pachi/peepo.git
9 years agoselfatari_cousin(): Suggest counter-capturesmaster
Petr Baudis [Mon, 4 Aug 2014 01:39:51 +0000 (4 03:39 +0200)]
selfatari_cousin(): Suggest counter-captures

When looking for alternatives of a self-atari move, aside of the
other liberty also look if we can counter-capture an opponent's
group.

This improves handling of some false-eyeish corner situations,
where we will more often capture the falsifying opponent stones
rather than connecting the pseudo-false eye (which can be fatal
to the status of the group).

9 years agoMoggy eyefillrate: Revert from 40 back to 60
Petr Baudis [Mon, 4 Aug 2014 00:29:57 +0000 (4 02:29 +0200)]
Moggy eyefillrate: Revert from 40 back to 60

With less than 50%, Pachi starts to massively misplay the concerned
corner situations.  I think measurements are way too noisy for
eyefillrate since the feature is very rarely applied.  Perhaps we
could get away fine with just pitching this at 100.

9 years agot-unit/sar.t: Add eye falsification selfatari testcase
Petr Baudis [Sat, 2 Aug 2014 21:55:59 +0000 (2 23:55 +0200)]
t-unit/sar.t: Add eye falsification selfatari testcase

9 years agoSelfatari: Allow capture-from-within moves
Petr Baudis [Sat, 2 Aug 2014 14:30:25 +0000 (2 16:30 +0200)]
Selfatari: Allow capture-from-within moves

10 years agoMoggy init: Also don't set nakaderate twice... sigh
Petr Baudis [Sat, 1 Feb 2014 03:05:26 +0000 (1 04:05 +0100)]
Moggy init: Also don't set nakaderate twice... sigh

10 years agoMoggy init: Also don't set nlibrate twice
Petr Baudis [Sat, 1 Feb 2014 03:03:07 +0000 (1 04:03 +0100)]
Moggy init: Also don't set nlibrate twice

10 years agoMoggy init: Rearrange rates initialization by their default value
Petr Baudis [Sat, 1 Feb 2014 03:01:32 +0000 (1 04:01 +0100)]
Moggy init: Rearrange rates initialization by their default value

10 years agoMoggy init: Do not set lcapturerate twice
Petr Baudis [Sat, 1 Feb 2014 02:59:55 +0000 (1 03:59 +0100)]
Moggy init: Do not set lcapturerate twice

10 years agoMoggy ladderrate: Add a note about obsolescence
Petr Baudis [Sat, 1 Feb 2014 02:59:36 +0000 (1 03:59 +0100)]
Moggy ladderrate: Add a note about obsolescence

10 years agoPlayout: New per-engine setboard_randomok setting
Petr Baudis [Wed, 29 Jan 2014 22:12:03 +0000 (29 23:12 +0100)]
Playout: New per-engine setboard_randomok setting

This is for a side-branch experiment where we generate some per-playout
settings for the Moggy playout policy.

10 years agoTwiddle several Moggy playout policy parameters based on CLOP
Petr Baudis [Wed, 29 Jan 2014 19:34:12 +0000 (29 20:34 +0100)]
Twiddle several Moggy playout policy parameters based on CLOP

These are CLOP experiments with pachi-29a24, single-threaded (i7-3770),
15x15 against GNUGo level 10; patterns were not used (?).
The number after [%] is |eigenvalue|.

abstime 500s            1000s           2000s           4000s
nakade  61% 3.39        80% 3.29        53% 7.62        66% 7.92
eyefill 54% 0.52        84% 0.61        49% 0.38        32% 2.45
nlib    29% 0.02        20% 0.09        22% 0.10        27% 0.28
ko      71% 0.08        63% 0.27        60% 0.33        48% 0.57
uElo SD 35              26              22              39      (smaller is more confident)

Just for your interest, mean winrate during tuning:
komi    -35.5           -40.5           -50.5           -50.5   (to give GNUGo an advantage)
Elo     140             164             84              101

Nota bene:
- 4000s S.D. single-threaded translates just to 500s S.D. (8 minutes)
  8-threaded, which is still faster than a regular KGS blitz.
- In general, 500s and 4000s values are unfortunately of low confidence,
  this Elo spread is more than influence of most of the parameters.

