes to move the projector around the screen, in order to remove those
black borders.  The more your footage is jerky, the more you have to zoom in to
discard the black borders.  That is why the result is better with HDV footage
than with DV footage.

@c cincvdoc_node_number_201
@node Motion blur
@subsection Motion blur
@cindex Motion blur

@image{manual_images_en/motionblur,12.5mm}

FIXME

@c cincvdoc_node_number_202
@node Oil painting
@subsection Oil painting
@cindex Oil painting
@cindex Painting, oil

@image{manual_images_en/oilpainting,12.5mm}

This effect makes video tracks appears as a painting.  It can be controlled by
Radius slider.  Intensity of colors can be chosen as option.

@c cincvdoc_node_number_203
@node Overlay video
@subsection Overlay video
@cindex Overlay video

@image{manual_images_en/overlay,13.25mm}

FIXME

@c cincvdoc_node_number_204
@node Perspective
@subsection Perspective
@cindex Perspective

@image{manual_images_en/perspective,12.5mm}

The perspective effect allows you to change the perspective of an object, and
is perfect for making objects appear as if they are fading into the distance.

@c cincvdoc_node_number_205
@node Polar
@subsection Polar
@cindex Polar

@image{manual_images_en/polar,12.5mm}

The Polar effect bends and warps your video in weird ways.  Mathematically, it
converts your video from either polar coordinates to rectangular coordinates,
or the reverse.

@c cincvdoc_node_number_206
@node RGB-601
@subsection RGB-601
@cindex RGB-601

FIXME

@image{manual_images_en/rgb601,12.5mm}

@c cincvdoc_node_number_207
@node Radial blur
@subsection Radial blur
@cindex Radial blur

@image{manual_images_en/radialblur,12.5mm}

It creates a whirlpool blur that simulates a swirling camera.  You can vary the
location, type, and quality of the blur.

@c cincvdoc_node_number_208
@node ReframeRT
@subsection ReframeRT
@cindex ReframeRT video effect

@image{manual_images_en/reframert,12.5mm}

ReframeRT changes number of frames in a sequence of video directly from the
timeline.  It has 2 modes, selected by the 2 toggles in the GUI@.

@b{Stretch} mode multiplies the current frame number of its output by the scale
factor to arrive at the frame to read from its input.  If its current output
frame is #55 and the scale factor is 2, frame #110 is read from its input.  The
stretch mode has the effect of changing the length of output video by the
inverse of the scale factor.  If the scale factor is greater than 1, the output
will end before the end of the sequence on the timeline.  If it is less than 1,
the output will end after the end of the sequence on the timeline.  The
ReframeRT effect must be lengthened to the necessary length to accommodate the
scale factor.  Change the length of the effect by clicking on the endpoint of
the effect and dragging.

Although stretch mode changes the number of the frame read from its input, it
does not change the frame rate of the input.  Effects before ReframeRT assume
the same frame rate as ReframeRT@.

@cindex Fast play effect
The ReframeRT in stretch mode can be use to create a @b{fast play effect}.
Select Stretch mode and enter a value greater than 1 to get accelerated
playback.

@cindex Slow motion effect
For @b{slow motion effect}, use a ReframeRT effect in stretch mode with a value
less than 1.  @b{Example}: you have a clip that you want to put in slow motion.
The clip starts at 33.792 seconds and ends at 39.765.  The clip is 5.973
seconds long.  You want to play it at 4/10ths normal speed.  You divide the
clip length by the playback speed (5.973/.4) to get a final clip length of
14.9325 seconds.  You create an in point at the start of your clip: 33.792
seconds.  You put an out point 14.9325 seconds later, at 48.7245 seconds
(33.792 + 14.9325).  You attach a ReframeRT effect, set it to .4 and stretch.
You change the out point at 48.7245 to an in point.  You start your next clip
after the slow motion effect at the 48.7245 out point.

You can also change the frame rate of the clip if you right click on it in the
@b{media viewer} and go to @b{Info}.  If you do not hit the drop down first, you
can type in a desired frame rate as well.  Cinelerra will pick the right frames
out for the project frame rate, effectively doing the time-lapsing as well

@b{Downsample} mode does not change the length of the output sequence.  It
multiplies the frame rate of the output by the scale factor to arrive at a
frame rate to read the input.  This has the effect of replicating the input
frames so that they only change at the scaled frame rate when sent to the
output.  It does not change the length of the sequence.  If the scale factor is
0.5 and the output frame rate is 30 fps, only 15 frames will be shown per
second and the input will be read at 15 fps.  Downsample is only useful for
scalefactors below 1, hence the name downsample.

Downsample mode changes the frame rate of the input as well as the number of
the frame to read, so effects before ReframeRT see the frame rate * the scale
factor as their frame rate.  If the scale factor is 2 and the output frame rate
is 30, the input frame rate will be 60 and the input frame number will by
doubled.  This will not normally do anything but some input effects may behave
differently at the higher frame rate.

@c cincvdoc_node_number_209
@node Reverse video
@subsection Reverse video
@cindex Reverse video effect

@image{manual_images_en/reversevideo,12.5mm}

Media can be reversed on the timeline in realtime.  This is not to be confused
with using the reverse playback on the transport.  The reverse effects reverse
the region covered by the effect regardless of the transport direction.

The region to be reversed is first determined by what part of the track the
effect is under and second by the locations of keyframes in the effect.  The
reverse effects have an @b{enabled} option which allows you to set keyframes.
This allows may possibilities.

Every @b{enabled} keyframe is treated as the start of a new reversed region and
the end of a previous reversed region.  Several @b{enabled} keyframes in
succession yield several regions reversed independent of each other.  An
@b{enabled} keyframe followed by a @b{disabled} keyframe yields one reversed
region followed by a forward region.

@c cincvdoc_node_number_210
@node Rotate
@subsection Rotate
@cindex Rotate

@image{manual_images_en/rotate,13.25mm}

The Rotate filter can rotate the video in 90 degree increments, reverse and
flip the video.

@c cincvdoc_node_number_211
@node SVG via Inkscape
@subsection SVG via Inkscape
@cindex SVG via Inkscape

@image{manual_images_en/svg,12.5mm}

FIXME

@c cincvdoc_node_number_212
@node Scale
@subsection Scale
@cindex Scale

@image{manual_images_en/scale,12.5mm}

FIXME

@c cincvdoc_node_number_213
@node Selective temporal averaging
@subsection Selective temporal averaging
@cindex Selective temporal averaging

@image{manual_images_en/timeavg,12.5mm}

This plugin is designed to smooth out non-moving areas of a video clip.  The
smoothing is performed by averaging the color component for each pixel across a
number of frames.  The smoothed value is used if both the standard deviation
and the difference between the current component value and the average
component value are below a threshold.

The average and standard deviation are calculated for each of the components
of the video.  The type of components averaged is determined by the color model
of the entire project.  The average and standard deviation of the frames can be
examined by selecting the specific radio button in the plugin options window.

The region over which the frames are averaged is determined by either a fixed
offset or a restart marker system.  In a restart marker system, certain
keyframes are marked as beginning of sections.  Then for each section, the
frames surrounding the current frame are used as the frames to average over,
except when approaching the beginning and end of a section, whereby the
averaging is performed over the @i{N} beginning or ending frames respectively.

@b{Common usage:}

If you have to select the number of frames you wish to average.

@enumerate 1
@item Enter a reasonable number of frames to average (e.g. 10).
@item Select the @b{Selective Temporal Averaging} method and enter 1 and 10 for
all the @b{Av. Thres.} and @b{S.D. Thres.} respectively.  This basically causes
all pixels to use the average value.
@item Turn the mask for a the first component on.  This should make the whole
frame have a solid color of that specific component.
@item Slowly reduce the @b{S.D. Thres.} value.  As you do so, you will notice
that the regions vastly different from the average will have a flipped mask
state.  Continue to reduce the threshold until you reach the point at which
non-moving regions of the video have a flipped masked state.  This value is
known as the @b{noise-floor} and is the level of natural noise generated by the
CCD in the camera.
@item Repeat the same procedure for the @b{Av. Thres.}
@item Turn off the mask
@item Repeat this for all channels
@end enumerate

@c cincvdoc_node_number_214
@node Sharpen
@subsection Sharpen
@cindex Sharpen

@image{manual_images_en/sharpen,12.5mm}

FIXME

@c cincvdoc_node_number_215
@node ShiftInterlace
@subsection ShiftInterlace
@cindex ShiftInterlace

@image{manual_images_en/shiftinterlace,12.5mm}

FIXME

@c cincvdoc_node_number_216
@node Swap channels
@subsection Swap channels
@cindex Swap channels

@image{manual_images_en/swapchannels,12.5mm}

FIXME

@c cincvdoc_node_number_217
@node Threshold
@subsection Threshold
@cindex Threshold video effect
@cindex Luminance

@image{manual_images_en/threshold,12.5mm}

Threshold converts the image to pure luminance.  Then luminance values below
and above the threshold range are converted to black and luminance values
inside the threshold range are converted to white.  The threshold window shows
a histogram of luminance values for the current frame.  Click dragging inside
the histogram creates a range to convert to white.  @b{SHIFT-clicking} extends
one border of this range.  Values for the threshold range can also be specified
in the text boxes.

This effect is basically a primitive luminance key.  A second track above the
track with the threshold effect can be multiplied, resulting in only the parts
of the second track within the threshold being displayed.

@c cincvdoc_node_number_218
@node Time average
@subsection Time average
@cindex Time average video effect

@image{manual_images_en/timeavg,12.5mm}

Time average is one effect which has many uses besides creating nifty trail
patterns of moving objects.  It is main use is reducing noise in still images.
Merely point a video camera at a stationary subject for 30 frames, capture the
frames, and average them using @b{time average} and you will have a super high
quality print.  In floating point colormodels, time average can increase the
dynamic range of lousy cameras.

Inside the time average effect is an accumulation buffer and a divisor.  A
number of frames are accumulated in the accumulation buffer and divided by the
divisor to get the average.

Because the time average can consume enormous amounts of memory, it is best
applied by first disabling playback for the track, dropping the time average in
it, configuring time average for the desired number of frames, and re-enabling
playback for the track.

@itemize @bullet
@item @b{Frames to average} @*
This determines the number of frames to be accumulated in the accumulation
buffer.  For extremely large integrations it is easier to edit the EDL in a
text editor and put in the number of frames.

@item @b{Accumulate} @*
This outputs the accumulation buffer without dividing it.

@item @b{Average} @*
This causes the accumulation buffer to be divided before being output.
This results in the average of all the frames.

@item @b{Inclusive Or} @*
This causes the accumulation buffer to be replaced by any pixels which
are not transparent.  In combination with motion tracking it allows entire
sequences to be combined to form panoramas.

@item @b{Reprocess frame again} @*
If an effect before the time average is adjusted the time average normally
does not reread the accumulation buffer to get the change.  This forces it to
reread the accumulation buffer when other effects change.

@item @b{Disable subtraction} @*
In order to represent the accumulation of only the specified number of
frames, the time average retains all the previous frames in memory and
subtracts them out as it plays forward.  It would run out of memory if it had
to accumulate thousands of frames.  By disabling subtraction the previous
frames are not stored in memory and only the average function is affected by
the frame count.
@end itemize

@c cincvdoc_node_number_219
@node TimeFront
@subsection TimeFront
@cindex TimeFront

@image{manual_images_en/timefront,12.5mm}

This is a warping framework plugin based on this article: @*
@uref{http://www.vision.huji.ac.il/videowarping/HUJI-CSE-LTR-2005-10_etf-tr.pdf}

@c cincvdoc_node_number_220
@node Title
@subsection Title
@cindex Title
@cindex Gimp

@image{manual_images_en/titler,12.5mm}

While it is possible to add text to movies by importing still images from The
Gimp and compositing them, the Titler allows you to add text from within
Cinelerra.

The titler has standard options for @b{font, size, and style}.  The best font
is a generic, normal font like Arial in a large size.

The titler also has options you will only find in moving pictures.  The
@b{Justify} operation justifies the text relative to the entire frame.  Once
justified, the @b{X and Y} offset is applied.  This allows text to be justified
while at the same time letting you push it within the title safe region.

The @b{motion type} scrolls the text in any of the four directions.  When using
this, the text may disappear.  Move the insertion point along the timeline
until the text is far enough along the animation to reappear.  The text scrolls
on and scrolls off.

Setting @b{loop} causes the text to scroll completely off and repeat.  Without
@b{loop} the text scrolls off and never reappears.

The speed of the animation is determined by @b{speed}.  Set it higher to speed
up the animation.

@b{Drop shadow} draws a black copy of the text to the bottom right of the
original text.  This is useful when drawing text over changing video to keep
the border always visible.

In addition to the scrolling, @b{Fade in/Fade out} are a second type of
animation.  If the fade seconds are 0, no fading is done.

@b{Color} picks the color to draw the text in.  Usually white is the only
practical color.

@b{Stamp timecode} replaces the text with the current position on the timeline
in seconds and frames.

@menu
* Adding fonts to the titler:: How to add fonts to the titler
* The title-safe region::      How to keep text visible on output
@end menu

@c cincvdoc_node_number_221
@node Adding fonts to the titler
@subsubsection Adding fonts to the titler

@cindex Fonts, adding to the titler
@cindex TTF fonts
The X Window system originally did not have a suitable font renderer for video.
It also is restricted to the current bit depth.  It does not have a convenient
way to know which fonts work with the suitable font renderer in the desired bit
depth.  The easiest way we have found to support fonts in the titler is to have a
directory for them at @file{/usr/lib/cinelerra/fonts}.

The titler supports mainly @b{TTF}, true type fonts.  It supports others but
TTF are the most reliable.  To add true type fonts, copy the @file{.TTF} files
to the @file{/usr/lib/cinelerra/fonts} directory.  In that directory run
@command{ttmkfdir && mv fonts.scale fonts.dir} and restart Cinelerra.  The new
fonts should appear.  The usage of ttmkfdir changes frequently so this
technique might not work.

@c cincvdoc_node_number_222
@node The title-safe region
@subsubsection The title-safe region
@cindex Title-safe region
@cindex TV display

If the video is displayed on a consumer TV, the outer border is going to be
cropped by 5% on each side.  Moreover, text which is too close to the edge
looks sloppy.  Make sure when adding titles to have the @b{title-safe}
@image{manual_images_en/titlesafe} tool active in the @b{compositor} window.  The text
should not cross the inner rectangle.

@c cincvdoc_node_number_223
@node Translate
@subsection Translate
@cindex Translate

@image{manual_images_en/translate,12.5mm}

FIXME

@c cincvdoc_node_number_224
@node Unsharp
@subsection Unsharp
@cindex Unsharp

@image{manual_images_en/unsharp,12.5mm}

This effect unsharps the video.  Its parameters are:
@itemize @bullet
@item @b{Amount} @*
Moving the slider to the right makes dark areas get darker and light areas
get lighter.

@item @b{Radius} @*
This slider controls how much blurring is used in the edge-finding stage.
The practical effect of this is to specify how large a region is darkened or
lightened.

@item @b{Threshold} @*
This slider permits to control how big is a difference between a pixel in
the blurred copy and the original copy is needed before any darkening or
lightening will be applied.
@end itemize

@c cincvdoc_node_number_225
@node Videoscope
@subsection Videoscope
@cindex Videoscope

@menu
* The waveform scope::
* The vectorscope::
@end menu

@image{manual_images_en/videoscope,12.5mm}

The videoscope is a tool that digitally represents levels of light and color on
a calibrated screen.  It is useful because the human eye is not specialized to
match precise level of light and color, but to detect differences between
lights and colors.

The Videoscope can be used in conjunction with other Cinelerra plugins such as
YUV, HUE, Brightness, Histogram to accurately correct video for contrast,
clarity, conformance (normalize various videos shot under different light
settings) or for cinematic purposes.

Some thought is being given to having a video scope for recording.
Unfortunately, this would require a lot of variations of the video scope for
all the different video drivers.

The Videoscope contains two displays: the @b{waveforme scope} and the
@b{vectorscope}

@c cincvdoc_node_number_226
@node The waveform scope
@subsubsection The waveform scope
@cindex Waveform scope

Screen labeled vertically from bottom to top bottom or 0% is maximum Black Top
or 100% is maximum White.  Divides the source video on vertical columns of
pixels, then each pixel on the column is metered, and plotted on corresponding
vertical scale the waveform scope according to its light intensity value.

A pixel on the bottom of the scope indicates for full black, a pixel on the top
(100%) of the scope is full white.

The image shows a stair step set of lines on scope, indicating the matching
levels of luminance on the test image bars.  Multiple levels on the same column
are represented by multiple lines on the scope.

The Waveform scope helps correct image light levels for contrast range or for
conforming light levels on various scenes originally shot on different light
settings.

Adjusting light levels.  (adjusting luminance) *Insert the xxx videoeffect on
your track, right click, Show Insert the videoscope effect on the track.  (make
sure that it is placed below the xx effect, so it can see the result of the xxx
effect.  If it is not, rightclick and move it down to make it so) rightclick,
show.

Looking at the luminance levels shown on the waveform, adjust the xxx control
to match the desired level of light for your image.

If you are looking for best contrast range, adjust the xxx level on the plugin
to align the darkest point on the scope on the 0% scale and the whitest portion
of interest on the 100%.  Anything above 100% is over saturated.

@c cincvdoc_node_number_227
@node The vectorscope
@subsubsection The vectorscope
@cindex Vectorscope

The Vectorscope is used to monitor color.  The screen represents a color wheel
where pixel color value is plotted on the radius of a line moving away from the
center, the smallest radius indicates full white and the outer circles
indicating highest values of intensity.

Color hue is also plotted, represented by the angle in degrees on the color
wheel, representing different hue colors.

The Vectorscope can used along other plugins to correct color, adjust image
tint, and apply other effects for cinematic effects, image correction or to
conform separate screens to look the same.

The vectorscope can also be used to monitor that the video output can be
properly displayed on various monitors.  Any points along the inner radius will
be displayed as pure white and any points above the 100% radius, will probably
not be correctly displayed on the screen.

@c cincvdoc_node_number_228
@node Wave
@subsection Wave
@cindex Wave

@image{manual_images_en/wave,12.5mm}

The wave effect adds waves on the image.

@center @image{manual_images_en/effect_wave_before_after,120mm}

You can adjust the following parameters:

@center @image{manual_images_en/effect_wave_window,60mm}

@c cincvdoc_node_number_229
@node Whirl
@subsection Whirl
@cindex Whirl

@image{manual_images_en/whirl,12.5mm}

FIXME

@c cincvdoc_node_number_230
@node YUV
@subsection YUV
@cindex YUV

@image{manual_images_en/yuv,12.5mm}

FIXME

@c cincvdoc_node_number_231
@node Zoom blur
@subsection Zoom blur
@cindex Zoom blur

@image{manual_images_en/zoomblur,12.5mm}

FIXME

@c cincvdoc_node_number_232
@node Rendered effects
@chapter Rendered effects
@cindex Rendered effects
@cindex Effects, rendered
Another type of effect is performed on a section of the track and the result
stored somewhere before it is played back.  The result is usually pasted into
the track to replace the original data.

The rendered effects are not listed in the resource window but instead are
accessed through the @b{Audio->Render effect} and @b{Video->Render effect} menu
options.  Each of these menu options brings up a dialog for the rendered
effect.  Rendered effects apply to only one type of track, either audio or
video.  If no tracks of the type exist, an error pops up.

A region of the timeline to apply the effect to must be defined before
selecting @b{Render effect...}.  If no in/out points and no highlighted region
exist, the entire region after the insertion point is treated as the affected
region.  Otherwise, the region between the in/out points or the highlighted
region is the affected region.

Secondly, the tracks to apply the rendered affect to need to be @b{armed}.  All
other tracks are ignored.

Finally, the rendered affect processes certain track attributes when it reads
its input data but not others.  Transitions in the affected track are applied.
Nudge is not and effects are not.  This allows the new data to be pasted into
the existing position without changing the nudge value.