Parameter analysis:
- nakade: let's go with the rough mean of 60%, which seems to hold.
- eyefill: we see a clear downward trend; let's go with 40%, more
  tuning needed.
- nlib: we see about constant 25% optimum, but the feature is not
  very interesting in general.
- ko: steady decrease, but 40% may hit closer to home in realistic
  KGS setting.

Overally, after 500 games, the performance changes as follows (komi -55):

-Elo uElo +Elo (+-1SD)
361  377  392  prior=pattern=80,playout=moggy:nakaderate=20:nlibrate=20:korate=20:eyefillrate=60
396  411  427  prior=pattern=160,playout=moggy:nakaderate=20:nlibrate=20:korate=20:eyefillrate=60
384  399  414  prior=pattern=160,playout=moggy:nakaderate=20:nlibrate=25:korate=40:eyefillrate=60
404  420  435  prior=pattern=160,playout=moggy:nakaderate=60:nlibrate=25:korate=40:eyefillrate=40

Not too conclusive, unfortuinately, but it seems we should have a slight
improvement.

10 years agoUCT pattern prior: Raise default from 80 to 160
Petr Baudis [Wed, 29 Jan 2014 19:00:20 +0000 (29 20:00 +0100)]
UCT pattern prior: Raise default from 80 to 160

After 500 games with single threaded (i7 HT) -t =4000, there is 44-40+22
Elo improvement here; let's optimize more for realistic scenarios for
our users even though on big cluster, 80 may be a better default.

10 years agoMoggy nakaderate: 20 -> default, i.e. much more often
Petr Baudis [Tue, 31 Dec 2013 02:15:30 +0000 (31 03:15 +0100)]
Moggy nakaderate: 20 -> default, i.e. much more often

Based on recent tuning; ongoing tweaks.

10 years agoMoggy nakade_check(): Call _after_ x-lib checks, not before
Petr Baudis [Tue, 31 Dec 2013 00:47:26 +0000 (31 01:47 +0100)]
Moggy nakade_check(): Call _after_ x-lib checks, not before

After 3000-games test, there is still no statistically significant
information about it being better or worse than previously, but it sure
makes more sense so let's do it.

10 years agoautotest-show: Add support for Elo display
Petr Baudis [Sun, 29 Dec 2013 21:37:11 +0000 (29 22:37 +0100)]
autotest-show: Add support for Elo display

10 years agoUCT debug_after: Extra printouts
Petr Baudis [Sat, 28 Dec 2013 17:49:25 +0000 (28 18:49 +0100)]
UCT debug_after: Extra printouts

10 years agoUCT dumpthres: Convert from something weird to true fraction of #root.playouts
Petr Baudis [Tue, 24 Dec 2013 02:12:42 +0000 (24 03:12 +0100)]
UCT dumpthres: Convert from something weird to true fraction of #root.playouts

10 years agoUCT debug_after: New aid to combine regular thinking + debugging simulation dump
Petr Baudis [Tue, 24 Dec 2013 01:55:56 +0000 (24 02:55 +0100)]
UCT debug_after: New aid to combine regular thinking + debugging simulation dump

10 years agoFix build in packaging systems like portage or debian
Petr Baudis [Mon, 16 Dec 2013 13:38:00 +0000 (16 14:38 +0100)]
Fix build in packaging systems like portage or debian

We now allow custom LIBS, LDFLAGS settings akin to custom CFLAGS
environment settings by using SYS_* for our overridable internal flags.

Pointed out by Guillaume BINET, solution proposed by Stanisław Halik.

10 years agoneighbor_count_at(): Fix S_NONE warning
Petr Baudis [Mon, 16 Dec 2013 03:07:14 +0000 (16 04:07 +0100)]
neighbor_count_at(): Fix S_NONE warning

Pointed out by Eugene Lovelace.