In the render effect dialog is a list of all the realtime and all the rendered
effects.  The difference here is that the realtime effects are rendered to disk
and not applied under the track.  Highlight an effect in the list to designate
it as the one being performed.

Define a file to render the effect to in the @b{Select a file to render to}
box.  The @image{manual_images_en/magnify,7mm} magnifying glass allows file selection
from a list.

Select a file format which can handle the track type.  The
@image{manual_images_en/wrench,4.33mm} wrench allows configuration specific to the
file format.

There is also an option for creating a new file at each label.  If you have a
CD rip on the timeline which you want to divide into different files, the
labels would become dividing points between the files if this option were
selected.  When the timeline is divided by labels, the effect is re-initialized
at every label.  Normalize operations take the peak in the current file and not
in the entire timeline.

Finally there is an insertion strategy just like in the render dialog.  It
should be noted that even though the effect applies only to audio or video, the
insertion strategy applies to all tracks just like a clipboard operation.

When you click @b{OK} in the effect dialog, it calls the GUI of the effect.  If
the effect is also a realtime effect, a second GUI appears to prompt for
acceptance or rejection of the current settings.  After accepting the settings,
the effect is processed.

@menu
* Rendered audio effects::     Rendered audio effects
* Rendered video effects::     Rendered video effects
@end menu

@c cincvdoc_node_number_233
@node Rendered audio effects
@section Rendered audio effects
@cindex Rendered audio effects

@menu
* Resample::        How to reduce the dynamic range of audio.
@end menu

@c cincvdoc_node_number_234
@node Resample
@subsection Resample
@cindex Resample effect

This multiplies the number of each output sample by a scale factor to arrive at
the number of the input sample.  The output file's sample rate is set to the
project sample rate but its length is changed to reflect the scaled number of
samples.  It also filters the resampled audio to remove aliasing.

If the scale factor is 2, every 2 input samples will be reduced to 1 output
sample and the output file will have half as many samples as the input
sequence.  If it is 0.5, every 0.5 input samples will be stretched to 1 output
sample and the output file will have twice as many samples as the input
sequence.

@c cincvdoc_node_number_235
@node Rendered video effects
@section Rendered video effects
@cindex Rendered video effects

@menu
* Reframe::        Reframe
@end menu

@c cincvdoc_node_number_236
@node Reframe
@subsection Reframe
@cindex Reframe video effect

This does exactly the same thing as @b{ReframeRT} in @b{Stretch} mode.  It
multiplies the output frame number by the scale factor to arrive at the input
frame number and changes the length of the sequence.  Unlike ReframeRT, this
must run from the @b{Video} menu and render its output.

Be aware @b{Reframe} does not write the scaled frame rate as the frame rate of
the rendered file.  It produces a file of scaled length and equal frame rate as
the project.  The new length is 1/scale factor as big as the original sequence.

@b{To create a slow-motion of fast moving video:}
@enumerate 1
@item Select the video clip you wish to re-frame and put it on a
video track
@item Select the area you wish to reframe
@item From the Video menu, select the Render Effect option
@item From the effect list, select Reframe
@item Enter the output format and insertion strategy for the new
clip to be created
@item Press ok
@item At the popup menu, enter the scale factor 2 to run twice as
fast, and .5 to run at half speed
@end enumerate

@c cincvdoc_node_number_237
@node Ladspa effects
@chapter Ladspa effects
@cindex Ladspa effects

LADSPA effects are supported in realtime and rendered mode for audio.  The
LADSPA plugins you get from the internet vary in quality.  Most can not be
tweeked in realtime very easily and work better when rendered.  Some crash and
some can only be applied to one track due to a lack of re-entrancy.  Although
Cinelerra implements the LADSPA interface as accurately as possible, multiple
tracks of realtime, simultaneous processing go beyond the majority of LADSPA
users.  LADSPA effects appear in the audio folder as the hammer and
screwdriver, to signify that they are Plugins for GNU/Linux Audio Developers.

LADSPA Effects are enabled merely by setting the @env{LADSPA_PATH} environment
variable to the location of your LADSPA plugins or putting them in the
@file{/usr/lib/cinelerra} directory.

If you use Debian, you can get a lot of plugins using apt: @*
@command{apt-cache search ladspa} @*
@command{apt-get install jack-rack cmt blop swh-plugins}

@c cincvdoc_node_number_238
@node Transitions
@chapter Transitions
@cindex Transitions

@menu
* Using transitions::
* Dissolve video transition::
@end menu

@c cincvdoc_node_number_239
@node Using transitions
@section Using transitions
@cindex Using transitions

When one edit ends and another edit begins, the default behavior is to have
the first edit's output immediately become the output of the second edit when
played back.  Transitions are a way for the first edit is output to become the
second edit is output with different variations.

Cinelerra supports audio and video transitions, all of which are listed in the
resource window.

@center @image{manual_images_en/resources_video_transitions,70mm}
@center @b{Video transitions in the resources window}

Transitions may only apply to the matching track type.  Transitions under
@b{audio transitions} can only apply to audio tracks.  Transitions under
@b{video transitions} can only apply to video tracks.

Load a video file and cut a section from the center so the edit point is
visible on the timeline.  Go the resource window and click on the @b{Video
transitions} folder.  Drag a transition from the transition list onto the
second video edit on the timeline.  A box highlights over where the transition
will appear.  Releasing it over the second edit applies the transition between
the first and second edit.

@center @image{manual_images_en/drop_transition}
@center @b{Dragging a dissolve transition to the timeline}

You can now scrub@footnote{Scrubbing refers to moving backwards and forwards in
time across a video or audio clip.} over the transition with the transport
controls and watch the output in the @b{Compositor window}.  Scrubbing with the
insertion point does not normally show transitions because the transition
durations are usually too short.

@cindex Detach Transitions
@cindex Edit Transitions
Once the transition is in place, it can be edited similarly to an effect.  Move
the pointer over the transition and right click to bring up the transition
menu.  The @b{show} option brings up specific parameters for the transition in
question if there are any.  The @b{length} option adjusts the length of the
transition in seconds.  Once these two parameters are set, they are applied to
future transitions until they are changed again.  Finally, the @b{detach}
option removes the transition from the timeline.

Dragging and dropping transitions from the Resource window to the Program
window can be really slow and tiring.  Fortunately, once you drag a transition
from the Resource window, the @b{U} and @b{u} keys will paste the same
transition.  The @b{U} key pastes the last video transition and the @b{u} key
pastes the last audio transition on all the recordable tracks.  If the
insertion point or in point is over an edit, the beginning of the edit is
covered by the transition.

It should be noted that when playing transitions from the timeline to hardware
accelerated video device, the hardware acceleration will usually be turned off
momentarily during the transition and on after the transition in order to
render the transition.  Using an un-accelerated video device for the entire
timeline normally removes the disturbance.

@b{Important:} The exact point in time when the transition takes effect is not
straightforward.  It starts when the second edit begins and lasts a certain
amount of time into the second edit.  Therefore, the first asset needs to have
enough data after the edit point to fill the transition into the second edit.

For example, the dissolve transition starts at the exact position where it is
located on the timeline.  If you set a duration of 1 second for a dissolve
transition, it @b{will not} start 0.5 second before the transition and continue
0.5 second after that point.  In fact, it will start exactly where the
transition is located on the timeline, and last for 1 second from that
position.

It is a common error to put a dissolve transition just after the last frame of
an asset.  Let's imagine a dissolve transition is put between asset A and asset
B, just after the last frame of asset A@.

Since the dissolve effect starts exactly at the position where it is located,
there is no more frames from asset A to display when the dissolve transition
starts.  Therefore, there is no other choice for Cinelerra than freezing the
last frame of asset A and dissolving it with asset B@.

You have to make sure there is enough frames of asset A to be displayed when
the dissolve transition starts.  The duration of those frames should be equal
or greater than the length of the transition effect.

@c cincvdoc_node_number_240
@node Dissolve video transition
@section Dissolve video transition
@cindex Dissolve video transition

@image{manual_images_en/video_disolve_icon}

This is a soft dissolve transition between two video segments, which we call in
and out segments.  The in segment turns increasingly transparent while the out
segment materializes into its place.  The length of time for the full effect to
take place can be controlled by the "Transition Length" control.

@b{Available controls:} @*
By right-clicking on the transition icon in the timeline, a menu will pop-up
with the following controls
@itemize @bullet
@item @b{Show:} Pop up the transition specific menu (not available on this
transition)
@item @b{On:} Toggle on/off the transition effect
@item @b{Transition length:} Set the span in seconds for the transition to
complete
@item @b{Detach:} Remove the transition from the timeline
@end itemize

@c cincvdoc_node_number_241
@node Keyframes
@chapter Keyframes
@cindex Keyframes

When you change the fade, camera, projector, or other parameters for a track,
they stay by default the same for the entire duration of the timeline.  Setting
static parameters is not very useful sometimes.  Normally you need to move the
camera around over time or change mask positions.  Masks need to follow
objects.  We create dynamic changes by defining keyframes.  A keyframe is a
certain point in time when the settings for one operation change.  In
Cinelerra, there are keyframes for almost every compositing parameter and
effect parameter.

@cindex Default keyframe
@cindex Keyframe, default
Whenever you adjust any parameter, the value is stored in a keyframe.  If the
value is stored in a keyframe, why does not it always change?  The keyframe it
is stored in by default is known as the @b{default keyframe}.  The default
keyframe applies to the entire duration if no other keyframes are present.  The
default keyframe is not drawn anywhere because it always exists.  The only way
change occurs over time is if non-default keyframes are created.

Display keyframes for any parameter by using the @b{view} menu.  A faster way
to toggle multiple keyframe types is to bring up @b{window->overlays}.  This
window allows toggling of every parameter in the view menu.  When keyframes are
selected, they are drawn on the timeline over the tracks they apply to.

Keyframes come in many forms: curves, toggles, modes, and so on.  How to handle
the different types of keyframes is described here.

@menu
* Curve keyframes::
* Toggle keyframes::
* Automatic keyframes::
* Compositor keyframes::
* Editing keyframes::
@end menu

@c cincvdoc_node_number_242
@node Curve keyframes
@section Curve keyframes
@cindex Curve keyframes
@cindex Keyframes, curve

Many parameters are stored in Bezier curves.  Go to @b{view->fade} or
@b{view->...zoom} to show curves on the timeline for those parameters.  In
either arrow editing mode or i-beam editing mode, move the cursor over the
curves in the timeline until it changes shape.  Then merely by clicking and
dragging on the curve you can create a keyframe at the position.

After the keyframe is created, click drag on it again to reposition it.  When
you click-drag a second keyframe on the curve, it creates a smooth ramp.
@b{CTRL-dragging} on a keyframe changes the value of either the input control
or the output control.  This affects the sharpness of the curve.  While the
input control and the output control can be moved horizontally as well as
vertically, the horizontal movement is purely for legibility and is not used in
the curve value.

You may remember that The Gimp and the Compositing masks all use @key{SHIFT} to
select control points so why does the timeline use @key{CTRL}?  When you
@b{SHIFT-drag} on a timeline curve, the keyframe snaps to the value of either
the next or previous keyframe, depending on which exists.  This lets you set a
constant curve value without having to copy the next or previous keyframe.

@menu
* Navigating curve keyframes::
@end menu

@c cincvdoc_node_number_243
@node Navigating curve keyframes
@subsection Navigating curve keyframes
@cindex Navigating curve keyframes

There is not much room on the timeline for a wide range of curve values.  You
need to zoom the curves in and out vertically to have any variability.  This is
done by 2 tools: the automation fit button and automation zoom menu
@image{manual_images_en/autozoom}.

The automation fit button scales and offsets the vertical range so the selected
curve area appears in the timeline.  If a region of the timeline is highlighted
by the cursor, only that region is scaled.  In/out points do not affect the
zoomed region.  @kbd{ALT-f} also performs automation fitting.

The automation zoom menu manually changes the vertical scaling of the curves in
multiples of 2.  Click on its tumbler to change the zoom.  @kbd{ALT-UP} and
@kbd{ALT-DOWN} change the automation zoom from the keyboard.

@c cincvdoc_node_number_244
@node Toggle keyframes
@section Toggle keyframes
@cindex Toggle keyframes
@cindex Keyframes, toggle

Mute is the only toggle keyframe.  Mute keyframes determine where the track is
processed but not rendered to the output.  Click-drag on these curves to create
a keyframe.  Unlike curves, the toggle keyframe has only two values: on or off.
@key{CTRL} and @key{SHIFT} do nothing on toggle keyframes.

@c cincvdoc_node_number_245
@node Automatic keyframes
@section Automatic keyframes
@cindex Automatic keyframes
@cindex Keyframes, automatic

You may have noticed when a few fade curves are set up, moving the insertion
point around the curves causes the faders to reflect the curve value under the
insertion point.  This is not just to look cool.  The faders themselves can set
keyframes in automatic keyframe mode.  Automatic keyframe mode is usually more
useful than dragging curves.

Enable automatic keyframe mode by enabling the automatic keyframe toggle
@image{manual_images_en/autokeyframe}.  In automatic keyframe mode, every time you tweek
a key-framable parameter it creates a keyframe on the timeline.  Since
automatic keyframes affect many parameters, it is best enabled just before you
need a keyframe and disabled immediately thereafter.

It is useful to go into the @b{View} menu and make the desired parameter
visible before performing a change.  The location where the automatic keyframe
is generated is under the insertion point.  If the timeline is playing back
during a tweek, several automatic keyframes will be generated as you change the
parameter.

When automatic keyframe mode is disabled, a similarly strange thing happens.
Adjusting a parameter adjusts the keyframe immediately preceding the insertion
point.  If two fade keyframes exist and the insertion point is between them,
changing the fader changes the first keyframe.

There are many parameters which can only be keyframed in automatic keyframe
mode.  These are parameters for which curves would take up too much space on
the track or which can not be represented easily by a curve.

Effects are only key-framable in automatic mode because of the number of
parameters in each individual effect.

Camera and projector translation can only be keyframed in automatic keyframe
mode while camera and projector zoom can be keyframed with curves.  It is here
that we conclude the discussion of compositing, since compositing is highly
dependant on the ability to change over time.

@c cincvdoc_node_number_246
@node Compositor keyframes
@section Compositor keyframes
@cindex Compositor keyframes
@cindex Keyframes, compositor

Camera and projector translation is represented by two parameters: x and y.
Therefore it is cumbersome to adjust with curves.  Cinelerra solves this
problem by relying on automatic keyframes.  With a video track loaded, move the
insertion point to the beginning of the track and enable automatic keyframe
mode.

Move the projector slightly in the compositor window to create a keyframe.
Then go forward several seconds.  Move the projector a long distance to create
another keyframe and emphasize motion.  This creates a second projector box in
the compositor, with a line joining the two boxes.  The joining line is the
motion path.  If you create more keyframes, more boxes are created.  Once all
the desired keyframes are created, disable automatic keyframe mode.

Now when scrubbing around with the compositor window's slider, the video
projection moves over time.  At any point between two keyframes, the motion
path is red for all time before the insertion point and green for all time
after the insertion point.  It is debatable if this is a very useful feature
but it makes you feel good to know what keyframe is going to be affected by the
next projector tweek.

Click-drag when automatic keyframes are off to adjust the preceding keyframe.
If you are halfway between two keyframes, the first projector box is adjusted
while the second one stays the same.  Furthermore, the video does not appear to
move in step with the first keyframe.  This is because halfway between two
keyframes the projector translation is interpolated.  In order to set the
second keyframe you will need to scrub after the second keyframe.

By default the motion path is a straight line, but it can be curved with
control points.  @b{CTRL-drag} to set either the in or out control point of the
preceding keyframe.  Once again, we depart from The Gimp because @key{SHIFT} is
already used for zoom.  After the in or out control points are extrapolated
from the keyframe, @b{CTRL-dragging} anywhere in the video adjusts the nearest
control point.  A control point can be out of view entirely yet still
controllable.

When editing the camera translation, the behavior of the camera boxes is
slightly different.  Camera automation is normally used for still photo
panning.  The current camera box does not move during a drag, but if multiple
keyframes are set, every camera box except the current keyframe appears to
move.  This is because the camera display shows every other camera position
relative to the current one.

The situation becomes more intuitive if you bend the motion path between two
keyframes and scrub between the two keyframes.  The division between red and
green, the current position between the keyframes, is always centered while the
camera boxes move.

@c cincvdoc_node_number_247
@node Editing keyframes
@section Editing keyframes
@cindex Editing keyframes
@cindex Keyframes, editing

@b{IMPORTANT:} when copying and pasting keyframes, make sure there is @b{no IN
or OUT point defined on the timeline}.

Keyframes can be shifted around and moved between tracks on the timeline using
similar cut and paste operations to editing media.  Only the keyframes selected
in the @b{view} menu are affected by keyframe editing operations, however.

The most popular keyframe editing operation is replication of some curve from
one track to the other, to make a stereo pair.  The first step is to solo the
source track's record @image{manual_images_en/recordpatch_up} patch by
@b{SHIFT-clicking} on it.  Then either set in/out points or highlight the
desired region of keyframes.  Go to @b{keyframes->copy keyframes} to copy them
to the clipboard.  Solo the destination track's record
@image{manual_images_en/recordpatch_up} patch by @b{SHIFT-clicking} on it and go to
@b{keyframes->paste keyframes} to paste the clipboard.

The media editing commands are mapped to the keyframe editing commands by using
the @key{SHIFT} key instead of just the keyboard shortcut.

This leads to the most complicated part of keyframe editing, the default
keyframe.  Remember that when no keyframes are set at all, there is still a
default keyframe which stores a global parameter for the entire duration.  The
default keyframe is not drawn because it always exists.  What if the default
keyframe is a good value which you want to transpose between other non-default
keyframes?  The @b{keyframes->copy default keyframe} and @b{keyframes->paste
default keyframe} allow conversion of the default keyframe to a non-default
keyframe.

@b{Keyframes->copy default keyframe} copies the default keyframe to the
clipboard, no matter what region of the timeline is selected.  The
@b{keyframes->paste keyframes} function may then be used to paste the clipboard
as a non-default keyframe.

If you have copied a non-default keyframe, it can be stored as the default
keyframe by calling @b{keyframes->paste default keyframe}.  After using paste
default keyframe to convert a non-default keyframe into a default keyframe, you
will not see the value of the default keyframe reflected until all the
non-default keyframes are removed.

Finally, there is a convenient way to delete keyframes besides selecting a
region and calling @b{keyframes->clear keyframes}.  Merely click-drag a
keyframe before its preceding keyframe or after its following keyframe on the
track.

@c cincvdoc_node_number_248
@node Capturing media
@chapter Capturing media
@cindex Capturing media
@cindex Media, capturing

@menu
* Capturing using Cinelerra::
* Capturing using dvgrab::
@end menu

@c cincvdoc_node_number_249
@node Capturing using Cinelerra
@section Capturing using Cinelerra

@menu
* Cinelerra recording functions::
* Batch recording::
* Editing tuner information::
@end menu

@c cincvdoc_node_number_250
@node Cinelerra recording functions
@subsection Cinelerra recording functions
@cindex Cinelerra recording functions

Ideally, all media would be stored on hard drives, CD-ROM, flash, or DVD and
loading it into Cinelerra would be a matter of loading a file.  In reality,
very few sources of media can be accessed like a filesystem but instead rely on
tape transport mechanisms and dumb I/O mechanisms to transfer the data to
computers.  These media types are imported into Cinelerra through the Record
dialog.

The first step in recording is to configure the input device.  In
@b{Settings->preferences} are a number of recording parameters described in
configuration @xref{Recording}.  These parameters apply to recording no matter
what the project settings are, because the recording parameters are usually the
maximum capability of the recording hardware while project settings come and
go.

Go to @b{File->record} to record a dumb I/O source.  This prompts for an output
format much like rendering does.  Once that is done, the record window and the
record monitor pop up.

The record window has discrete sections.  While many parameters change
depending on if the file has audio or video, the discrete sections are always
the same.