10 years agois_border_ladder() shortage of libs test: Properly test for potential self-atari...
Petr Baudis [Mon, 16 Dec 2013 03:03:56 +0000 (16 04:03 +0100)]
is_border_ladder() shortage of libs test: Properly test for potential self-atari condition

10 years agoTODO: GNUGo cross-pollination?
Petr Baudis [Sun, 15 Dec 2013 15:03:42 +0000 (15 16:03 +0100)]
TODO: GNUGo cross-pollination?

10 years agoTODO: Game analysis features
Petr Baudis [Sun, 15 Dec 2013 14:55:18 +0000 (15 15:55 +0100)]
TODO: Game analysis features

10 years agoTODO: is_private() hack
Petr Baudis [Sun, 15 Dec 2013 14:48:00 +0000 (15 15:48 +0100)]
TODO: is_private() hack

10 years agoTODO: Revamp, elaborate on things
Petr Baudis [Sun, 15 Dec 2013 14:45:49 +0000 (15 15:45 +0100)]
TODO: Revamp, elaborate on things

10 years agoConsider a tree during which we drop komi as untrustworthy - don't resign or reuse it
Petr Baudis [Fri, 13 Dec 2013 11:41:46 +0000 (13 12:41 +0100)]
Consider a tree during which we drop komi as untrustworthy - don't resign or reuse it

In late endgame, raising a komi by 1pt can fatally destroy the tree.

Problem discovered in KGS pachi-Marknguyen on 2013-12-13.

10 years agosgf2gtp.py: Default SZ to 19
Petr Baudis [Fri, 13 Dec 2013 11:38:24 +0000 (13 12:38 +0100)]
sgf2gtp.py: Default SZ to 19

10 years agopattern_rate_moves(): Add a debug print listing all patterns considered
Petr Baudis [Fri, 13 Dec 2013 11:38:09 +0000 (13 12:38 +0100)]
pattern_rate_moves(): Add a debug print listing all patterns considered

10 years agoPATTERN_SPEC_MATCH_DEFAULT: Disable PF_CAPTURE_COUNTSTONES as it was not used for...
Petr Baudis [Tue, 10 Dec 2013 02:08:18 +0000 (10 03:08 +0100)]
PATTERN_SPEC_MATCH_DEFAULT: Disable PF_CAPTURE_COUNTSTONES as it was not used for generating stock pattern files

Using non-matching pattern files and patternspecs is rather fatal as
common situations will not generate matching patterns anymore. A symptom
of that was kindly reported by Martin Miller.

10 years agoadd arm-linux-androideabi v16 instructions
Stanislaw Halik [Fri, 15 Nov 2013 14:40:19 +0000 (15 15:40 +0100)]
add arm-linux-androideabi v16 instructions

10 years agooverriding {c,ld}flags eg when crossing android
Stanislaw Halik [Fri, 15 Nov 2013 14:30:38 +0000 (15 15:30 +0100)]
overriding {c,ld}flags eg when crossing android

10 years agouct_progress_text(): Typo fix (add missing space)
Petr Baudis [Thu, 14 Nov 2013 00:54:34 +0000 (14 01:54 +0100)]
uct_progress_text(): Typo fix (add missing space)

10 years agounderflow less by s/floating_t/double/
Stanislaw Halik [Wed, 13 Nov 2013 23:56:50 +0000 (14 00:56 +0100)]
underflow less by s/floating_t/double/

10 years agoshow color information for easy twogtp regress test
Stanislaw Halik [Wed, 13 Nov 2013 23:47:11 +0000 (14 00:47 +0100)]
show color information for easy twogtp regress test

10 years agonetwork.c: Add missing netinet/in.h include (by Stanisław Halik)
Petr Baudis [Mon, 11 Nov 2013 22:39:18 +0000 (11 23:39 +0100)]
network.c: Add missing netinet/in.h include (by Stanisław Halik)

10 years agoMoggy 3x3 cut keima pattern: Adjust to cover cutting hane-nobi shapes and bamboo...
Petr Baudis [Thu, 12 Sep 2013 07:45:42 +0000 (12 09:45 +0200)]
Moggy 3x3 cut keima pattern: Adjust to cover cutting hane-nobi shapes and bamboo joints too

Inspired by pachi vs. Aya in the September KGS slow tournament.

10 years agoUCB1AMAF vloss_sqrt: Enable, tested to fix high-threading tree deformations
Petr Baudis [Sun, 25 Aug 2013 17:30:50 +0000 (25 19:30 +0200)]
UCB1AMAF vloss_sqrt: Enable, tested to fix high-threading tree deformations

10 years agoMoggy: Bump selfatarirate to 95.
Petr Baudis [Sun, 25 Aug 2013 15:22:36 +0000 (25 17:22 +0200)]
Moggy: Bump selfatarirate to 95.

10 years agoUCB1AMAF: vloss_sqrt to re-normalize virtual loss for better many-threads performance
Petr Baudis [Sat, 24 Aug 2013 23:20:06 +0000 (25 01:20 +0200)]
UCB1AMAF: vloss_sqrt to re-normalize virtual loss for better many-threads performance

Rescale virtual loss value to square root of #threads. This mitigates
the number of virtual losses added in case of a large amount of
threads; it seems that with linear virtual losses, overly diverse
exploration caused by this may cause a wrong mean value computed
for the parent node.

Off by default for now.

10 years agoRewrite virtual loss support
Petr Baudis [Sat, 17 Aug 2013 20:18:47 +0000 (17 22:18 +0200)]
Rewrite virtual loss support

Instead of directly manipulating the u stats, track the number
of parallel descents separately and use during node evaluation.
Also, manipulate the counter atomically.

On highly parallel machines, the amount of inconsistency introduced by
virtual loss was just too huge - mostly, there would be a tendency of
"forgetting" the virtual loss addition, but performing the removal,
therefore swinging the results off balance, an effect that would be
self-reinforcing (a funny consequence of how the search works) and
cause the tree to quickly deteriorate into very long, not particulary
wise branches.

The change makes virtual loss support reliable (though I did not test
*too* rigorously if it still works so well, game performance wise)
and doesn't seem to introduce any noticeable performance degradation.

10 years agoMerge branch 'master' of ssh://repo.or.cz/srv/git/pachi
Petr Baudis [Thu, 15 Aug 2013 18:02:25 +0000 (15 20:02 +0200)]
Merge branch 'master' of ssh://repo.or.cz/srv/git/pachi

10 years agoplay_random_game(): Don't enforce minimum gamelen (10)
Petr Baudis [Thu, 15 Aug 2013 18:02:16 +0000 (15 20:02 +0200)]
play_random_game(): Don't enforce minimum gamelen (10)

This intereferes with extremely long tree branches, triggering an
assertion failure wrt. amafmap->gamelen.

10 years agoStop pondering on kgs-game_over so that we don't waste CPU when waiting for the next...
Petr Baudis [Thu, 15 Aug 2013 00:54:03 +0000 (15 02:54 +0200)]
Stop pondering on kgs-game_over so that we don't waste CPU when waiting for the next game

11 years agouct_search_result(): Resign if only pass is feasible but it won't win the game
Petr Baudis [Fri, 19 Apr 2013 19:27:19 +0000 (19 21:27 +0200)]
uct_search_result(): Resign if only pass is feasible but it won't win the game

Reported by Martin Mueller. This change needs careful testing for
unintended consequences, in case Pachi starts resigning at weird
times...