@itemize @bullet
@item The output format area describes the format of the output file and the
current position within it.
@item The edit batch area lets you change parameters in the current batch.
@item The transport controls start and stop recording different ways.
@item The batch list displays all the defined batches.
@item The confirmation area lets you determine how the output files are
imported into the timeline and quit.
@end itemize

@center @image{manual_images_en/recording,140mm}
@center @b{Recording window areas}

Recording in Cinelerra is organized around batches.  A batch essentially
defines a distinct output file for the recording.  For now you can ignore the
batch concept entirely and record merely by hitting the record button
@image{manual_images_en/record}.

The record button opens the current output file if it is not opened and writes
captured data to it.  Use the stop button to stop the recording.  Recording can
be resumed with the record button without erasing the file at this point.  In
the case of a video file, there is a single frame record button
@image{manual_images_en/singleframe} which records a single frame.

When enough media is recorded, choose an insertion method from the @b{Insertion
Strategy} menu and hit @b{close}.

@c cincvdoc_node_number_251
@node Batch recording
@subsection Batch recording
@cindex Batch recording

Now we come to the concept of batches.  Batches try to make the dumb I/O look
more like a filesystem.  Batches are traditionally used to divide tape into
different programs and save the different programs as different files instead
of recording straight through an entire tape.  Because of the high cost of
developing frame-accurate deck control mechanisms, the only use of batches now
is recording different programs during different times of day.  This is still
useful for recording TV shows or time lapse movies as anyone who can not afford
proper appliances knows.

The record window supports a list of batches and two recording modes:
interactive and batch recording.  Interactive recording happens when the record
button is pressed.  Interactive recording starts immediately and uses the
current batch to determine everything except start time.  By default, the
current batch is configured to behave like tape.

Batch recording happens when the @b{start} button is pressed.  In batch
recording, the @b{start time} is the time the batch starts recording.

First, you will want to create some batches.  Each batch has certain parameters
and methods of adjustment.

@itemize @bullet

@item @b{On} @*
Determines whether the batch is included in batch recording operations.  Click
the list row under @b{On} to enable or disable batches.

@item @b{Path} @*
It is the file the batch is going to be recorded to.  The filename specified in
the record dialog is the name of the first batch, to simplify interactive
recording, but the filename may be changed in the record window for any batch
in the @b{edit batch} area.

@item @b{News} @*
Shows whether the file exists or not.  This is a very important attribute since
there is no confirmation dialog if the file exists.  The first time you hit
record, the file is opened.  If the file exists at this point it is erased.
News says @b{File exists} if it exists and @b{OK} if it does not.  Every time
you resume recording in the same batch, the news should say @b{Open},
indicating the file is already opened and will not be erased in the next record
button press. @*
If you change out of the current batch after recording, the file is closed.
Next time you change into the batch, the file will be erased.

@item @b{Start time} @*
It is the 24 hour time of day the batch will start recording if in batch mode.
The start time may become a time of tape and reel number if deck control is
implemented but for now it is a time of day.

@item @b{Duration} @*
This is the length of the batch.  It only has meaning if the @b{Mode} of the
batch is @b{Timed}.  Once the recording length reaches @b{duration} the
recording stops, whether in interactive or batch mode.

@item @b{Source} @*
This has meaning only when the capturing hardware has multiple sources.
Usually the source is a tuner channel or input.  When the current batch
finishes and the next batch begins recording, the source is changed to what the
next batch is set to.  This way multiple TV stations can be recorded at
different times.
@end itemize

The record window has a notion of the @b{current batch}.  The current batch is
not the same as the batch which is highlighted in the batch list.  The current
batch text is colored red in the batch list.  The highlighted batch is merely
displayed in the edit batch section for editing.

By coloring the current batch red, any batch can be edited by highlighting it,
without changing the batch to be recorded.

All recording operations take place in the current batch.  If there are
multiple batches, highlight the desired batch and hit @b{activate} to make it
the current batch.  If the @b{start} button is pressed, the current batch
flashes to indicate it is waiting for the start time in batch mode.  If the
@b{record} button is pressed, the current batch is recorded immediately in
interactive mode.

In batch and interactive recording modes, when the current batch finishes
recording the next batch is activated and performed.  All future recording is
done in batch mode.  When the first batch finishes, the next batch flashes
until its start time is reached.

Interrupt either the batch or the interactive operation by hitting the stop
button.

Finally there is the @image{manual_images_en/rewind} rewind button.  In either
interactive or batch recording, the rewind button causes the current batch to
close its file.  The next recording operation in the current batch deletes the
file.

@c cincvdoc_node_number_252
@node Editing tuner information
@subsection Editing tuner information
@cindex Editing tuner information
@cindex Tuner, editing information

Sometimes in the recording process and the configuration process, you will need
to define and select tuner channels to either record or play back to.  In the
case of the Video4Linux and Buz recording drivers, tuner channels define the
source.  When the Buz driver is also used for playback, tuner channels define
the destination.

Defining tuner channels is accomplished by pushing the @image{manual_images_en/channel}
channel button.  This brings up the channel editing window.  In this window you
add, edit, and sort channels.  Also, for certain video drivers, you can adjust
the picture quality.

The @b{add} operation brings up a channel editing box.  The title of the
channel appears in the channel list.  The source of the channel is the entry in
the physical tuner's frequency table corresponding to the title.

Fine tuning in the channel edit dialog adjusts the physical frequency slightly
if the driver supports it.  The norm and frequency table together define which
frequency table is selected for defining sources.  If the device supports
multiple inputs, the input menu selects these.

To sort channels, highlight the channel in the list and push @b{move up} or
@b{move down} to move it.

Once channels are defined, the @b{source} item in the record window can be used
to select channels for recording.  The same channel selecting ability also
exists in the record monitor window.  Be aware channel selections in the record
monitor window and the record window are stored in the current batch.

For some drivers an option to @b{swap fields} may be visible.  These drivers
do not get the field order right every time without human intervention.  Toggle
this to get the odd and even lines to record in the right order.

@c cincvdoc_node_number_253
@node Capturing using dvgrab
@section Capturing using dvgrab
@cindex Dvgrab, capturing using

dvgrab is great and simple to use a command line tool to capture videos from a
DV camcorder.  When invoked it will automatically put your camera in play
mode, and start storing the videos on your hard disk.  Video files will be
labeled sequentially, as: @file{001.avi}, @file{002.avi} and so on.

To install dvgrab, use your distribution preferred installation mechanism (apt,
rpm, deb, etc) or refer to the dvgrab webpage.

Capturing videos in four easy steps:
@enumerate 1
@item Create a directory where you want the capture videos to be stored
@item @command{cd} to that directory
@item Type: @command{dvgrab --buffers 500} and @kbd{RETURN}
@item Press @kbd{CTRL-C} to stop capturing video
@end enumerate

The @option{--autosplit} option is very useful. It splits scenes according to
the timecode. However, that only works when grabbing from a DV camcorder. It
will not work when grabbing from a analog/digital converter such as a Canopus
ADVC110.

Read the dvgrab manual to get more information about dvgrab features.

@c cincvdoc_node_number_254
@node Rendering files
@chapter Rendering files
@cindex Rendering files
@cindex Files, rendering

Rendering takes a section of the timeline, performs all the editing, effects
and compositing, and stores it in a pure movie file.  You can then delete all
the source assets, play the rendered file in a movie player, or bring it back
into Cinelerra for more editing.  It is very difficult to retouch any editing
decisions in the pure movie file, however, so keep the original assets and XML
file around several days after you render it.

All rendering operations are based on a region of the timeline to be rendered.
You need to define this region on the timeline.  The navigation section
describes methods of defining regions.  @xref{Timebar}.  The
rendering functions define the region based on a set of rules.  When a region
is highlighted or in/out points are set, the affected region is rendered.  When
no region is highlighted, everything after the insertion point is rendered.
Merely by positioning the insertion point at the beginning of a track and
unsetting all in/out points, the entire track is rendered.

@menu
* Single file rendering::      Rendering a single file
* Separate files rendering::
* Insertion strategy of rendered files::
* Batch rendering::            Rendering several files unattended
* The render farm::            Rendering using many computers
* Command line rendering::     Rendering from the command line without a GUI
* Rendering options and encoding tips::
* Using background rendering::
@end menu

@c cincvdoc_node_number_255
@node Single file rendering
@section Single file rendering
@cindex Single file rendering

The fastest way to get media to disk is to use the single file rendering
function.

Go to @b{File->render} or press @kbd{SHIFT-R} to bring up the render dialog.
Select the magnifying glass @image{manual_images_en/magnify,7mm} to bring up a file
selection dialog.  This determines the filename to write the rendered file to
and the encoding parameters.

@center @image{manual_images_en/render_window,80mm}
@center @b{The render window}

In the render dialog select a format from the @b{File Format} menu.  The format
of the file determines whether you can render audio or video or both.  Select
the @b{Render audio tracks} toggle to generate audio tracks and @b{Render video
tracks} to generate video tracks.

Select the wrench @image{manual_images_en/wrench,4.33mm} next to each toggle to set
compression parameters.  If the file format can not store audio or video the
compression parameters will be blank.  If @b{Render audio tracks} or @b{Render
video tracks} is selected and the file format does not support it, trying to
render will pop up an error.

@c cincvdoc_node_number_256
@node Separate files rendering
@section Separate files rendering
@cindex Separate files rendering

The @b{Create new file at each label} option causes a new file to be created
when every label in the timeline is encountered.  This is useful for dividing
long audio recordings into individual tracks.  When using the renderfarm,
@b{Create new file at each label} causes one renderfarm job to be created at
every label instead of using the internal load balancing algorithm to space
jobs.

When @b{Create new file at each label} is selected, a new filename is created
for every output file.  If the filename given in the render dialog has a 2
digit number in it, the 2 digit number is overwritten with a different
incremental number for every output file.  If no 2 digit number is given,
Cinelerra automatically concatenates a number to the end of the given filename
for every output file.

In the filename @file{/hmov/track01.wav} the @samp{01} would be overwritten for
every output file.  The filename @file{/hmov/track.wav}; however, would become
@file{/hmov/track.wav001} and so on and so forth.  Filename regeneration is
only used when either renderfarm mode is active or creating new files for every
label is active.

@c cincvdoc_node_number_257
@node Insertion strategy of rendered files
@section Insertion strategy of rendered files
@cindex Insertion strategy of rendered files

Finally the render dialog lets you select an insertion mode.  The insertion
modes are the same as with loading files.  In this case if you select @b{insert
nothing} the file will be written out to disk without changing the current
project.  For other insertion strategies be sure to prepare the timeline to
have the output inserted at the right position before the rendering operation
is finished.  @xref{Editing}.  Editing describes how to cause output to be
inserted at the right position.

It should be noted that even if you only have audio or only have video
rendered, a @b{paste} insertion strategy will behave like a normal paste
operation, erasing any selected region of the timeline and pasting just the
data that was rendered.  If you render only audio and have some video tracks
armed, the video tracks will get truncated while the audio output is pasted
into the audio tracks.

@c cincvdoc_node_number_258
@node Batch rendering
@section Batch rendering
@cindex Batch rendering

If you want to render many projects to media files without having to repeatedly
attend to the @b{Render} dialog, @b{batch rendering} is the function to use.
In this function, you specify many EDL files to render and the unique output
files for each.  Then Cinelerra loads each EDL file and renders it
automatically, without any user intervention.  Each EDL file and its output to
be rendered are called a @b{batch}.  This allows a huge amount of media to be
processed and greatly increases the value of an expensive computer.

The first thing to do when preparing to do batch rendering is define projects
to be rendered.  The batch renderer requires a separate EDL file for every
batch to be rendered.  Set up a project and define the region to be rendered
either by highlighting it, setting in/out points around it, or positioning the
insertion point before it.  Then save the project as an EDL@.  Define as many
projects as needed this way.  The batch renderer takes the active region from
the EDL file for rendering.

With all the EDL files prepared with active regions, go to @b{File->batch
render}.  This brings up the batch rendering dialog.  The interface for batch
rendering is a bit more complex than for single file rendering.

A list of batches must be defined before starting a batch rendering operation.
The table of batches appears on the bottom of the batch render dialog and is
called @b{batches to render}.  Above this are the configuration parameters for
a single batch.

Set the @b{output path}, @b{file format}, @b{Audio}, @b{Video}, and @b{Create
new file at each label} parameters as if it was a single file.  These
parameters apply to only one batch.  In addition to the standard rendering
parameters, you must select the source EDL to use in the batch.  Do this by
setting the @b{EDL path}.

If the @b{batches to render} list is empty or nothing is highlighted, click
@b{New} to create a new batch.  The new batch will contain all the parameters
you just set.

Repeatedly press the @b{New} button to create more batches with the same
parameters.  Highlight any batch and edit the configuration on the top of the
batch render window.  The highlighted batch is always synchronized to the
information displayed.

Click and drag batches to change the order in which they are rendered.  Hit
@b{delete} to permanently remove the highlighted batch.

In the list box is a column which enables or disables the batch.  This way
batches can be skipped without being deleted.  Click on the @b{Enabled} column
in the list box to enable or disable a batch.  If it is checked, the batch is
rendered.  If it is blank, the batch is skipped.

The other columns in the batch list are informative.
@itemize @bullet
@item @b{Output} The output path of the batch.
@cindex EDL
@item @b{EDL} The source EDL of the batch.
@item @b{Elapsed} The amount of time taken to render the batch if it is
finished.
@end itemize

To start rendering from the first enabled batch, hit @b{Start}.

Once rendering, the main window shows the progress of the batch.  Once the
batch finishes, the elapsed column in the batch list is updated and the next
batch is rendered until all the enabled batches are finished.  The currently
rendering batch is always highlighted red.

To stop rendering before the batches are finished without closing the batch
render dialog, hit @b{Stop}.

To stop rendering before the batches are finished and close the batch render
dialog, hit @b{Cancel}.

To exit the batch render dialog whether or not anything is being rendered, hit
@b{Cancel}.

@c cincvdoc_node_number_259
@node The render farm
@section The render farm
@cindex Render farm

When bicubic interpolation and HDTV was first done on Cinelerra, the time
needed to produce the simplest output became unbearable even on the fastest
dual 1.7 GHz Xeon of the time.  Renderfarm support even in the simplest form
brings HDTV times back in line with SD while making SD faster than real-time.

While the renderfarm interface is not spectacular, it is simple enough to use
inside an editing suite with less than a dozen nodes without going through the
same amount of hassle you would with a several hundred node farm.  Renderfarm
is invoked transparently for all file->render operations when it is enabled in
the preferences.

Cinelerra divides the selected region of the timeline into a certain number of
jobs which are then dispatched to the different nodes depending on the load
balance.  The nodes process the jobs and write their output to individual files
on the filesystem.  The output files are not concatenated.  It is important for
all the nodes to have access to the same filesystem on the same mount point for
assets.

If a node can not access an input asset it will display error messages to its
console but probably not die.  If it can not access an output asset it will cause
the rendering to abort.

It should be noted that in the render dialog, the @b{Create new file at each
label} option causes a new renderfarm job to be created at each label instead
of by the load balancer.  If this option is selected when no labels exist, only
one job will be created.

A Cinelerra renderfarm is organized into a master node and any number of slave
nodes.  The master node is the computer which is running the GUI@.  The slave
nodes are anywhere else on the network and are run from the command line.  Run
a slave node from the command line with @command{cinelerra -d}

That is the simplest configuration.  Type @command{cinelerra -h} to see more
options.  The default port number may be overridden by passing a port number
after the @option{-d}.

Most of the time you will want to bring in the rendered output and fine tune the
timing on the timeline.  Also some file formats like MPEG can not be direct
copied.  Because of this, the jobs are left in individual files.

You can load these by creating a new track and specifying @b{concatenate to
existing tracks} in the load dialog.  Files which support direct copy can be
concatenated into a single file by rendering to the same file format with
renderfarm disabled.  Also to get direct copy, the track dimensions, output
dimensions, and asset dimensions must be equal.

MPEG files or files which do not support direct copy have to be concatenated
with a command line utility.  MPEG files can be concatenated with @b{cat}.

Configuration of the renderfarm is described in the configuration chapter
@xref{Renderfarm}.  The slave nodes traditionally read and write data to a
common filesystem over a network, thus they do not need hard drives.

Ideally all the nodes on the renderfarm have similar CPU performance.
Cinelerra load balances on a first come first serve basis.  If the last segment
is dispatched to the slowest node, all the fastest nodes may end up waiting for
the slowest node to finish while they themselves could have rendered it faster.

@c cincvdoc_node_number_260
@node Command line rendering
@section Command line rendering
@cindex Command line rendering
@cindex Rendering, command line

The command line rendering facility consists of a way to load the current set
of batch rendering jobs and process them without a GUI@.  This is useful if
you are planning on crashing X repeatedly or want to do rendering on the other
side of a low bandwidth network.  You might have access to a supercomputer in
India but still be stuck in America, exiled you might say.  A command line
interface is ideal for this.

To perform rendering from the command line, first run Cinelerra in graphical
mode.  Go to @b{file->batch render}.  Create the batches you intend to render
in the batch window and close the window.  This saves the batches in a file.
Set up the desired renderfarm attributes in @b{settings->preferences} and exit
Cinelerra.  These settings are used the next time command line rendering is
used.

On the command line run: @command{cinelerra -r}

to processes the current batch jobs without a GUI@.  Setting up all the
parameters for this operation is hard.  That is why the command line aborts if
any output files already exist.

Other parameters exist for specifying alternative files for the preferences and
the batches.  Attempting to use anything but the defaults is very involved so
it has not been tested.

@c cincvdoc_node_number_261
@node Rendering options and encoding tips
@section Rendering options and encoding tips
@cindex Rendering options and encoding tips

@menu
* Rendering to Quicktime4linux to re-encode using mencoder::
* Making a DVD::
@end menu

@c cincvdoc_node_number_262
@node Rendering to Quicktime4linux to re-encode using mencoder
@subsection Rendering to Quicktime4linux to re-encode using mencoder
@cindex Rendering to Quicktime4linux to re-encode using mencoder

If you want to reencode your rendered file with mencoder, you should export it
as a Quicktime4linux file:

@itemize @bullet
@item Audio option Two Complements 16bits (Pcm)
@item Video option DV
@end itemize

@b{Mencoder encoding options:}

@itemize @bullet
@item @b{Avi (mpeg4/divx) not for upload purposes:} @*
@command{mencoder -of avi -o cinelerra.avi -oac mp3lame -ovc lavc -lavcopts
vcodec=mpeg4:vbitrate=2500 cinelerra_render.mov}

@item @b{Internet (download):} @*
@command{mencoder -of avi -o cinelerra_render.avi -oac mp3lame -ovc lavc
-lavcopts vcodec=mpeg4:vbitrate=400 -vf scale=320:240 cinelerra_render.mov} @*
or @*
First pass: @*
@command{mencoder input.mov -ovc xvid -xvidencopts bitrate=600:pass=1 -vf
scale=320:240 -oac mp3lame -lameopts abr:br=64 -o output.avi} @*
Second pass: @*
@command{mencoder input.mov -ovc xvid -xvidencopts bitrate=600:pass=2 -vf
scale=320:240 -oac mp3lame -lameopts abr:br=64 -o output.avi}
@end itemize

Scott Frase wrote a Quicktime for GNU/Linux Compatibility Chart.  It contains an
exhaustive list of all the Quicktime compression schemes available and their
compatibility in Cinelerra, Mplayer and some other media players.  That
document has two main sections, one based on an HDV resolution-formatted
project and another based on a DV resolution-format project.

It is available here: @*
@uref{http://content.serveftp.net/video/qtcompatibility.ods}

Some interesting notes:
@itemize @bullet
@item Mplayer does behave better with smaller, DV resolution video
@item Cinelerra compatibility with files rendered from a DV project is not much
different than its compatibility with files rendered from an HDV project.
@item Comparison chart of DV/HDV mplayer/cinelerra compatibility included
@end itemize

@c cincvdoc_node_number_263
@node Making a DVD
@subsection Making a DVD
@cindex Making a DVD
@cindex DVD, making a

@menu
* Rendering to mpeg2::
* Making a DVD menu::
* Authoring a DVD::
* Burning a DVD::
@end menu

@c cincvdoc_node_number_264
@node Rendering to mpeg2
@subsubsection Rendering to mpeg2
@cindex Rendering to mpeg2
@cindex Mpeg2, rendering to

Here is a method to export mpeg2 video and make a single chapter DVD@.  This
method allows you to precisely set the encoding option you want and produces an
mpeg2 file which is 100% compatible with all DVD standalone players.