11 years agoUCT banner: Allow + instead of space, because kgsGtp is braindamaged
Petr Baudis [Sun, 14 Apr 2013 17:29:48 +0000 (14 19:29 +0200)]
UCT banner: Allow + instead of space, because kgsGtp is braindamaged

11 years agoMoggy pat3_gammas: Tuned default values
Petr Baudis [Sun, 14 Apr 2013 17:03:49 +0000 (14 19:03 +0200)]
Moggy pat3_gammas: Tuned default values

11 years agoMerge branch 'moggy3patgamma'
Petr Baudis [Sun, 14 Apr 2013 17:00:11 +0000 (14 19:00 +0200)]
Merge branch 'moggy3patgamma'

11 years agomq_gamma_pick(): Fix case of zero totalmoggy3patgamma
Petr Baudis [Sun, 24 Mar 2013 21:22:11 +0000 (24 22:22 +0100)]
mq_gamma_pick(): Fix case of zero total

11 years agoMoggy test_pattern3_here(): Disable spurious debug print
Petr Baudis [Sun, 24 Mar 2013 20:19:36 +0000 (24 21:19 +0100)]
Moggy test_pattern3_here(): Disable spurious debug print

11 years agoMoggy eyefixrate: Set to 100 after brief tuning
Petr Baudis [Sun, 24 Mar 2013 20:15:54 +0000 (24 21:15 +0100)]
Moggy eyefixrate: Set to 100 after brief tuning

11 years agoMoggy, Pattern3: Add support for assigning gamma values to 3x3 patterns (by default...
Petr Baudis [Fri, 22 Mar 2013 15:37:07 +0000 (22 16:37 +0100)]
Moggy, Pattern3: Add support for assigning gamma values to 3x3 patterns (by default all gammas are still 1.0)

11 years agoPattern3: Store pattern index in the top six bits of the hash table
Petr Baudis [Fri, 22 Mar 2013 15:36:34 +0000 (22 16:36 +0100)]
Pattern3: Store pattern index in the top six bits of the hash table

11 years agomq.h: Add mq_gamma_{add,pick} routines for dealing with gamma probdists
Petr Baudis [Fri, 22 Mar 2013 15:36:07 +0000 (22 16:36 +0100)]
mq.h: Add mq_gamma_{add,pick} routines for dealing with gamma probdists

11 years agoMoggy eye fix: Fill falsifying move when approach liberty has been filledfalseeye
Petr Baudis [Fri, 22 Mar 2013 12:13:05 +0000 (22 13:13 +0100)]
Moggy eye fix: Fill falsifying move when approach liberty has been filled

Inspired by the Zen-Ishida Densei-sen game.

11 years agouct_playout(): Add comment warning about the result number mess, I got caught up...
Petr Baudis [Tue, 19 Mar 2013 15:23:43 +0000 (20 00:23 +0900)]
uct_playout(): Add comment warning about the result number mess, I got caught up in it myself again

11 years agouct_playout(): Add playout# to UCT walk debug print
Petr Baudis [Tue, 19 Mar 2013 15:23:25 +0000 (20 00:23 +0900)]
uct_playout(): Add playout# to UCT walk debug print

11 years agoMoggy joseki_check(): Do not suggest invalid joseki moves (on hash collisions)
Petr Baudis [Tue, 19 Mar 2013 15:23:03 +0000 (20 00:23 +0900)]
Moggy joseki_check(): Do not suggest invalid joseki moves (on hash collisions)

11 years agoREADME (test opponent): +pondering=0, pass_all_alive
Petr Baudis [Tue, 19 Mar 2013 12:40:37 +0000 (19 21:40 +0900)]
README (test opponent): +pondering=0, pass_all_alive

11 years agoMoggy mq_tagged_choose: Debug print with probability distribution
Petr Baudis [Sat, 9 Mar 2013 00:06:33 +0000 (9 01:06 +0100)]
Moggy mq_tagged_choose: Debug print with probability distribution

11 years agoMoggy fullchoose: Do not match features with zero rate
Petr Baudis [Sat, 9 Mar 2013 00:06:15 +0000 (9 01:06 +0100)]
Moggy fullchoose: Do not match features with zero rate

11 years agoMoggy fullchoose: Fix global atari matching
Petr Baudis [Sat, 9 Mar 2013 00:05:40 +0000 (9 01:05 +0100)]
Moggy fullchoose: Fix global atari matching