The mplex program from @b{mjpegtools} must be installed.  The mjpegtools
package is built in the hvirtual distribution and the mplex utility may be
extracted from there.

First, make sure you properly defined your cinelerra project format before
rendering your video (menu @b{Settings->Format}).  PAL is 720x576 at 25 frames
per second, and NTSC is 720x480 at 29.97 frames per second.

@enumerate 1
@item Create a script @file{~/cine_render.sh}
@item Copy in @file{~/cine_render.sh file} the following lines: @*
@command{#/bin/bash} @*
@command{mpeg2enc -v 0 -K tmpgenc -r 16 -4 1 -2 1 -D 10 -E 10 -g 15 -G 15 -q 6
-b 8600 -f 8 -o $1}
@item Put the execute permissions on that file:
@command{chmod 777 ~/cine_render.sh}
@item Open cinelerra, and select the part of the video you want to render with
the [ and ] points
@item Press @kbd{SHIFT-R}
@item Select the @b{YUV4MPEG Stream} file format
@item Deselect @b{Render audio tracks} and select @b{Render video tracks}
@item Click on the wrench
@item In the newly opened window, indicate the name of the @file{m2v} file you
want to create.  That file will contain video only.
@item Click on @b{Use pipe} and write this command:
@command{/home/<your_user>/cine_render.sh %}
@item Click OK to close the second window, and OK again to render your
@file{m2v} file
@item When the m2v file is rendered, open the rendering window again, and
render an AC3 file at 224kbits
@item Finally, combine video and audio with this command:
@command{mplex -f 8 your_video_file.m2v your_audio_file.ac3 -o
video_audio_file.mpeg}
@end enumerate

You can modify the mpeg2enc parameters if you want to.  Look at the mpeg2enc
manpage.  Some details about the settings:
@itemize
@item @option{-b 8600} : This is the maximum bitrate of your @file{m2v} file (it
does not include the audio bitrate).  We recommend you to do not increase that
value or you could get errors when mplexing the video and the audio.
@item @option{-q 6} : This is the quantizer setting.  If you reduce it (do not go
below 3), the quality increases.  But the bitrate will increase.  It's
recommended to keep the medium bitrate achieved (that's displayed when mplexing
the audio and video files) around 10% lower than the bitrate defined with the
@option{-b} setting.
@end itemize

If your material is noisy (Hi8 analog material for example), you can add some
mjpegtools in the command line written in @file{~/cine_render.sh}:
@itemize @bullet
@item @command{y4mshift} and @command{y4mscaler} can be used to remove the
noisy borders around the video.  For example, those commands added at the
beginning of the command line in @file{cine_render.sh} remove the black borders
around a Hi8 video: @*
@command{y4mshift -n -2 | yuvscaler -I USE_744x560+12+8 -O DVD -M BICUBIC |}
@item @command{yuvdenoise} and @command{yuvmedianfilter} can help removing
noise.  Example: @*
@command{yuvdenoise -F | yuvmedianfilter -T 3 |} @*
Denoising is a complex task, and the options given above are just an example.
Please read the mjpegtools'manual and subscribe to its mailing-list for more
information.
@end itemize

@c cincvdoc_node_number_265
@node Making a DVD menu
@subsubsection Making a DVD menu
@cindex Making a DVD menu

A DVD menu is composed of:

@itemize @bullet
@item a background (still image or video)
@item buttons
@item sound/music
@end itemize

You can build a menu with a GUI such as qdvdauthor, dvdstyler, dvdwizard or
tovid.  However, using those GUI is not perfect for the moment, since they are
bugged or limited for the moment.

The method we explain below is more complicated than using a GUI, however it:

@itemize @bullet
@item produces DVD playable on all standalone players
@item is not subject to bugs
@item will save you a lot of time since all you have to do to author a new DVD
is to modify text files
@end itemize

If you prefer to use a GUI, we recommend you to try tovid: @*
@uref{http://tovid.wikia.com/wiki/Main_Page}

Here are the steps needed to create your DVD menu:

@itemize @bullet
@item create the menu background with cinelerra
@item add the buttons by creating PNG images
@item combine the menu and the buttons with spumux
@end itemize

We suppose that you want to create a menu with an animated background.  Launch
Cinelerra and create a project containing what you want to be the background of
the menu.  You can add a music if you wish to.  Pay attention to the fact that
this menu will play in loop.

To draw the buttons, you have two possibilities:

@itemize @bullet
@item display them in Cinelerra.  That way, you will be able to make animated
buttons, such as a video thumbnail for each part of your video.
@item do not draw the buttons in Cinelerra.  You will add them later on, from
PNG images "added" to the mpeg2 menu file.  This is the simpliest method, but
you won't be able to display animated buttons.
@end itemize

Render that video into m2v and ac3 using the @command{cine_render.sh} method
explain above.  Combine the audio and video with mplex as you would do with any
"normal" video.

You obtain a mpeg2 file containing the menu background, and some buttons
displayed above it if you added them in Cinelerra.

We have to use spumux to define each button position in that mpeg2 file.  If
you did not draw the buttons in Cinelerra, you will be able to put them in with
spumux.

Spumux is a command line utility which takes 2 arguments:

@itemize @bullet
@item an XML file explaining where the buttons are
@item the mpeg2 file name (the one you rendered for the menu)
@end itemize

Here is a spumux example XML file:
@verbatim
<subpictures>
 <stream>
  <spu start="00:00:00.0" image="buttons_normal.png" highlight=
  "buttons_highlight.png" select="buttons_select.png">
   <button name="1" x0="94 " y0="234 " x1="253 " y1="278"
   down="2" right="4" />
   <button name="2" x0="63 " y0="287 " x1="379 " y1="331" up="1"
   down="3" right="5" />
  </spu>
 </stream>
</subpictures>
@end verbatim

@itemize @bullet
@item @b{image="buttons_normal.png"} This png image contains the buttons as
they should appear when they are not selected nor highlighted.
@item @b{highlight="buttons_highlight.png"} This png image contains the buttons
in their highlighted state.
@item @b{select="buttons_select.png} This png image contains the buttons in
their selected state.
@end itemize

If you already made the buttons in Cinelerra, you have to specify empty (100%
transparent) PNG images here.

The PNG images used in spumux must:

@itemize @bullet
@item contain an @b{alpha channel} (ie support transparency)
@item be in @b{4 indexed colors}. You can easily convert an image to 4 indexed
colors using Gimp.
@end itemize

There is one line per button. Each line contains the button coordinates, a
button having a rectangular shape:

@itemize @bullet
@item @b{x0, y0}: upper left corner
@item @b{x1, y1}: bottom right corner
@end itemize

You also have to set which button to move to when using the up, down, left and
right buttons of the DVD remote.  Here is an example:

@verbatim
<button name="3" ...coordinates... up="1" down="5" left="2" right="4" />
@end verbatim

When button 3 is selected, if the "Up" key is pressed on the remote then the
button 1 will be highlighted.  If the "Right" key is pressed on the remote,
then button 4 will be highlighted.

When you have finished editing your spumux XML file, you have to type this
command: @*
@command{spumux menu.xml < menu.mpeg > menu_with_buttons.mpeg} @*
That will make a @file{menu_with_buttons.mpeg}. It is an mpeg2 files with
buttons.

@c cincvdoc_node_number_266
@node Authoring a DVD
@subsubsection Authoring a DVD
@cindex Authoring a DVD

After having rendered to mpeg2 your video files, and having prepared a menu
with spumux, you need to "author" the DVD with dvdauthor.

Dvdauthor uses XML files to describe the DVD structure.  The syntax is
rigorous, and one should really pay a lot of attention to the .xml file syntax.
The risk is the DVD to be readable on some standalone players, but not all of
them.

To help you start using dvdauthor, we will show you some example XML files.

@verbatim
<dvdauthor dest="/path/to/the/folder/which/will/contain/the/dvd">
    <vmgm />
    <titleset>
        <titles>
            <pgc>
                <vob file="/the/mpeg/file.mpeg" />
                <post>
                    jump chapter 1;
                </post>
            </pgc>
        </titles>
    </titleset>
</dvdauthor>
@end verbatim

This is a very simple dvdauthor XML file. There are no menus, so the video file
@file{/the/mpeg/file.mpeg} will be played as soon as you insert the DVD in the
player.

The command in <post> tag means the video should be played in a loop.  When the
DVD player reaches the end of the video, it will jump to the first chapter of
the video (which dvdautor assumes to be the beginning of the video since
chapters haven't been defined).

To author the DVD, just type the following command: @*
@command{dvdauthor -x simple_example.xml}

Now, let's have a look at a more complex example. When the DVD is inserted, a
menu is displayed and you can choose to play any of 4 videos.

@verbatim
<dvdauthor dest="/path/to/the/folder/which/will/contain/the/dvd" jumppad="yes" >
<vmgm>
 <fpc> jump menu 1; </fpc>
  <menus>
   <video format="pal" aspect="4:3" resolution="720x576" />
   <pgc entry="title" >
    <vob file="menu.mpeg" pause="0" />
    <button name="1" > { g3=1; jump titleset 1 menu entry root; } </button>
    <button name="2" > { g3=2; jump titleset 1 menu entry root; } </button>
    <button name="3" > { g3=3; jump titleset 1 menu entry root; } </button>
    <button name="4" > { g3=4; jump titleset 1 menu entry root; } </button>
     <post> { jump cell 1; } </post>
   </pgc>
  </menus>
 </vmgm>
 <titleset>
  <menus>
   <pgc entry="root" >
    <pre> { if ( g3 gt 0 )  {
                if ( g3 eq 1 ) { g3=0; jump title 1  chapter 1; }
                if ( g3 eq 2 ) { g3=0; jump title 1  chapter 3; }
                if ( g3 eq 3 ) { g3=0; jump title 1  chapter 5; }
                if ( g3 eq 4 ) { g3=0; jump title 1  chapter 7; }
                jump vmgm menu entry title;
                }
        } </pre>
    <post> { jump vmgm menu entry title; } </post>
   </pgc>
  </menus>
  <titles>
   <video format="pal" aspect="4:3" resolution="720x576" />
   <pgc pause="0" >
    <vob file="video_1.mpeg" pause="0" />
    <vob file="blackvideo.mpg" pause="0" />
    <vob file="video_2.mpeg" pause="0" />
    <vob file="blackvideo.mpg" pause="0" />
    <vob file="video_3.mpeg" pause="0" />
    <vob file="blackvideo.mpg" pause="0" />
    <vob file="video_4.mpeg" pause="0" />
    <post> { call vmgm menu entry title; } </post>
   </pgc>
  </titles>
 </titleset>
</dvdauthor>
@end verbatim

The file @file{blackvideo.mpg} is used to add a 2 second black screen between
each video. Here is how to create it: @*
@command{convert -size 720x576 xc:black -depth 8 blackframe.ppm} @*
@command{dd if=/dev/zero bs=4 count=960000 | toolame -b 128 -s 48 /dev/stdin
emptyaudio.mpa} @*
@command{ppmtoy4m -S 420mpeg2 -n 50 -F 25:1 -r blackframe.ppm | mpeg2enc -a 2
-n p -f 8 -o blackvideo.mpv} @*
@command{mplex -f 8 -o blackvideo.mpg blackvideo.mpv emptyaudio.mpa}

@c cincvdoc_node_number_267
@node Burning a DVD
@subsubsection Burning a DVD
@cindex Burning a DVD

When you have finished authoring the DVD, you will find in the destination folder the
following directories: @file{AUDIO_TS} and @file{VIDEO_TS}.  To test your DVD
before burning it, cd into this folder, and type: @*
@command{xine dvd:`pwd`}

If your DVD plays fine on your computer, it is time to burn it.  When you are
in the folder containing @file{AUDIO_TS} and @file{VIDEO_TS}, type this
command (adjusting for your dvd burner device, eg /dev/dvdrw): @*
@command{nice -n -20 growisofs -dvd-compat -speed=2 -Z /dev/dvd -dvd-video -V
VIDEO ./ && eject /dev/dvd}

If you have a lot of copies to do, you can first make an .iso master using this
command: @*
@command{nice -n -20 mkisofs -dvd-video -V VIDEO -o ../dvd.iso .} @*
This @file{../dvd.iso} file can be burnt using this command: @*
@command{nice -n -20 growisofs -dvd-compat -speed=2 -Z /dev/dvd=../dvd.iso &&
eject /dev/cdrom}

We recommend you do not burn at a speed higher than 4x.  Use good quality DVD-R
only.

To test your DVD on a standalone player without wasting several DVD-R, you can
burn on DVD-RW@.  First, format your DVD-RW using this command: @*
@command{dvd+rw-format -lead-out /dev/dvd} @*
Then, burn the DVD-RW using the commands above.

@c cincvdoc_node_number_268
@node Using background rendering
@section Using background rendering
@cindex Background rendering
@cindex Rendering, background

Background rendering allows impossibly slow effects to play back in real-time
shortly after the effect is pasted in the timeline.  It continuously renders
temporary output.  When renderfarm is enabled, background rendering uses the
renderfarm continuously.  This way, any size video can be seen in real-time
merely by creating a fast enough network with enough nodes.

Background rendering is enabled in settings->preferences->performance.  It has
one interactive function: @b{settings->set background render}.  This sets the
point where background rendering begins to where the in point is.  If any video
exists, a red bar appears in the timeline showing what has been background
rendered.

It is often useful to insert an effect or a transition and then select
settings->set background render right before the effect to preview it at full
framerates.

@c cincvdoc_node_number_269
@node Tips
@chapter Tips
@cindex Tips

In this section, you will find ways to solve common problems using Cinelerra.
This section is arranged in order of the problems and what tools are used to
solve them.  Following sections are arranged in order of the tools and their
uses.

@menu
* Encoding into Dolby Pro Logic::
* Cleaning analog TV::
* Defeating interlacing::
* Making video look like film::
* Clearing out haze::
* Making a ringtone::
* Time stretching audio::
* Video screen captures::
* Improving performance::             Making Cinelerra run better on GNU/Linux.
* Translating Cinelerra::             How to translate Cinelerra to different languages.
* Panning and zooming still images::
@end menu

@c cincvdoc_node_number_270
@node Encoding into Dolby Pro Logic
@section Encoding into Dolby Pro Logic
@cindex Dolby Pro Logic, encoding into

Dolby pro logic is an easy way to output 6 channel audio from a 2-channel
soundcard with degraded but useful results.  Rudimentary Dolby pro logic
encoding can be achieved with clever usage of the effects.

First, create the front left and right channels. Create 2 audio tracks, each
carrying either the left or right channel. Pan the left channel to the left and
the right channel to the right with @b{pan}.

Next, create the rear left and right channels. Create another 2 audio tracks as
above: the left channel panned left and the right channel panned right.  Then
apply @b{invert audio} to both new channels and the signals will come out of
the rear speakers.

Next, create the center channel by creating a single audio track with monaural
audio from a different source.  Center it with the @b{pan} control and the
signal will come out of the center speaker.

If a copy of the signal in the back speakers is desired in any single
front speaker, the signal in the back speakers must be delayed by at least 0.05
seconds and a single new track should be created.  Pan the new track to orient
the signal in the front speakers.

If the same signal is desired in all the speakers except the center speaker,
delay the back speakers by 0.5 seconds and delay either the front left or front
right by 0.2 seconds.

If you want to hear something from the subwoofer, create a new track, select a
range, drop a synthesizer effect, and set the frequency below 60 Hz.  The
subwoofer merely plays anything below 60Hz or so.

Other tricks you can perform to separate the speakers are parametric
equalization to play only selected ranges of frequencies through different
speakers and lowpass filtering to play signals through the subwoofer.

@c cincvdoc_node_number_271
@node Cleaning analog TV
@section Cleaning analog TV
@cindex Cleaning analog TV
@cindex TV, cleaning analog

Unless you live in a rich nation like China or are a terrorist, you probably
record analog TV more than you record digital TV@.  The picture quality on
analog TV is horrible but you can do things in Cinelerra to make it look more
like it did in the studio.

First, when capturing the video, capture it in the highest resolution possible.
For Europeans it is 720x576 and for North Americans it is 720x480.  Do not bother
adjusting the brightness or contrast in the recording monitor, although maxing
out the color is useful.  Capture it using MJPEG or uncompressed Component
Video if possible.  If those are too demanding, then capture it using JPEG@.
RGB should be a last resort.

Now on the timeline use @b{Settings->Format} to set a YUV colorspace.  Drop a
@b{Downsample} effect on the footage.  Set it for
@verbatim
Horizontal:        2
Horizontal offset: 0
Vertical:          2
Vertical offset:   0
      red
  x   green
  x   blue
      alpha
@end verbatim

Use the camera tool to shift the picture up or down a line to remove the most
color interference from the image.  This is the difference we are looking for:

@center @image{manual_images_en/cleaning1}

If you have vertical blanking information or crawls which constantly change in
each frame, block them out with the @b{Mask} tool.  This improves compression
ratios.

This is about all you can do without destroying more data than you would
naturally lose in compression.  The more invasive cleaning techniques involve
deinterlacing.

@c cincvdoc_node_number_272
@node Defeating interlacing
@section Defeating interlacing
@cindex Interlacing, defeating

Interlacing is done on most video sources because it costs too much to build
progressive scanning cameras and progressive scanning CRT's.  Many a consumer
has been disappointed to spend 5 paychecks on a camcorder and discover what
horrible jagged images it produces on a computer monitor.

As for progressive scanning camcorders, forget it.  Cost factors are probably
going to keep progressive scanning cameras from ever equaling the spatial
resolution of interlaced cameras.  Interlacing is here to stay.  That is why
they made deinterlacing effects in Cinelerra.

We do not believe there has ever been a perfect deinterlacing effect.  They are
either irreversible or do not work.  Cinelerra cuts down the middle by providing
deinterlacing tools that are irreversible sometimes and do not work sometimes
but are neither one or the other.

@itemize @bullet
@item
@cindex Interlacing, line doubling
@b{Line Doubling}
This one is done by the @b{Deinterlace} effect when set to @b{Odd lines} or
@b{Even lines}.  When applied to a track it reduces the vertical resolution by
1/2 and gives you progressive frames with stairstepping.  This is only useful
when followed by a scale effect which reduces the image to half its size.

@item
@cindex Interlacing, line averaging
@b{Line averaging}
The @b{Deinterlace} effect when set to @b{Average even lines} or @b{Average odd
lines} does exactly what line doubling does except instead of making straight
copies of the lines it makes averages of the lines.  This is actually useful
for all scaling. @*
There is an option for adaptive line averaging which selects which lines to line
average and which lines to leave interlaced based on the difference between the
lines.  It does not work.

@item
@cindex Interlacing, inverse telecine
@b{Inverse Telecine}
This is the most effective deinterlacing tool when the footage is an NTSC TV
broadcast of a film.  @xref{Inverse telecine}.

@item
@cindex Interlacing, time base correction
@b{Time base correction}
The first three tools either destroy footage irreversibly or do not work
at times.  @b{Time base correction} is last because it is the perfect
deinterlacing tool.  It leaves the footage intact.  It does not reduce
resolution, perceptually at least.  It does not cause jittery timing.

@item
@cindex Interlacing, frames to fields
The @b{Frames to Fields} effect converts each frame to two frames, so it must
be used on a timeline whose project frame rate is twice the footage's frame
rate.  In the first frame it puts a line-averaged copy of the even lines.  In
the second frame it puts a line-averaged copy of the odd lines.  When played
back at full framerates it gives the illusion of progressive video with no loss
of detail. @*
Best of all, this effect can be reversed with the @b{Fields to frames} effect.
That one combines two frames of footage back into the one original interlaced
frame of half the framerate. @*
Be aware that frames to fields inputs frames at half the framerate as the
project.  Effects before frames to fields process at the reduced framerate. @*
Unfortunately, the output of @b{Frames to Fields} can not be compressed as
efficiently as the original because it introduces vertical twitter and a super
high framerate. @*
Interlaced 29.97 fps footage can be made to look like film by applying
@b{Frames to Fields} and then reducing the project frame rate of the resulting
59.94 fps footage to 23.97 fps.  This produces no timing jitter and the
occasional odd field gives the illusion of more detail than there would be if
you just line averaged the original.
@end itemize

@cindex Interlacing, HDTV exceptions
@cindex HDTV interlacing
@b{HDTV exceptions} @*
1920x1080 HDTV is encoded a special way.  If it is a broadcast of original HDTV
film, an inverse telecine works fine.  If it is a rebroadcast of a 720x480
source, you need to use a time base and line doubling algorithm to deinterlace
it, @xref{1080 to 480}.

@c cincvdoc_node_number_273
@node Making video look like film
@section Making video look like film
@cindex Making video look like film
@cindex Film look

Video sweetening is constantly getting better.  Lately the best thing you can
do for dirt cheap consumer camcorder video is to turn it into progressive 24
fps output.  While you can not really do that, you can get pretty close for the
money.  Mind you, since this procedure can degrade high quality video just as easily
as it improves low quality video, it should only be used for low quality
video.

@enumerate 1
@item Set project framerate to twice the video framerate.
@item Apply a @b{Sharpen} effect.  Set it to sharpness: 25, no interlacing, and
horizontal only.
@item Drop a @b{Frame to Fields} effect on the same track.  Set Average Empty
Rows to on and play through the video a few times to figure out which field is
first.  If the wrong field is first, the motion is shaky.  Secondly, any
editing in the doubled frame rate may now screw up the field order.  We are
still figuring out the easiest way to support warnings for field glitches but
for now you need to go back to the normal framerate to do editing or play test
to make sure the fields are right.
@item Render just the video to the highest quality file possible.
@item Import the video back to a new track.  Set the project framerate to
24.  The new track should now display more filmish and sharper images than the
original footage.
@end enumerate

This entire procedure could be implemented in one non-realtime effect, but the
biggest problem with that is you will most often want to keep the field based
output and the 24 fps output for posterity.  A non-realtime effect would
require all that processing just for the 24 fps copy.  Still debating that one.

@c cincvdoc_node_number_274
@node Clearing out haze
@section Clearing out haze
@cindex Clearing out haze
@cindex Haze, clearing out
@cindex Gradient effect

You probably photograph a lot of haze and never see blue sky.  Even if you
can afford to briefly go somewhere where there is blue sky, horizon shots
usually can stand for more depth.  This is what the @b{gradient effect} is for.

Drop the gradient effect on hazy tracks.  Set the following parameters:
@itemize @bullet
@item Angle: 0
@item Inner radius: 0
@item Outer radius: 40
@item Inner color: blue 100% alpha
@item Outer color: blue 0% alpha
@end itemize

It is important to set the 0% alpha color to blue even though it is 0% alpha.
The color of the outer alpha is still interpolated with the inner color.  This
is a generally applicable setting for the gradient.  Some scenes may work
better with orange or brown for an evening feel.

@c cincvdoc_node_number_275
@node Making a ringtone
@section Making a ringtone
@cindex Ringtone, making a

This is how we made ringtones for the low end Motorola V180's and it will
probably work with any new phone.  Go to @b{File->Load files...} and load a
sound file with Insertion strategy: @b{Replace current project}.  Go to
@b{Settings->Format} change @b{Channels} to 1 and @b{Samplerate} to 16000 or
22050.

Either highlight a region of the timeline or set in/out points to use for the
ringtone.  To improve sound quality on the cell phone, you need the maximum
amplitude in as many parts of the sound as possible.  Right click on track
Audio 1 and select @b{Attach effect..}.  Highlight the @b{Compressor} effect
and hit @b{Attach} in the attachment popup.

Make sure the insertion point or highlighted area is in the region with the
Compressor effect.  Right click on track Audio 2 and select @b{Attach
effect..}.  Highlight @b{Audio 1: Compressor} and hit @b{Attach}.  Click the
Audio1 Compressor's magnifying glass @image{manual_images_en/magnify,7mm} to bring up
the compressor GUI@.

Set the following parameters:
@itemize @bullet
@item Reaction secs: @b{-0.1}
@item Decay secs: @b{0.1}
@item Trigger Type: @b{Total}
@item Trigger: @b{0}
@item Smooth only: @b{No}
@end itemize

Click @b{Clear} to clear the graph.  Click anywhere in the grid area and drag a
new point to 0 Output and -50 Input.  The graph should look like this.

@center @image{manual_images_en/compress,70mm}

Go to @b{File->Render}.  Specify the name of an mp3 file to output to.  Set the
file format to @b{MPEG Audio}.  Click the wrench
@image{manual_images_en/wrench,4.33mm} for Audio and set @b{Layer} to @b{III} and
@b{Kbits per second} to @b{24} or @b{32}.  Check @b{Render audio tracks} and
uncheck @b{Render video tracks}.  Hit OK to render the file.

The resulting @file{.mp3} file must be uploaded to a web server.  Then, the
phone's web browser must download the @file{.mp3} file directly from the URL@.
There also may be a size limit on the file.

@c cincvdoc_node_number_276
@node Time stretching audio
@section Time stretching audio
@cindex Time stretching audio
@cindex Audio, time stretching

It may appear that time stretching audio is a matter of selecting a region of
the audio tracks, enabling recording for the desired tracks, going to
@b{Audio->Render Effect}, and applying @b{Time Stretch}.  In actuality there
are 3 audio effects for time stretching: @b{Time Stretch}, @b{Resample}, and
@b{Asset info dialog}.

Time Stretch applies a fast Fourier transform to try to change the duration
without changing the pitch, but this introduces windowing artifacts to the
audio.  It is only useful for large changes in time because obvious changes in
duration make windowing artifacts less obtrusive.

For smaller changes in duration, in the range of 5%, @b{Resample} should be
used.  This changes the pitch of the audio but small enough changes are not
noticeable.  Resample does not introduce any windowing artifacts, so this is
most useful for slight duration changes where the listener is not supposed to
know what is going on.

Another way to change duration slightly is to go to the @b{Resources} window,
highlight the @b{media} folder, right click on an audio file, click on
@b{Info}.  Adjust the sample rate in the @b{Info} dialog to adjust the
duration.  This method also requires left clicking on the right boundary of the
audio tracks and dragging left or right to correspond to the length changes.

@c cincvdoc_node_number_277
@node Video screen captures
@section Video screen captures
@cindex Video screen captures

We explain here how to record video screen captures and edit them in Cinelerra.

First, you have to record the video with xvidcap.  You can find that utility in
most distributions' repositories, or download it here: @*
@uref{http://xvidcap.sourceforge.net}

First, capture the screen: @*
@command{xvidcap --fps 10 --cap_geometry 1280x1024+0+0 --file "file1.mpeg"
--gui no --audio no}

Do not forget to change the geometry option according to your screen size.
Then, convert the @file{file1.mpeg} file you obtained into an mpeg file
suitable for Cinelerra: @*
@command{ffmpeg -r 10 -i file1.mpeg -s 1280x1024 -b 3000 -aspect 1.33 -r 25
file2.mpeg}

You can now load that file into Cinelerra.  Make sure you properly set the
video format of your project (size, frame-rate, aspect-ratio)

When you have finished editing your video, you have to render it.  Render it as
a jpeg-sequence. It is recommended to write the jpeg files in a new folder,
since there probably will have a lot of files created.

Then, open a shell window, and cd into that folder.  Encode the jpeg files
using those commands:

First pass: @*
@command{mencoder "mf://*.jpg" -mf fps=25 -oac pcm -sws 2 -vf
scale=1280:1024,hqdn3d=2:1:2 -ovc lavc -lavcopts
vcodec=mpeg4:vbitrate=800:aspect=4/3:vpass=1 -ofps 10 -of avi -o /dev/null
-ffourcc DIVX} @*
Second pass: @*
@command{mencoder "mf://*.jpg" -mf fps=25 -oac pcm -sws 2 -vf
scale=1280:1024,hqdn3d=2:1:2 -ovc lavc -lavcopts
vcodec=mpeg4:vbitrate=800:aspect=4/3:vpass=2 -ofps 10 -of avi -o
../rendered_file.avi -ffourcc DIVX}

You can also render the video to mpeg4 directly from Cinelerra if you wish to.

@c cincvdoc_node_number_278
@node Improving performance
@section Improving performance
@cindex Performance, improving

For the moment GNU/Linux is not an excellent desktop.  It is more of a server.
Most of what you will find on modern GNU/Linux distributions are faceless,
network-only programs strategically designed to counteract one Microsoft server
feature or another and not to perform very well at user interaction.  There are
a number of parameters on GNU/Linux, which ordinary people can adjust to make
it behave more like a thoroughbred in desktop usage.

@menu
* Disabling swap space::
* Enlarging sound buffers::
* Freeing more shared memory::
* Speeding up the hard drive::
* Disabling cron::
* Reducing USB mouse sensitivity::
* Assorted X tweeks::
* Speeding up the file system::
* Improving Zoran video::
@end menu

@c cincvdoc_node_number_279
@node Disabling swap space
@subsection Disabling swap space
@cindex Disabling swap space
@cindex Swap space, disabling

On systems with lots of memory, Cinelerra sometimes runs better without a swap
space.  If you have 4 GB of RAM, you are probably better off without a swap
space.  If you have 512MB of RAM, you should keep the swap.  If you want to do
recording, you should probably disable swap space in any case.  There is a
reason for this.  GNU/Linux only allows half the available memory to be used.
Beyond that, it starts searching for free pages to swap, in order to cache more
disk access.  In a 4 GB system, you start waiting for page swaps after using
only 2 GB@.

The question then is how to make GNU/Linux run without a swap space.
Theoretically it should be a matter of running @*
@command{swapoff -a}

Unfortunately, without a swap space the kswapd tasklet normally spins at 100%.
To eliminate this problem, edit @file{linux/mm/vmscan.c}.  In this file, put a
line saying @command{return 0;} before it says
@verbatim
    /*
     * Kswapd main loop.
     */
@end verbatim

Then recompile the kernel.

@c cincvdoc_node_number_280
@node Enlarging sound buffers
@subsection Enlarging sound buffers
@cindex Sound buffers, enlarging

In order to improve realtime performance, the audio buffers for all the
GNU/Linux sound drivers were limited from 128k to 64k.  For recording audio and
video simultaneously and for most audio recording this causes dropouts.
Application of low latency and preemptible kernel patches make it possible to
record more audio recordings but it does not improve recording video with audio.
This is where you need to hack the kernel.

To see if your sound buffers are suitable, run the included soundtest program
with nothing playing or recording.  This allocates the largest possible buffers
and displays them.  If the @b{Total bytes available} is under 131072, you need
to see about getting the buffers enlarged in the driver.  While many drivers
differ, we have a hack for at least one driver.

This only applies to the OSS version of the Soundblaster Live driver.  Since
every sound card and every sound driver derivative has a different
implementation you will need to do some searching for other sound cards.  Edit
@file{linux/drivers/sound/emu10k1/audio.c}

Where it says
@verbatim
if (bufsize >= 0x10000)
@end verbatim
change it to:
@verbatim
if (bufsize > 0x40000)
@end verbatim

Where it says
@verbatim
    for (i = 0; i < 8; i++)
        for (j = 0; j < 4; j++)
@end verbatim
change it to:
@verbatim
    for (i = 0; i < 16; i++)
        for (j = 0; j < 4; j++)
@end verbatim

In @file{linux/drivers/sound/emu10k1/hwaccess.h}, change
@verbatim
#define MAXBUFSIZE 65536
@end verbatim
to
@verbatim
#define MAXBUFSIZE 262144
@end verbatim

Finally, in @file{linux/drivers/sound/emu10k1/cardwi.h}, change
@verbatim
#define WAVEIN_MAXBUFSIZE         65536
@end verbatim
to
@verbatim
#define WAVEIN_MAXBUFSIZE         262144
@end verbatim

Then recompile the kernel modules.

@c cincvdoc_node_number_281
@node Freeing more shared memory
@subsection Freeing more shared memory
@cindex Freeing more shared memory
@cindex Shared memory, freeing
@cindex Memory, freeing

The GNU/Linux kernel only allows 32MB of shared memory to be allocated by
default.  This needs to be increased to do anything useful.  Run the following
command: @*
@command{echo "0x7fffffff" > /proc/sys/kernel/shmmax}

@c cincvdoc_node_number_282
@node Speeding up the hard drive
@subsection Speeding up the hard drive
@cindex Speeding up the hard drive
@cindex Hard drive, speeding up the
@cindex hdparm

This is a very popular command sequence among GNU/Linux gurus, which is not
done by default on GNU/Linux distributions. @*
@command{hdparm -c3 -d1 -u1 -k1 /dev/hda}

@itemize @bullet
@item @option{-c3} puts the hard drive into 32 bit I/O with sync.  This
normally does not work due to inept kernel support for most IDE controllers.
If you get lost interrupt or SeekComplete errors, quickly use @option{-c0}
instead of @option{-c3} in your command.
@item @option{-d1} enables DMA of course.  This frees up the CPU partially
during data transfers.
@item @option{-u1} allows multiple interrupts to be handled during hard drive
transactions.  This frees up even more CPU time.
@item @option{-k1} prevents GNU/Linux from resetting your settings in case of a
glitch.
@end itemize

@c cincvdoc_node_number_283
@node Disabling cron
@subsection Disabling cron
@cindex Disabling cron
@cindex Cron, disabling

GNU/Linux runs some daily operations like compressing man pages.  These may be
acceptable background tasks while compiling or word processing but not while
playing video.  Disable these operations by editing
@file{/etc/rc.d/init.d/anacron}.

Put @command{exit} before the first line not beginning in @command{#}.

In @file{/etc/rc.d/init.d/crond} put @command{exit} before the first line not
beginning in @command{#}.  Then reboot.

You can not use the @command{at} command anymore, but who uses that command
anyways?

@c cincvdoc_node_number_284
@node Reducing USB mouse sensitivity
@subsection Reducing USB mouse sensitivity
@cindex Reducing USB mouse sensitivity
@cindex Mouse, reducing sensitivity
@cindex USB mouse, reducing sensitivity

Gamers like high resolution mice, but this can be painful for precisely
positioning the mouse on a timeline or video screen.  XFree86 once allowed you
to reduce PS/2 mouse sensitivity using commands like @command{xset m 1 1} but
you are out of luck with USB mice or KVM's.

We have a way to reduce USB mouse sensitivity but it requires editing the
kernel source code.  Even though USB mice have been supported for years, the
kernel source code for USB mice is constantly being rewritten.  These
instructions were relevant for 2.6.12.3.  Edit
@file{/usr/src/linux/drivers/input/mousedev.c}.

After the line saying
@verbatim
struct mousedev_hw_data {
@end verbatim
put
@verbatim
#define DOWNSAMPLE_N 100
#define DOWNSAMPLE_D 350
int x_accum, y_accum;}
@end verbatim

Next, the section which says something like:
@verbatim
switch (code) {
    case REL_X: mousedev->packet.dx += value; break;
    case REL_Y: mousedev->packet.dy -= value; break;
    case REL_WHEEL:     mousedev->packet.dz -= value; break;
}
@end verbatim
must be replaced by
@verbatim
switch (code) {
    case REL_X:
    mousedev->packet.x_accum += value * DOWNSAMPLE_N;
    mousedev->packet.dx += (int)mousedev->packet.x_accum
    / (int)DOWNSAMPLE_D;
    mousedev->packet.x_accum -=
    ((int)mousedev->packet.x_accum / (int)DOWNSAMPLE_D)
    * (int)DOWNSAMPLE_D;
    break;
    case REL_Y:
    mousedev->packet.y_accum += value * DOWNSAMPLE_N;
    mousedev->packet.dy -= (int)mousedev->packet.y_accum
    / (int)DOWNSAMPLE_D;
    mousedev->packet.y_accum -=
    ((int)mousedev->packet.y_accum
    / (int)DOWNSAMPLE_D) * (int)DOWNSAMPLE_D;
    break;
    case REL_WHEEL: mousedev->packet.dz -= value; break;
}
@end verbatim

Change the value of @b{DOWNSAMPLE_N} to change the mouse sensitivity.

@c cincvdoc_node_number_285
@node Assorted X tweeks
@subsection Assorted X tweeks
@cindex X, assorted tweeks

XFree86 by default can not display Cinelerra's advanced pixmap rendering very
fast.  The X server stalls during list box drawing.  Fix this by adding a line
to your XF86Config* files.

In the @b{Section "Device"} area, add a line saying:
@verbatim
Option "XaaNoOffscreenPixmaps"
@end verbatim
and restart the X server.

Screen blanking is really annoying, unless you are fabulously rich and can
afford to leave your monitor on 24 hours a day without power saving mode.  In
@file{/etc/X11/xinit/xinitrc} put
@verbatim
xset s off
xset s noblank
@end verbatim
before the first @command{if} statement.

How about those windows keys which no GNU/Linux distribution even thinks to
use.  You can make the window keys provide ALT functionality by editing
@file{/etc/X11/Xmodmap}.  Append the following to it.
@verbatim
keycode 115 = Hyper_L
keycode 116 = Hyper_R
add mod4 = Hyper_L
add mod5 = Hyper_R
@end verbatim

The actual changes to a window manager to make it recognize window keys for
@key{ALT} are complex.  In @b{FVWM} at least, you can edit
@file{/etc/X11/fvwm/system.fvwm2rc} and put
@verbatim
Mouse 0 T A move-and-raise-or-raiselower
#Mouse 0 W M move
Mouse 0 W 4 move
Mouse 0 W 5 move
Mouse 0 F A resize-or-raiselower
Mouse 0 S A resize-or-raiselower
@end verbatim

in place of the default section for moving and resizing.  Your best performance
is going to be on FVWM@.  Other window managers seem to slow down video with
extra event trapping and are not as efficient in layout.

@c cincvdoc_node_number_286
@node Speeding up the file system
@subsection Speeding up the file system
@cindex Speeding up the file system
@cindex File system, speeding up the

You will often store video on an expensive, gigantic disk array separate from
your boot disk.  You will thus have to manually install an EXT filesystem on this
disk array, using the @command{mke2fs} command.  By far the fastest file system
is @*
@cindex mke2fs
@cindex tune2fs
@command{mke2fs -i 65536 -b 4096 my_device} @*
@command{tune2fs -r0 -c10000 my_device}

This has no journaling, reserves as few blocks as possible for filenames, and
accesses the largest amount of data per block possible.  A slightly slower file
system, which is easier to recover after power failures is @*
@command{mke2fs -j -i 65536 -b 4096 my_device} @*
@command{tune2fs -r0 -c10000 my_device}

This adds a journal which slows down the writes but makes filesystem checks
faster.

@c cincvdoc_node_number_287
@node Improving Zoran video
@subsection Improving Zoran video
@cindex Zoran video, improving

Video recorded from the ZORAN inputs is normally unaligned or not completely
encoded on the right.  This can be slightly compensated by adjusting parameters
in the driver sourcecode.

In @file{/usr/src/linux/drivers/media/video/zr36067.c} the structures defined
near line 623 affect alignment.  At least for NTSC, the 2.4.20 version of the
driver could be improved by changing
@verbatim
    static struct tvnorm f60ccir601 = { 858, 720, 57, 788, 525, 480, 16 };
to
    static struct tvnorm f60ccir601 = { 858, 720, 57, 788, 525, 480, 17 };
@end verbatim

In @file{/usr/src/linux/drivers/media/video/bt819.c} more structures near line
76 affect alignment and encoding. @*
For NTSC
@verbatim
    {858 - 24, 2, 523, 1, 0x00f8, 0x0000},
could be changed to
    {868 - 24, 2, 523, 1, 0x00f8, 0x0000},
@end verbatim
Adjusting these parameters may or may not move your picture closer to the
center.  More of the time, they will cause the driver to lock up before capturing
the first frame.

@b{New in 2.6.5:} @*
In the 2.6 kernels, the video subsystem was rewritten again from scratch.  To
adjust the Zoran parameters go to @file{drivers/media/video/zoran_card.c} and
look for a group of lines like
@verbatim
    static struct tvnorm f50sqpixel = { 944, 768, 83, 880, 625, 576, 16 };
    static struct tvnorm f60sqpixel = { 780, 640, 51, 716, 525, 480, 12 };
    static struct tvnorm f50ccir601 = { 864, 720, 75, 804, 625, 576, 18 };
    static struct tvnorm f60ccir601 = { 858, 720, 57, 788, 525, 480, 16 };

    static struct tvnorm f50ccir601_lml33 = { 864, 720, 75+34, 804, 625, 576, 18 };
    static struct tvnorm f60ccir601_lml33 = { 858, 720, 57+34, 788, 525, 480, 16 };

    /* The DC10 (57/16/50) uses VActive as HSync, so HStart must be 0 */
    static struct tvnorm f50sqpixel_dc10 = { 944, 768, 0, 880, 625, 576, 0 };
    static struct tvnorm f60sqpixel_dc10 = { 780, 640, 0, 716, 525, 480, 12 };

    /* FIXME: I cannot swap U and V in saa7114, so i do one
     * pixel left shift in zoran (75 -> 74)
     * (Maxim Yevtyushkin <max@linuxmedialabs.com>) */
    static struct tvnorm f50ccir601_lm33r10 = { 864, 720, 74+54, 804, 625, 576, 18 };
    static struct tvnorm f60ccir601_lm33r10 = { 858, 720, 56+54, 788, 525, 480, 16 };
@end verbatim

These seem to control the image position.  At least for the LML33 the following
definition for @b{f60ccir601_lml33} does the trick.
@verbatim
static struct tvnorm f60ccir601_lml33 = { 858, 720, 67+34, 788, 525, 480, 13 };
@end verbatim

@c cincvdoc_node_number_288
@node Translating Cinelerra
@section Translating Cinelerra
@cindex Translating Cinelerra

@menu
* Available localizations::
* Updating an existing translation::
* Creating a new translation::
@end menu

This information is needed if you wish to partipate in translating Cinelerra.
@xref{Environment variables}, for running Cinelerra in your own language.

@c cincvdoc_node_number_289
@node Available localizations
@subsection Available localizations
@cindex Localizations, available

There are some existing localizations for cinelerra:
@itemize @bullet
@item @b{DE} - German
@item @b{ES} - Spanish
@item @b{EU} - Basque
@item @b{FR} - French
@item @b{IT} - Italian
@item @b{PT_BR} - Brazilian Portuguese
@item @b{SL} - Slovenian
@end itemize

@c cincvdoc_node_number_290
@node Updating an existing translation
@subsection Updating an existing translation
@cindex Updating an existing translation

To generate an updated @file{*.po} file with the newer strings of Cinelerra
source code not yet present in the @file{.po} file, run after
@command{./configure}: @*
@command{cd po && make}

Then, edit the @file{.po} file located in @file{po/} directory of your target
language and submit the diff file to the Cinelerra-CV team.

@c cincvdoc_node_number_291
@node Creating a new translation
@subsection Creating a new translation
@cindex Creating a new translation

To create a new translation, run after @command{./configure}: @*
@command{cd po && make}

Then, edit the @file{cinelerra.pot} file located in @file{po/} and add the
appropriate translated strings.  Rename the file to @file{(lang_prefix).po} and
add the language prefix to @file{po/LINGUAS}.  Finally, submit the diff file to
the cinelerra-CV team.

@c cincvdoc_node_number_292
@node Panning and zooming still images
@section Panning and zooming still images
@cindex Panning and zooming still images
@cindex Still images, panning and zooming

Cinelerra's powerful keyframe features allow you to use pan and zoom
effects on still pictures.
@enumerate 1
@item Load and create a clip from a still image as described above.  Make the
clip 10 seconds long.
@item Activate the @b{automatic generation of keyframes}
@item Using the @b{transport controls}, go to the beginning of the clip
@item Using the @b{compositing camera control} set the clip's initial position
@item Using the @b{transport controls}, move forward a couple of seconds on the
clip
@item Dragging on the @b{compositing camera} move the camera center to a new
position further
@item Now, rewind to the beginning of the clip and play it.
@end enumerate
You can see that the camera smoothly flows from keyframe point to next keyframe
point, as Cinelerra automatically adjusts the camera movement in straight lines
from point to point.

@c cincvdoc_node_number_293
@node Troubleshooting
@chapter Troubleshooting
@cindex Troubleshooting

@menu
* Reporting bugs::
* Buz driver crashes::
* Dragging in and out points does not work::
* Locking up when loading files::
* Synchronization lost while recording::
* Applying gamma followed by blur does not work::
* Copy/Paste of track selections does not work in the timeline::
* Cinelerra often crashes::
* Theme Blond not found error::
@end menu

@c cincvdoc_node_number_294
@node Reporting bugs
@section Reporting bugs
@cindex Reporting bugs
@cindex Bugs, reporting

When you notice a bug, the first thing to do is to go to
@uref{http://bugs.cinelerra.org} and check if it has not been already reported.
If there is no bug report for the bug you noticed, you can file a bug report.
Open an account on @uref{http://bugs.cinelerra.org} if you do not have one.
Then, file the bug report, including the following information:

@itemize @bullet
@item Revision number of Cinelerra CV@.  Example: r959

@item Distribution name and version.  Example: Debian SID

@item Steps to replicate the bug.  That is very important since it really helps
people trying to fix bugs.  Example:
@enumerate 1
@item launch cinelerra
@item open the recording window
@item click on OK
@item Cinelerra crashes
@end enumerate

@item When Cinelerra CV crashes, a debugger output is welcome.  Run: @*
@command{gdb cinelerra} @*
@command{run} @*
(You trigger the bug and Cinelerra CV crashes) @*
@command{thread apply all bt} @*
Then copy all of the information displayed into your bug report.
@end itemize

Do not hesitate to attach any file which you think could be useful, such as a
screenshot for example.  The gdb output is more useful when Cinelerra is
compiled with debugging symbols.  @xref{Compiling with debugging symbols}, for
compilation instructions.

Moreover, if the bug you noticed concerns a problem loading a specific file
into Cinelerra-CV, uploading a small sample of such a file on the internet is
appreciated.  That would allow people fixing bugs to load that file themselves
in Cinelerra and look at what happens.

@c cincvdoc_node_number_295
@node Buz driver crashes
@section Buz driver crashes
@cindex Buz, driver crashes

First, Zoran capture boards must be accessed using the @b{Buz} video driver in
@b{Preferences->Recording} and @b{Preferences->Playback}.  Some performance
tweeks are available in another section.  @xref{Improving performance}.

Once tweeked, the Buz driver seems to crash if the number of recording buffers
is too high.  Make sure @b{Preferences->Recording->Frames to buffer in device}
is below 10.

@c cincvdoc_node_number_296
@node Dragging in and out points does not work
@section Dragging in and out points does not work
@cindex Dragging in and out points does not work
@cindex In/out points, dragging does not work

Sometimes there will be two edits really close together.  The point selected
for dragging may be next to the indended edit if those edits are too small to see at
the current zoom level.  Zoom in horizontally.

@c cincvdoc_node_number_297
@node Locking up when loading files
@section Locking up when loading files
@cindex Locking up when loading files
@cindex Files, locking up when loading

The most common reason loading files locks up Cinelerra is because the codec is not
supported.  Another reason is because Cinelerra is building picons for the
Resources window.  If you load a large number of images, it needs to decompress
every single image to build a picon.  Go into
@b{settings->preferences->interface} and disable @b{Use thumbnails in resource
window} to skip this process.

@c cincvdoc_node_number_298
@node Synchronization lost while recording
@section Synchronization lost while recording
@cindex Synchronization lost while recording
@cindex Recording, synchronization

If the rate at which frames are captured during recording is much lower than the framerate of the
source, the video will accumulate in the recording buffers over time and the
audio and video will become well out of sync.  Decrease the @b{number of frames to
buffer in the device} in @b{preferences->recording} so the excess frames are
dropped instead of buffered.

@c cincvdoc_node_number_299
@node Applying gamma followed by blur does not work
@section Applying gamma followed by blur does not work

The gamma effect uses the pow function while the blur effect uses a number of
exp functions in the math library.  For some reason, using the pow function
breaks later calls to the exp functions in the math library.  You need to apply
gamma after blur to get it to work.

@c cincvdoc_node_number_300
@node Copy/Paste of track selections does not work in the timeline
@section Copy/Paste of track selections does not work in the timeline
@cindex Copy/Paste of track selections does not work in the timeline

If you are running the KDE Klipper application, either disable it, or
right-click its taskbar icon, select @b{Configure Klipper} and ensure
@b{Prevent empty clipboard} is not selected.

@c cincvdoc_node_number_301
@node Cinelerra often crashes
@section Cinelerra often crashes
@cindex Crashes

Do a clean install.  Be sure that you do not have libraries from previous
installations.  Delete your @file{$HOME/.bcast/} directory too. @*
@command{rm -f /usr/local/lib/libguicast*} @*
@command{rm -f /usr/lib/libguicast*} @*
@command{rm -f /usr/local/lib/libquicktimehv*} @*
@command{rm -f /usr/lib/libquicktimehv*} @*
@command{rm -f /usr/local/lib/libmpeg3hv*} @*
@command{rm -f /usr/lib/libmpeg3hv*}

@c cincvdoc_node_number_302
@node Theme Blond not found error
@section Theme Blond not found error
@cindex Theme Blond not found error

If the following error message appears: @command{Aborted, MWindow::init_theme:
Theme Blond not found}, then:
@itemize @bullet
@item You should check for the file @file{defaulttheme.*} in
@file{/usr/lib/cinelerra} or @file{/usr/local/lib/cinelerra}.  If it does not
exist, you need to install the plugins again.
@item Try to delete the @file{$HOME/.bcast/} directory
@item Look into @file{$HOME/.bcast/Cinelerra_rc} and find THEME, it should be
=> THEME Blond
@end itemize

@c cincvdoc_node_number_303
@node Plugin authoring
@chapter Plugin authoring
@cindex Plugin authoring

The plugin API in Cinelerra dates back to 1997, before the LADSPA and before
VST became popular.  It is fundamentally the same as it was in 1997, with minor
modifications to handle keyframes and GUI feedback.  The GUI is not abstracted
from the programmer.  This allows the programmer to use whatever toolkit they
want and allows more flexibility in appearance but it costs more.

There are several types of plugins, each with a common procedure of
implementation and specific changes for that particular type.  The easiest way
to implement a plugin is to take the simplest existing one out of the group and
rename the symbols.

@menu
* Introducing the pull method:: The current paradigm for plugin writing
* Common plugin functions:: What all effects have to do.
* Realtime plugins:: What realtime effects have to do.
* Nonrealtime plugins:: What rendered effects have to do.
* Audio plugins:: What audio effects have to do.
* Video plugins:: What video effects have to do.
* Transition plugins:: What transitions have to do.
* Plugin GUI's which update during playback:: How to use currently playing data to draw the GUI.
* Using OpenGL:: How to use hardware to speed up operations.
* Plugin queries:: How plugins get information about the data to be processed.
@end menu

@c cincvdoc_node_number_304
@node Introducing the pull method
@section Introducing the pull method
@cindex Pull method, introducing the

Originally plugins were designed with the push method.  The push method is
intuitive and simple.  A source pushes data to a plugin, the plugin does math
operations on it, and the plugin pushes it to a destination.  For 6 years this
was the way all realtime plugins were driven internally but it did not allow you
to reduce the rate of playback in realtime.  While plugins can still be
designed as if they are pushing data, this is not the way they are processed
internally anymore.

The latest evolution in Cinelerra's plugin design is the pull method.  The
rendering pipeline starts at the final output and the final steps in the
rendering pipeline are reading the data from disk.  Every step in the rendering
chain involves requesting data from the previous step.  When the rendering
pipleline eventually requests data from a plugin chain, each plugin requests
data from the plugin before it.

This is less intuitive than the push method but is much more powerful.
Realtime plugins written using the pull method can not only change the rate data is
presented to the viewer but also the direction of playback.  The pull method also allows
plugins to take in data at a higher rate than they send it out.

To get the power of rate independence, the pull method requires plugins to know
more about the data than they needed to under the push method.  Plugins need to
know what rate the project is at, what rate their output is supposed to be at
and what rate their input is supposed to be at.  These different data rates
have to be correlated for a plugin to configure itself properly.

Keyframes for a plugin are stored relative to the project frame rate.  Queries
from a plugin for the current playback position are given relative to the
project frame rate.  If the plugin's output was requested to be at twice the
project frame rate, the positions need to be converted to the project rate for
keyframes to match up.  Two classes of data rates were created to handle this
problem.

Rate conversions are done in terms of the @b{project rate} and the @b{requested
rate}.  The project rate is identical for all plugins.  It is determined by the
@b{settings->format} window.  The requested rate is determined by the
downstream plugin requesting data from the current plugin.  The requested rate
is arbitrary.  Exactly how to use these rates is described below.

@c cincvdoc_node_number_305
@node Common plugin functions
@section Common plugin functions
@cindex Common plugin functions

All plugins inherit from a derivative of PluginClient.  This PluginClient
derivative implements most of the required methods in PluginClient, but users
must still define methods for PluginClient.  The most commonly used methods are
predefined in macros to reduce the typing yet still allow flexibility.

The files they include depend on the plugin type.  Audio plugins include
@file{pluginaclient.h} and video plugins include @file{pluginvclient.h}.  They
inherit @b{PluginAClient} and @b{PluginVClient} respectively.

Cinelerra instantiates all plugins at least twice when they are used in a
movie.  One instance is the GUI@.  The other instance is the signal processor.
User input, through a complicated sequence, is propogated from the GUI instance
to the signal processor instance.  If the signal processor wants to alter the
GUI, it propogates data back to the GUI instance.  There are utility functions
for doing all this.

All plugins define at least three objects:

@itemize @bullet
@item @b{Processing object} @*
Contains pointers to all the other objects and performs the signal processing.
This object contains a number of queries to identify itself and is the object
you register to register the plugin.
@item @b{User interface object} @*
This is defined according to the programmer's discretion.  It can either use
Cinelerra's toolkit or another toolkit.  It shows data on the screen and
collects parameters from the user. @*
Using Cinelerra's toolkit, the only user interface object a developer needs to
worry about is the Window.  The window has pointers to a number of widgets, a
few initialization methods, and a back pointer to the plugin's processing
object.  The documentation refers to the usage of Cinelerra's toolkit. @*
Depending on the user interface toolkit, a user interface thread may be created
to run the user interface asynchronous of everything else.  Synchronizing the
user interface to changes in the plugin's configuration is the most complicated
aspect of the plugin, so the user interface thread and object are heavily
supported by macros if you use Cinelerra's toolkit.
@item @b{Configuration object} @*
This stores the user parameters and always needs interpolation, copying, and
comparison functions.  Macros for the plugin client automatically call
configuration methods to interpolate keyframes.
@end itemize

@menu
* The processing object::
* The configuration object::
* The user interface object::
@end menu

@c cincvdoc_node_number_306
@node The processing object
@subsection The processing object
@cindex The processing object

Load up a simple plugin like gain to see what this object looks like.  The
processing object should inherit from the intended PluginClient derivative.
Its constructor should take a PluginServer argument.
@verbatim
MyPlugin(PluginServer *server);
@end verbatim

In the implementation, the plugin must contain a registration line with the
name of the processing object like
@verbatim
REGISTER_PLUGIN(MyPlugin)
@end verbatim

The constructor should contain
@verbatim
PLUGIN_CONSTRUCTOR_MACRO
@end verbatim
to initialize the most common variables.

The processing object should have a destructor containing
@verbatim
PLUGIN_DESTRUCTOR_MACRO
@end verbatim
to delete the most common variables.

Another function which is useful but not mandatory is
@verbatim
int is_multichannel();
@end verbatim
It should return 1 if one instance of the plugin handles multiple tracks
simultaneously or 0 if one instance of the plugin only handles one track.  The
default is 0 if it is omitted.

Multichannel plugins in their processing function should refer to a function
called @b{PluginClient::get_total_buffers()} to determine the number of
channels.

To simplify the implementation of realtime plugins, a macro for commonly used
members has been created for the class header, taking the configuration object
and user interface thread object as arguments.  The macro definitions apply
mainly to realtime plugins and are not useful in nonrealtime plugins.
Fortunately, nonrealtime plugins are simpler.
@verbatim
PLUGIN_CLASS_MEMBERS(config_name, thread_name)
@end verbatim

The commonly used members in PLUGIN_CLASS_MEMBERS are described below.

@b{int load_configuration();} @*
Loads the configuration based on surrounding keyframes and current position. @*
The class definition for load_configuration should contain
@verbatim
LOAD_CONFIGURATION_MACRO(plugin_class, config_class)
@end verbatim
to implement the default behavior for load_configuration.  This stores whatever
the current configuration is inside the plugin's configuration object and
returns 1 if the new configuration differs from the previous configuration.
The return value of load_configuration is used by another commonly used
function, update_gui to determine if the GUI really needs to be updated. @*
The plugin's configuration object is always called @b{config} inside
PLUGIN_CLASS_MEMBERS@.

@b{VFrame* new_picon();} @*
Creates a picon for display in the resource window.  Use
@verbatim
#include "picon_png.h"
NEW_PICON_MACRO(plugin_class)
@end verbatim
to implement new_picon.  In addition, the user should create a
@file{picon_png.h} header file from a PNG image using @command{pngtoh}.
@command{pngtoh} is compiled in the @file{guicast/ARCH} directory. @*
The source PNG image should be called @file{picon.png} and can be any format
supported by PNG@.

@b{char* plugin_title();} @*
Returns a text string identifying the plugin in the resource window.  The
string has to be unique.

@b{void update_gui();} @*
Should first load the configuration, test for a return of 1, and then redraw
the GUI with the new parameters.  All the plugins using GuiCast have a format
like
@verbatim
    void MyPlugin::update_gui()
    {
        if(thread)
        {
        if(load_configuration())
        {
            thread->window->lock_window();
            // update widgets here
            thread->window->unlock_window();
        }
        }
    }
@end verbatim
to handle concurrency and conditions of no GUI@.

@b{int show_gui();} @*
Instantiate the GUI and switch the plugin to GUI mode.  This is implemented
with
@verbatim
SHOW_GUI_MACRO(plugin_class, thread_class)
@end verbatim

@b{int set_string();} @*
Changes the title of the GUI window to a certain string.  This is implemented
with
@verbatim
SET_STRING_MACRO(plugin_class)
@end verbatim

@b{void raise_window();} @*
Raises the GUI window to the top of the stack.  This is implemented with
@verbatim
RAISE_WINDOW_MACRO(plugin_class)
@end verbatim

Important functions that the processing object must define are the functions which
load and save configuration data from keyframes.  These functions are called by
the macros so all you need to worry about is accessing the keyframe data.
@verbatim
void save_data(KeyFrame *keyframe);
void read_data(KeyFrame *keyframe);
@end verbatim

The read data functions are only used in realtime plugins.  The read data
functions translate the plugin configuration between the KeyFrame argument and
the configuration object for the plugin.  The keyframes are stored on the
timeline and can change for every project.

Use an object called @b{FileXML} to do the translation and some specific
commands to get the data out of the KeyFrame argument.  See any existing plugin
to see the usage of KeyFrame and FileXML.
@verbatim
int load_defaults();
int save_defaults();
@end verbatim

The load defaults functions are used in realtime and non-realtime plugins.  The
load defaults functions translate the plugin configuration between a BC_Hash
object and the plugin's configuration.  The BC_Hash object stores
configurations in a discrete file on disk for each plugin but does not isolate
different configurations for different projects.

The function overriding @b{load_defaults} also needs to create the BC_Hash
object.  See any existing plugin to see the usage of BC_Hash.

Other standard members may be defined in the processing object, depending on
the plugin type.

@c cincvdoc_node_number_307
@node The configuration object
@subsection The configuration object
@cindex Configuration object

The configuration object is critical for GUI updates, signal processing, and
default settings in realtime plugins.  Be aware it is not used in nonrealtime
plugins.  The configuration object inherits from nothing and has no
dependancies.  It is merely a class containing three functions and variables
specific to the plugin's parameters.

Usually the configuration object starts with the name of the plugin followed by
Config.
@verbatim
    class MyPluginConfig
    {
    public:
        MyPluginConfig();
@end verbatim

Following the name of the configuration class, we put in three required
functions and the configuration variables.
@verbatim
        int equivalent(MyPluginConfig &that);
        void copy_from(MyPluginConfig &that);
        void interpolate(MyPluginConfig &prev,
        MyPluginConfig &next,
        int64_t prev_position,
        int64_t next_position,
        int64_t current_position);
        float parameter1;
        float parameter2;
        int parameter3;
    };
@end verbatim

Now you must define the three functions.  @b{Equivalent} is called by
LOAD_CONFIGURATION_MACRO to determine if the local configuration parameters are
identical to the configuration parameters in the arguement.  If equivalent
returns 0, the LOAD_CONFIGURATION_MACRO causes the GUI to redraw.  If
equivalent returns 1, the LOAD_CONFIGURATION_MACRO does not redraw the GUI@.

Then there is @b{copy_from} which transfers the configuration values from the
argument to the local variables.  This is once again used in
LOAD_CONFIGURATION_MACRO to store configurations in temporaries.  Once
LOAD_CONFIGURATION_MACRO has replicated the configuration, it loads a second
configuration.  Then it interpolates the two configurations to get the current
configuration.  The interpolation function performs the interpolation and
stores the result in the local variables.

Normally the interpolate function calculates a previous and next fraction,
using the arguments.
@verbatim
    void MyPluginConfig::interpolate(MyPluginConfig &prev,
        MyPluginConfig &next,
        int64_t prev_position,
        int64_t next_position,
        int64_t current_position
    {
        double next_scale =
        (double)(current_position - prev_position)
        / (next_position - prev_position);
        double prev_scale =
        (double)(next_position - current_position) /
        (next_position - prev_position);
@end verbatim

Then the fractions are applied to the previous and next configuration variables
to yield the current values.
@verbatim
        this->parameter1 =
        (float)(prev.parameter1 * prev_scale
        + next.parameter1 * next_scale);
        this->parameter2 =
        (float)(prev.parameter2 * prev_scale
        + next.parameter2 * next_scale);
        this->parameter3 =
        (int)(prev.parameter3 * prev_scale
        + next.parameter3 * next_scale);
    }
@end verbatim

Alternatively you can copy the values from the previous configuration argument
if no interpolation is desired.

This usage of the configuration object is the same in audio and video plugins.
In video playback, the interpolation function is called for every frame,
yielding smooth interpolation.  In audio playback, the interpolation function
is called only once for every console fragment and once every time the
insertion point moves.  This is good enough for updating the GUI while
selecting regions on the timeline but it may not be accurate enough for really
smooth rendering of the effect.

For really smooth rendering of audio, you can still use load_configuration when
updating the GUI@.  For process_buffer; however, ignore load_configuration and
write your own interpolation routine which loads all the keyframes in a console
fragment and interpolates every sample.  This would be really slow and hard to
debug, yielding improvement which may not be audible.  Then of course, every
country has its own wierdos.

An easier way to get smoother interpolation is to reduce the console fragment
to 1 sample.  This would have to be rendered and played back with the console
fragment back over 2048 of course.  The GNU/Linux sound drivers can not play
fragments of 1 sample.

@c cincvdoc_node_number_308
@node The user interface object
@subsection The user interface object
@cindex User interface object

The user interface object at the very least consists of a pointer to a window
and pointers to all the widgets in the window.  Using Cinelerra's toolkit, it
consists of a @b{BCWindow} derivative and a @b{Thread} derivative.  The Thread
derivative is declared in the plugin header using
@verbatim
PLUGIN_THREAD_HEADER(plugin_class, thread_class, window_class)
@end verbatim

Then it is defined using
@verbatim
PLUGIN_THREAD_OBJECT(plugin_class, thread_class, window_class)
@end verbatim

This, in combination with the SHOW_GUI macro does all the work in instantiating
the Window.  This two-class system is used in realtime plugins but not in
nonrealtime plugins.  Nonrealtime plugins create and destroy their GUI in their
@b{get_parameters} function and there is no need for a Thread.

Now the window class must be declared in the plugin header.  It is easiest to
implement the window by copying an existing plugin and renaming the symbols.
The following is an outline of what happens.  The plugin header must declare
the window's constructor using the appropriate arguments.
@verbatim
    #include "guicast.h"
    class MyPluginWindow : public BC_Window
    {
    public:
        MyPluginWindow(MyPluginMain *plugin, int x, int y);
@end verbatim

This becomes a window on the screen, positioned at x and y.

It needs two methods
@verbatim
int create_objects();
int close_event();
@end verbatim
and a back pointer to the plugin
@verbatim
MyPlugin *plugin;
@end verbatim

The constructor's definition should contain extents and flags causing the
window to be hidden when first created.  The create_objects member puts widgets
in the window according to GuiCast's syntax.  A pointer to each widget which
you want to synchronize to a configuration parameter is stored in the window
class.  These are updated in the @b{update_gui} function you earlier defined
for the plugin.  The widgets are usually derivatives of a GuiCast widget and
they override functions in GuiCast to handle events.  Finally create_objects
calls
@verbatim
show_window();
flush();
@end verbatim
to make the window appear all at once.

The close_event member should be implemented using
@verbatim
WINDOW_CLOSE_EVENT(window_class)
@end verbatim

Every widget in the GUI needs to detect when its value changes.  In GuiCast the
@b{handle_event} method is called whenever the value changes.  In
@b{handle_event}, the widget then needs to call
@b{plugin->send_configure_change()} to propogate the change to any copies of
the plugin which are processing data.

@c cincvdoc_node_number_309
@node Realtime plugins
@section Realtime plugins
@cindex Realtime plugins

Realtime plugins should use PLUGIN_CLASS_MEMBERS to define the basic set of
members in their headers.  All realtime plugins must define an @b{int
is_realtime()}

member returning 1.  This causes a number of methods to be called during live
playback and the plugin to be usable on the timeline.

Realtime plugins must override a member called @b{process_buffer}

This function takes different arguments depending on if the plugin handles
video or audio.  See an existing plugin to find out which usage applies.

The main features of the process_buffer function are a buffer to store the
output, the starting position of the output, and the requested output rate.
For audio, there is also a size argument for the number of samples.

The starting position of the output buffer is the lowest numbered sample on the
timeline if playback is forward and the highest numbered sample on the timeline
if playback is reverse.  The direction of playback is determined by one of the
plugin queries described below.

The position and size arguments are all relative to the frame rate and sample
rate passed to process_buffer.  This is the requested data rate and may not be
the same as the project data rate.

The process_realtime function should start by calling @b{load_configuration}.
The LOAD_CONFIGURATION_MACRO returns 1 if the configuration changed.

After determining the plugin's configuration, input media has to be loaded for
processing.  Call:
@verbatim
    read_frame(VFrame *buffer,
        int channel,
        int64_t start_position,
        double frame_rate)
or
    read_samples(double *buffer,
        int channel,
        int sample_rate,
        int64_t start_position,
        int64_t len)
@end verbatim

to request input data from the object which comes before this plugin.  The read
function needs a buffer to store the input data in.  This can either be a
temporary you create in the plugin or the output buffer supplied to
process_buffer if you do not need a temporary.

It also needs a set of position arguments to determine when you want to read
the data from.  The start position, rate, and len passed to a read function
need not be the same as the values recieved by the process_buffer function.
This way plugins can read data at a different rate than they output data.

The channel argument is only meaningful if this is a multichannel plugin.  They
need to read data for each track in the get_total_buffers() value and process
all the tracks.  Single channel plugins should pass 0 for channel.

Additional members are implemented to maintain configuration in realtime
plugins.  Some of these are also needed in nonrealtime plugins.

@itemize @bullet
@item @b{void read_data(KeyFrame *keyframe);} @*
Loads data from a keyframe into the plugin's configuration.  Inside the
keyframe is an XML string.  It is most easily parsed by creating a @b{FileXML}
object.  See an existing plugin to see how the read_data function is
implemented. @*
Read data loads data out of the XML object and stores values in the plugin's
configuration object.  Since configuration objects vary from plugin to plugin,
these functions can not be automated.

@item @b{void save_data(KeyFrame *keyframe);} @*
Saves data from the plugin's configuration to a keyframe.  Inside the keyframe
you will put an XML string which is normally created by a FileXML object.  See an
existing plugin to see how the save_data function is implemented. @*
Save data saves data from the plugin's configuration object into the XML
object.

@item @b{int load_defaults();} @*
Another way the plugin gets parameters is from a defaults file.  The load and
save defaults routines use a BC_Hash object to parse the defaults file.  The
defaults object is created in @b{load_defaults} and destroyed in the plugin's
destructor.  See an existing plugin to see how the BC_Hash object is used.

@item @b{int save_defaults();} @*
Saves the configuration in the defaults object.
@end itemize

@c cincvdoc_node_number_310
@node Nonrealtime plugins
@section Nonrealtime plugins
@cindex Nonrealtime plugins

Some references for non-realtime plugins are @b{Normalize} for audio and
@b{Reframe} for video.

Like realtime plugins, @b{load_defaults} and @b{save_defaults} must be
implemented.  In nonrealtime plugins, these are not just used for default
parameters but to transfer values from the user interface to the signal
processor.  There does not need to be a configuration class in nonrealtime
plugins.

Unlike realtime plugins, the LOAD_CONFIGURATION_MACRO can not be used in the
plugin header.  Instead, the following methods must be defined.

The nonrealtime plugin should contain a pointer to a defaults object.
@verbatim
BC_Hash *defaults;
@end verbatim
It should also have a pointer to a MainProgressBar.
@verbatim
MainProgressBar *progress;
@end verbatim

The progress pointer allows nonrealtime plugins to display their progress in
Cinelerra's main window.

The constructor for a nonrealtime plugin cannot use PLUGIN_CONSTRUCTOR_MACRO
but must call @b{load_defaults} directly.

The destructor, likewise, must call @b{save_defaults} and @b{delete defaults}
directly instead of PLUGIN_DESTRUCTOR_MACRO@.

@itemize @bullet
@item @b{VFrame* new_picon();} @*
@b{char* plugin_title();} @*
The usage of these is the same as realtime plugins.

@item @b{int is_realtime();} @*
This function must return 0 to indicate a nonrealtime plugin.

@item @b{int get_parameters();} @*
Here, the user should create a GUI, wait for the user to hit an OK button or a
cancel button, and store the parameters in plugin variables.  This routine must
return 0 for success and 1 for failure.  This way the user can cancel the
effect from the GUI@. @*
Unlike the realtime plugin, this GUI need not run asynchronously of the plugin.
It should block the get_parameters function until the user selects OK or
Cancel.

@item @b{int load_defaults();} @*
This should create a defaults object and load parameters from the defaults
object into plugin variables.

@item @b{int save_defaults();} @*
This should save plugin variables to the defaults object.

@item @b{int start_loop();} @*
If @b{get_parameters} returned 0 for success, this is called once to give the
plugin a chance to initialize processing.  The plugin should instantiate the
progress object with a line like @*
@code{progress = start_progress("MyPlugin progress...",} @*
@code{PluginClient::get_total_len());} @*
The usage of @b{start_progress} depends on whether the plugin is multichannel
or single channel.  If it is multichannel you always call start_progress.  If
it is single channel, you first need to know whether the progress bar has
already started in another instance of the plugin. @*
If @b{PluginClient::interactive} is 1, you need to start the progress bar.  If
it is 0, the progress bar has already been started. @*
The PluginClient defines @b{get_total_len()} and @b{get_source_start()} to
describe the timeline range to be processed.  The units are either samples or
frames and in the project rate.

@item @b{int process_loop} @*
This is called repeatedly until the timeline range is processed.  It has either
a samples or frames buffer for output and a reference to write_length to store
the number of samples processed.  If this is an audio plugin, the user needs to
call @b{get_buffer_size()} to know how many samples the output buffer can hold.
@*
The plugin must use @b{read_samples} or @b{read_frame} to read the input.
These functions are a bit different for a non realtime plugin than they are for
a realtime plugin. @*
They take a buffer and a position relative to the start of the timeline, in the
timeline's rate.  Then you must process it and put the output in the buffer
argument to process_loop.  write_length should contain the number of samples
generated if it is audio. @*
Finally, process_loop must test @b{PluginClient::interactive} and update the
progress bar if it is 1. @*
@code{progress->update(total_written);} @*
returns 1 or 0 if the progress bar was cancelled.  If it is 1, process_loop
should return 1 to indicate a cancellation.  In addition to progress bar
cancellation, @b{process_loop} should return 1 when the entire timeline range
is processed.

@item @b{int stop_loop();} @*
This is called after process_loop processes its last buffer. @*
If PluginClient::is_interactive is 1, this should call @b{stop_progress} in the
progress bar pointer and delete the pointer.  Then it should delete any objects
it created for processing in @b{start_loop}.
@end itemize

@c cincvdoc_node_number_311
@node Audio plugins
@section Audio plugins
@cindex Audio plugins

The simplest audio plugin is Gain.  The processing object should include
@file{pluginaclient.h} and inherit from @b{PluginAClient}.  Realtime audio
plugins need to define
@verbatim
    int process_buffer(int64_t size,
        double **buffer,
        int64_t start_position,
        int sample_rate);
if it is multichannel or
    int process_buffer(int64_t size,
        double *buffer,
        int64_t start_position,
        int sample_rate);
@end verbatim
if it is single channel.  These should return 0 on success and 1 on failure.
In the future, the return value may abort failed rendering.

The processing function needs to request input samples with
@verbatim
    int read_samples(double *buffer,
        int channel,
        int sample_rate,
        int64_t start_position,
        int64_t len);
@end verbatim
It always returns 0.  The user may specify any desired sample rate and start
position.

Nonrealtime audio plugins need to define
@verbatim
int process_loop(double *buffer, int64_t &write_length);
for single channel or
int process_loop(double **buffers, int64_t &write_length);
@end verbatim
for multi channel.  Non realtime plugins use a different set of read_samples
functions to request input data.  These are fixed to the project sample rate.

@c cincvdoc_node_number_312
@node Video plugins
@section Video plugins
@cindex Video plugins

The simplest video plugin is Flip.  The processing object should include
@file{pluginvclient.h} and inherit from @b{PluginVClient}.  Realtime video
plugins need to define
@verbatim
    int process_buffer(VFrame **frame,
        int64_t start_position,
        double frame_rate);
@end verbatim
if it is multichannel or
@verbatim
    int process_buffer(VFrame *frame,
        int64_t start_position,
        double frame_rate);
@end verbatim
if it is single channel.

The nonrealtime video plugins need to define
@verbatim
int process_loop(VFrame *buffer);
for single channel or
int process_loop(VFrame **buffers);
@end verbatim
for multi channel.  The amount of frames generated in a single process_loop is
always assumed to be 1, hence the lack of a write_length argument.  Returning 0
causes the rendering to continue.  Returning 1 causes the rendering to abort.

A set of read_frame functions exist for requesting input frames in non-realtime
video plugins.  These are fixed to the project frame rate.

@c cincvdoc_node_number_313
@node Transition plugins
@section Transition plugins
@cindex Transition plugins

The simplest video transition is @b{wipe} and the simplest audio transition is
@b{crossfade}.  These use a subset of the default class members of realtime
plugins, but so far no analogue to PLUGIN_CLASS_MEMBERS has been done for
transitions.

The processing object for audio transitions still inherits from PluginAClient
and for video transitions it still inherits from PluginVClient.

Transitions may or may not have a GUI@.  If they have a GUI, they must also
manage a thread like realtime plugins.  Do this with the same
PLUGIN_THREAD_OBJECT and PLUGIN_THREAD_HEADER macros as realtime plugins.
Since there is only one keyframe in a transition, you do not need to worry about
updating the GUI from the processing object like you do in a realtime plugin.

If the transition has a GUI, you can use PLUGIN_CONSTRUCTOR_MACRO and
PLUGIN_DESTRUCTOR_MACRO to initialize the processing object.  You will also need
a BC_Hash object and a Thread object for these macros.

Since the GUI is optional, overwrite a function called @b{uses_gui()} to
signifiy whether or not the transition has a GUI@.  Return 1 if it does and 0 if
it does not.

Transitions need a @b{load_defaults} and @b{save_defaults} function so the
first time they are dropped on the timeline they will have useful settings.

A @b{read_data} and @b{save_data} function takes over after insertion to access
data specific to each instance of the transition.

The most important difference between transitions and realtime plugins is the
addition of an @b{is_transition} method to the processing object.
@b{is_transition} should return 1 to signify the plugin is a transition.

Transitions process data in a @b{process_realtime} function.
@verbatim
    int process_realtime(VFrame *input,
        VFrame *output);
    int process_realtime(int64_t size,
        double *input_ptr,
        double *output_ptr);
@end verbatim
The input argument to process_realtime is the data for the next edit.  The
output argument to process_realtime is the data for the previous edit.

Routines exist for determining where you are relative to the transition's start
and end.

@itemize @bullet
@item
@b{PluginClient::get_source_position()} - returns the current position since
the start of the transition of the lowest sample in the buffers.

@item
@b{PluginClient::get_total_len()} - returns the integer length of the
transition.  The units are either samples or frames, in the data rate requested
by the first plugin.
@end itemize

Users should divide the source position by total length to get the fraction of
the transition the current @b{process_realtime} function is at.

Transitions run in the data rate requested by the first plugin in the track.
This may be different than the project data rate.  Since process_realtime lacks
a rate argument, use @b{get_framerate()} or @b{get_samplerate} to get the
requested rate.

@c cincvdoc_node_number_314
@node Plugin GUI's which update during playback
@section Plugin GUI's which update during playback
@cindex Plugin GUI's which update during playback

Effects like @b{Histogram} and @b{VideoScope} need to update the GUI during
playback to display information about the signal.  This is achieved with the
@b{send_render_gui} and @b{render_gui} methods.  Normally in process_buffer,
when the processing object wants to update the GUI it should call
@b{send_render_gui}.  This should only be called in process_buffer.
Send_render_gui goes through a search and eventually calls @b{render_gui} in
the GUI instance of the plugin.

Render_gui should have a sequence like
@verbatim
    void MyPlugin::render_gui(void *data)
    {
        if(thread)
        {
        thread->window->lock_window();
        // update GUI here
        thread->window->unlock_window();
        }
    }
@end verbatim

Send_render_gui and render_gui use one argument, a void pointer to transfer
information from the processing object to the GUI@.  The user should typecast
this pointer into something useful.

@c cincvdoc_node_number_315
@node Plugin queries
@section Plugin queries
@cindex Plugin queries

There are several useful queries in PluginClient which can be accessed from the
processing object.  Some of them have different meaning in realtime and
non-realtime mode.  They all give information about the operating system or the
project which can be used to improve the quality of the processing.

@menu
* System queries:: Utilities for determining the system resources.
* Timing queries:: Utilities for performing time-dependant processing.
@end menu

@c cincvdoc_node_number_316
@node System queries
@subsection System queries
@cindex System queries

@itemize @bullet
@item @b{get_interpolation_type()} @*
Returns the type of interpolation the user wants for all scaling operations.
This is a macro from overlayframe.inc.  It can be applied to any call to the
@b{OverlayFrame} object.

@item @b{get_project_smp()} @*
Gives the number of CPU's on the system minus 1.  If it is a uniprocessor it is
0.  If it is a dual processor, it is 1.  This number should be used to gain
parallelism.

@item @b{get_total_buffers()} @*
Gives the number of tracks a multichannel plugin needs to process.
@end itemize

@c cincvdoc_node_number_317
@node Timing queries
@subsection Timing queries
@cindex Timing queries

There are two rates for media a realtime plugin has to be aware of: the project
rate and the requested rate.  Functions are provided for getting the project
and requested rate.  In addition, doing time dependant effects requires using
several functions which tell where you are in the effect.

@itemize @bullet
@item @b{get_project_framerate()} @*
Gives the frames per second of the video as defined by the project settings.

@item @b{get_project_samplerate()} @*
Gives the samples per second of the audio as defined by the project settings.

@item @b{get_framerate()} @*
Gives the frames per second requested by the plugin after this one.  This is
the requested frame rate and is the same as the frame_rate argument to
process_buffer.

@item @b{get_samplerate()} @*
Gives the samples per second requested by the plugin after this one.  This is
the requested sample rate and is the same as the sample_rate argument to
process_buffer.

@item @b{get_total_len()} @*
Gives the number of samples or frames in the range covered by the effect,
relative to the requested data rate.

@item @b{get_source_start()} @*
For realtime plugins it gives the lowest sample or frame in the effect range in
the requested data rate.  For nonrealtime plugins it is the start of the range
of the timeline to process.

@item @b{get_source_position()} @*
For realtime plugins it is the lowest numbered sample in the requested region
to process if playing forward and the highest numbered sample in the region if
playing backward.  For video it is the start of the frame if playing forward
and the end of the frame if playing in reverse.  The position is relative to
the start of the EDL and in the requested data rate. @*
For transitions this is always the lowest numbered sample of the region to
process relative to the start of the transition.

@item @b{get_direction()} @*
Gives the direction of the current playback operation.  This is a macro defined
in transportque.inc.  This is useful for calling read functions since the read
functions position themselves at the start or end of the region to read,
depending on the playback operation.

@item @b{local_to_edl()} @*
@b{edl_to_local()} @*
These convert between the requested data rate and the project data rate.  They
are used to convert keyframe positions into numbers which can be interpolated
at the requested data rate.  The conversion is automatically based on frame
rate or sample rate depending on the type of plugin.

@item @b{get_prev_keyframe(int64_t position, int is_local)} @*
@b{get_next_keyframe(int64_t position, int is_local)} @*
These give the nearest keyframe before or after the position given.  The macro
defined version of load_configuration automatically retrieves the right
keyframes but you may want to do this on your own. @*
The position argument can be either in the project rate or the requested rate.
Set is_local to 1 if it is in the requested rate and 0 if it is in the project
rate. @*
In each keyframe, another position value tells the keyframe's position relative
to the start of the timeline and in the project rate. @*
The only way to get smooth interpolation between keyframes is to convert the
positions in the keyframe objects to the requested rate.  Do this by using
edl_to_local on the keyframe positions.
@end itemize

@c cincvdoc_node_number_318
@node Using OpenGL
@section Using OpenGL
@cindex OpenGL, using

Realtime video plugins support OpenGL@.  Using OpenGL to do plugin routines can
speed up playback greatly since it does most of the work in hardware.
Unfortunately, every OpenGL routine needs a software counterpart for rendering,
doubling the amount of software to maintain.  Fortunately, having an OpenGL
routine means the software version does not need to be as optimized as it did
when software was the only way.

As always, the best way to design a first OpenGL plugin is to copy an existing
one and alter it.  The @b{Brightness} plugin is a simple OpenGL plugin to copy.
There are 3 main points in OpenGL rendering and 1 point for optimizing OpenGL
rendering.

@menu
* Getting OpenGL data:: Getting video data in a form usable by OpenGL
* Drawing using OpenGL:: The method of drawing video in OpenGL
* Using shaders:: Routines to simplify shader usage
* Aggregating plugins:: Combining OpenGL routines from different plugins into one.
@end menu

@c cincvdoc_node_number_319
@node Getting OpenGL data
@subsection Getting OpenGL data
@cindex OpenGL, getting data

The first problem is getting OpenGL-enabled plugins to interact with
software-only plugins.  To solve this, all the information required to do
OpenGL playback is stored in the VFrame object which is passed to
@b{process_buffer}.  To support 3D, the VFrame contains a PBuffer and a
texture, in addition to VFrame's original rows.

In OpenGL mode, VFrame has 3 states corresponding to the location of its video
data.  The opengl state is recovered by calling @b{get_opengl_state} and is set
by calling @b{set_opengl_state}.  The states are:

@itemize @bullet
@item @b{VFrame::RAM} @*
This means the video data is stored in the traditional row pointers.  It must
be loaded into a texture before being drawn using OpenGL routines.

@item @b{VFrame::TEXTURE} @*
The video data is stored in texture memory.  It is ready to be drawn using
OpenGL routines.

@item @b{VFrame::SCREEN} @*
The video data is stored in a frame buffer in the graphics card.  For plugins,
the frame buffer is always a PBuffer.  The image on the frame buffer can not be
replicated again unless it is read back into the texture and the opengl state
is reset to TEXTURE@.  The frame buffer is limited to 8 bits per channel.  If
an OpenGL effect is used in a floating point project, it only retains 8 bits.
@end itemize

In the plugin's @b{process_buffer} routine, there is normally a call to
@b{read_frame} to get data from the previous plugin in the chain.
@b{read_frame} takes a new parameter called @b{use_opengl}.

The plugin passes 1 to @b{use_opengl} if it intends to handle the data using
OpenGL@.  It passes 0 to @b{use_opengl} if it can only handle the data using
software.  The value of @b{use_opengl} is passed up the chain to ensure a
plugin which only does software only gets the data in the row pointers.  If
@b{use_opengl} is 0, the opengl state in VFrame is RAM@.

The plugin must not only know if it is software-only but if its output must be
software only.  Call @b{get_use_opengl} to determine if the output can be
handled by OpenGL@.  If @b{get_use_opengl} returns 0, the plugin must pass 0 for
@b{use_opengl} in @b{read_frame} and do its processing in software.  If
@b{get_use_opengl} is 1, the plugin can decide based on its implementation
whether to use OpenGL@.

The main problem with OpenGL is that all the gl... calls need to be run from
the same thread.  To work around this, the plugin interface has routines for
running OpenGL in a common thread.

@b{run_opengl} transfers control to the common OpenGL thread.  This is normally
called by the plugin in @b{process_buffer} after it calls @b{read_frame} and
only if @b{get_use_opengl} is 1.

Through a series of indirections, @b{run_opengl} eventually transfers control
to a virtual function called @b{handle_opengl}.  @b{handle_opengl} must be
overridden with a function to perform all the OpenGL routines.  The contents of
@b{handle_opengl} must be enclosed in @b{#ifdef HAVE_GL} ... @b{#endif} to
allow it to be compiled on systems with no graphics support, like render nodes.
The return value of @b{handle_opengl} is passed back from @b{run_opengl}.

@b{read_frame} can not be called from inside @b{handle_opengl}.  This would
create a recursive lockup because it would cause other objects to call
@b{run_opengl}.

Once inside @b{handle_opengl}, the plugin has full usage of all the OpenGL
features.  VFrame provides some functions to automate common OpenGL sequences.

The VFrame argument to @b{process_buffer} is always available through the
@b{get_output(int layer)} function.  If the plugin is multichannel, the layer
argument retrieves a specific layer of the output buffers.  The PBuffer of the
output buffer is where the OpenGL output must go if any processing is done.

@c cincvdoc_node_number_320
@node Drawing using OpenGL
@subsection Drawing using OpenGL
@cindex Drawing using OpenGL

The sequence of commands to draw on the output PBuffer stars with getting the
video in a memory area where it can be recalled for drawing:
@verbatim
get_output()->to_texture();
get_output()->enable_opengl();
@end verbatim

@itemize @bullet
@item @b{to_texture} transfers the OpenGL data from wherever it is to the
output's texture memory and sets the output state to TEXTURE@.
@item @b{enable_opengl} makes the OpenGL context relative to the output's
PBuffer.
@end itemize

The next step is to draw the texture with some processing on the PBuffer.  The
normal sequence of commands to draw a texture is:
@verbatim
get_output()->init_screen();
get_output()->bind_texture(0);
get_output()->draw_texture();
@end verbatim

@itemize @bullet
@item @b{VFrame::init_screen} sets the OpenGL frustum and parameters to known
values.
@item @b{VFrame::bind_texture(int texture_unit)} binds the texture to the given
texture unit and enables it.
@item @b{VFrame::draw_texture()} calls the vertex functions to draw the texture
normalized to the size of the PBuffer.  Copy this if you want custom vertices.
@end itemize

The last step in the handle_opengl routine, after the texture has been drawn on
the PBuffer, is to set the output's opengl state to SCREEN with a call to
@b{VFrame::set_opengl_state}.  The plugin should not read back the frame buffer
into a texture or row pointers if it has no further processing.  Plugins should
only leave the output in the texture or RAM if its location results from normal
processing.  They should set the opengl state to RAM or TEXTURE if they do.

@b{Colormodels in OpenGL:} @*
The colormodel exposed to OpenGL routines is always floating point since that
is what OpenGL uses, but it may be YUV or RGB depending on the project
settings.  If it is YUV, it is offset by 0.5 just like in software.  Passing
YUV colormodels to plugins was necessary for speed.  The other option was to
convert YUV to RGB in the first step that needed OpenGL@.  Every effect and
rendering step would have needed a YUV to RGB routine.  With the YUV retained,
only the final compositing step needs a YUV to RGB routine.

@c cincvdoc_node_number_321
@node Using shaders
@subsection Using shaders
@cindex OpenGL, using shaders

Very few effects can do anything useful with just a straight drawing of the
texture on the PBuffer.  They normally define a shader.  The shader is a C
program which runs on the graphics card.  Since the graphics card is optimized
for graphics, it can be much faster than running it on the CPU@.

Shaders are written in OpenGL Shading Language.  The shader source code is
contained in a string.  The normal sequence for using a shader comes after a
call to @b{enable_opengl}.

@verbatim
char *shader_source = "...";
unsigned char shader_id = VFrame::make_shader(0, shader_source, 0);
glUseProgram(shader_id);
// Set uniform variables using glUniform commands
@end verbatim

The compilation and linking step for shaders is encapsulated by the
VFrame::make_shader command.  It returns a shader_id which can be passed to
OpenGL functions.  The first and last arguments must always by 0.  And
arbitrary number of source strings may be put between the 0's.  The source
strings are concatenated by make_shader into one huge shader source.  If
multiple main functions are in the sources, the main functions are renamed and
run in order.

There are a number of useful macros for shaders in @file{playback3d.h}.  All
the shaders so far have been fragment shaders.  After the shader is
initialized, draw the texture starting from @b{init_screen}.  The shader
program must be disabled with another call to @b{glUseProgram(0)} and 0 as the
argument.

The shader_id and source code is stored in memory as long as Cinelerra runs.
Future calls to make_shader with the same source code run much faster.

@c cincvdoc_node_number_322
@node Aggregating plugins
@subsection Aggregating plugins
@cindex Aggregating plugins

Further speed improvements may be obtained by combining OpenGL routines from
two plugins into a single handle_opengl function.  This is done when @b{Frame
to Fields} and @b{RGB to 601} are attached in order.  Aggregations of more than
two plugins are possible but very hard to get working.  Aggregation is useful
for OpenGL because each plugin must copy the video from a texture to a PBuffer.
In software there was no copy operation.

In aggregation, one plugin processes everything from the other plugins and the
other plugins fall through.  The fall through plugins must copy their
parameters to the output buffer so they can be detected by the processing
plugin.

The VFrame used as the output buffer contains a parameter table for parameter
passing between plugins and it is accessed with @b{get_output()->get_params()}.
Parameters are set and retrieved in the table with calls to @b{update} and
@b{get} just like with defaults.

The fall through plugins must determine if the processor plugin is attached
with calls to @b{next_effect_is} and @b{prev_effect_is}.  These take the name
of the processor plugin as a string argument and return 1 if the next or
previous plugin is the processor plugin.  If either returns 1, the fall through
plugin must still call @b{read_frame} to propogate the data but return after
that.

The processor plugin must call @b{next_effect_is} and @b{prev_effect_is} to
determine if it is aggregated with a fall through plugin.  If it is, it must
perform the operations of the fall through plugin in its OpenGL routine.  The
parameters for the the fall through plugin should be available by
@b{get_output()->get_params()} if the fall through plugin set them.

@c cincvdoc_node_number_323
@node Keyboard shortcuts
@chapter Keyboard shortcuts
@cindex Keyboard shortcuts
@cindex Shortcuts

Alex Ferrer started summarizing most of the keyboard shortcuts.  Most of the
keys work without any modifier like @key{SHIFT} or @key{CTRL}.  Most windows
can be closed with a @kbd{CTRL-w}.  Most operations can be cancelled with
@key{ESC} and accepted with @key{RET}.

@section Program window
@cindex Program window
@subsection Editing Media
@cindex Media, editing

@multitable @columnfractions .2 .8
@item @kbd{z}
@tab Undo
@item @kbd{SHIFT Z}
@tab Re-Do
@item @kbd{x}
@tab Cut
@item @kbd{c}
@tab Copy
@item @kbd{v}
@tab Paste
@item @kbd{Del}
@tab Clear
@item @kbd{SHIFT Space}
@tab Paste Silence
@item @kbd{m}
@tab Mute region
@item @kbd{a}
@tab Select all
@item @kbd{SHIFT + click}
@tab When done over an edit causes the highlighted selection to extend to the
cursor position.  When done over the boundary of an effect causes the trim
operation to apply to one effect.
@end multitable

@subsection Editing Labels and In/Out Points
@cindex Editing labels and in/out points

@multitable @columnfractions .2 .8
@item @kbd{[}
@tab Toggle In point
@item @kbd{]}
@tab Toggle Out point
@item @kbd{l}
@tab Toggle label at current position
@item @kbd{CTRL <-}
@tab Go to Previous Label
@item @kbd{CTRL ->}
@tab Go to Next Label
@end multitable

@subsection Navigation
@cindex Navigation

@multitable @columnfractions .2 .8
@item @kbd{Right arrow}
@tab Move right *
@item @kbd{Left arrow}
@tab Move left *
@item @kbd{Up arrow}
@tab Zoom out *
@item @kbd{Down arrow}
@tab Zoom in *
@item @kbd{CTRL Up}
@tab Expand current curve amplitude
@item @kbd{CTRL Dn}
@tab Shrink current curve amplitude
@item @kbd{CTRL Alt Up}
@tab Expand all curve amplitude
@item @kbd{Ctrl Alt Dn}
@tab Shrink all curve amplitude
@item @kbd{Alt Up}
@tab Expand curve amplitude
@item @kbd{Alt Dn}
@tab Shrink curve amplitude
@item @kbd{f}
@tab Fit time displayed to selection
@item @kbd{Alt f}
@tab Make the range of all the automation types.  Fit the maximum and minimum
range of the current highlighted selection
@item @kbd{Ctrl Alt f}
@tab Make the range of the currently selected automation type fit the maximum
and minimum range of the current highlighted selection
@item @kbd{Alt Left}
@tab Move left one edit
@item @kbd{Alt Right}
@tab Move right one edit
@item @kbd{Page Up}
@tab Move up *
@item @kbd{Page Dn}
@tab Move down *
@item @kbd{Ctrl Page Up}
@tab Expand track height
@item @kbd{Ctrl Page Dn}
@tab Shrink track height
@item @kbd{Home}
@tab Go to beginning of timeline *
@item @kbd{End}
@tab Go to end of timeline *
@end multitable

* You may have to click on the timeline to deactivate any text boxes
or tumblers before these work.

@subsection File operations
@cindex File operations

@multitable @columnfractions .2 .8
@item @kbd{n}
@tab New project
@item @kbd{o}
@tab Load Files
@item @kbd{s}
@tab Save Project
@item @kbd{r}
@tab Record
@item @kbd{SHIFT R}
@tab Render
@item @kbd{q}
@tab Quit
@item @kbd{SHIFT P}
@tab Preferences
@item @kbd{SHIFT B}
@tab Batch Render
@item @kbd{SHIFT F}
@tab Set Format
@item @kbd{}
@tab
@end multitable

@subsection Key frame editing
@cindex Keyframes, editing

@multitable @columnfractions .2 .8
@item @kbd{SHIFT X}
@tab Cut keyframes
@item @kbd{SHIFT C}
@tab Copy keyframes
@item @kbd{SHIFT V}
@tab Paste keyframes
@item @kbd{SHIFT Del}
@tab Clear keyframes
@item @kbd{Alt c}
@tab Copy default keyframe
@item @kbd{Alt v}
@tab Paste default keyframe
@end multitable

@subsection Track Manipulation
@cindex Track manipulation

@multitable @columnfractions .2 .8
@item @kbd{t}
@tab Add Audio Track
@item @kbd{u}
@tab Insert default Audio Transition
@item @kbd{SHIFT T}
@tab Add Video Track
@item @kbd{SHIFT U}
@tab Insert Default Video Transition
@item @kbd{d}
@tab Delete last Track
@item @kbd{SHIFT L}
@tab Loop playback
@item @kbd{TAB}
@tab Toggle single track arming status
@item @kbd{SHIFT-TAB}
@tab Toggle every other track's arming status
@end multitable

@subsection What is drawn on the timeline
@cindex Timeline, what is drawn on the

@multitable @columnfractions .2 .8
@item @kbd{1}
@tab Show titles
@item @kbd{2}
@tab Show transitions
@item @kbd{3}
@tab Fade keyframes
@item @kbd{4}
@tab Mute keyframes
@item @kbd{5}
@tab Mode keyframes
@item @kbd{6}
@tab Pan keyframes
@item @kbd{7}
@tab Camera keyframes
@item @kbd{8}
@tab Projector keyframes
@item @kbd{9}
@tab Plugin keyframes
@item @kbd{0}
@tab Mask keyframes
@item @kbd{-}
@tab Camera Zoom
@item @kbd{=}
@tab Projector Zoom
@end multitable

@section Viewer and compositor windows
@cindex Viewer and compositor windows

@multitable @columnfractions .2 .8
@item @kbd{x}
@tab Cut
@item @kbd{c}
@tab Copy
@item @kbd{v}
@tab Paste
@item @kbd{v}
@tab Splice
@item @kbd{b}
@tab Overwrite
@item @kbd{[}
@tab Toggle In point
@item @kbd{]}
@tab Toggle Out point
@item @kbd{l}
@tab Toggle label at current position
@item @kbd{Ctrl <-}
@tab Go to Previous Label
@item @kbd{Ctrl ->}
@tab Go to Next Label
@item @kbd{Home}
@tab Go to beginning
@item @kbd{End}
@tab Go to end
@item @kbd{z}
@tab Undo
@item @kbd{SHIFT Z}
@tab Re-Do
@item @kbd{+}
@tab Zoom in
@item @kbd{-}
@tab Zoom out
@end multitable

@section Playback transport
@cindex Playback transport

Transport controls work in any window which has a playback transport.  They are
accessed through the number pad with num lock disabled.

@multitable @columnfractions .08 .17 .08 .17 .08 .17 .08 .17
@item @kbd{4}
@tab Frame back
@tab @kbd{5}
@tab Reverse Slow
@tab @kbd{6}
@tab Reverse
@tab @kbd{+}
@tab Reverse Fast
@item @kbd{1}
@tab Frame Forward
@tab @kbd{2}
@tab Forward Slow
@tab @kbd{3}
@tab Play
@tab @kbd{Enter}
@tab Fast Forward
@item @kbd{0}
@tab Stop
@tab
@tab
@tab
@tab
@tab
@tab
@end multitable

@key{SPACE} is normal Play, Hitting any key twice is Pause.

Hitting any transport control with @key{CTRL} down causes only the region
between the in/out points to be played, if in/out points are defined.

@section Record window
@cindex Record window

@multitable @columnfractions .2 .8
@item @kbd{Space}
@tab Start and pause recording of the current batch
@item @kbd{l}
@tab Toggle label at current position
@end multitable

@include gpl_en.texi

@ifnotplaintext
@ifnothtml
@ifnotdocbook
@c cincvdoc_node_number_324
@node Index
@unnumbered Index
@printindex cp
@end ifnotdocbook
@end ifnothtml
@end ifnotplaintext

@bye