11 years agotools/sgf2gtp.py: Graceful handling of SGFs without black/white setup
Petr Baudis [Thu, 7 Mar 2013 01:39:22 +0000 (7 02:39 +0100)]
tools/sgf2gtp.py: Graceful handling of SGFs without black/white setup

11 years agoDynkomi linear: Do not call adjusting linear_permove() from linear_persim()
Petr Baudis [Wed, 6 Mar 2013 03:21:10 +0000 (6 04:21 +0100)]
Dynkomi linear: Do not call adjusting linear_permove() from linear_persim()

The adjustment is not persistent if not called over permove API. This
caused the endless streams of dynkomi messages in some yose situations.

11 years agoDynkomi linear: Factor out linear_simple() from linear_permove()
Petr Baudis [Wed, 6 Mar 2013 03:19:49 +0000 (6 04:19 +0100)]
Dynkomi linear: Factor out linear_simple() from linear_permove()

11 years agoThread managers: Raise spawned/joined worker prints debug level to reduce noise at...
Petr Baudis [Mon, 4 Mar 2013 18:34:43 +0000 (4 19:34 +0100)]
Thread managers: Raise spawned/joined worker prints debug level to reduce noise at -d 4

11 years agosgf2gtp.py: Include setup stones
D. Cooper Stevenson [Wed, 20 Feb 2013 23:24:09 +0000 (21 00:24 +0100)]
sgf2gtp.py: Include setup stones

This is useful if you want to export a position for processing by the
engine: simply save the position from the GUI and run this modified
sgf2gtp script on it.

11 years agosgf2gtp.py: Convert tabs to four spaces, as is the Python custom
Petr Baudis [Wed, 20 Feb 2013 23:18:35 +0000 (21 00:18 +0100)]
sgf2gtp.py: Convert tabs to four spaces, as is the Python custom

11 years agonetwork.c: SO_REUSEADDR value shall be an int (found by Farmer Lu)
Petr Baudis [Wed, 20 Feb 2013 10:44:36 +0000 (20 11:44 +0100)]
network.c: SO_REUSEADDR value shall be an int (found by Farmer Lu)

11 years agoMerge branch 'pachi-10-satsugen'
Petr Baudis [Mon, 13 Aug 2012 19:49:38 +0000 (13 21:49 +0200)]
Merge branch 'pachi-10-satsugen'

11 years agouct_progress_json(): Remove extra curly braces around choice valuepachi-10-satsugen
Petr Baudis [Mon, 13 Aug 2012 19:48:20 +0000 (13 21:48 +0200)]
uct_progress_json(): Remove extra curly braces around choice value

11 years agoPachi Retsugen-devel 10.99mob
Petr Baudis [Sun, 12 Aug 2012 20:23:25 +0000 (12 22:23 +0200)]
Pachi Retsugen-devel 10.99

11 years agoPachi Satsugen 10.00pachi-10.00-satsugen
Petr Baudis [Sun, 12 Aug 2012 20:22:13 +0000 (12 22:22 +0200)]
Pachi Satsugen 10.00

11 years agoDistributed mode: no time limit by default on slaves, control on master only
Jean-loup Gailly [Sun, 12 Aug 2012 19:36:52 +0000 (12 21:36 +0200)]
Distributed mode: no time limit by default on slaves, control on master only

11 years agoChange default time settings from -t =80000 to -t 15
Petr Baudis [Sun, 12 Aug 2012 14:56:28 +0000 (12 16:56 +0200)]
Change default time settings from -t =80000 to -t 15

This loses a semi-nice property that Pachi will by default have
the same strength on any computer, but on the other hand it will
make Pachi adjust its strength based on the computer without playing
needlessly fast or taking way too long time per move. This should be
more user friendly. Suggested by Jean-Loup.

11 years agoUCT pondering: Enable by default
Petr Baudis [Sun, 12 Aug 2012 14:12:32 +0000 (12 16:12 +0200)]
UCT pondering: Enable by default

11 years agoUCT max_maintime_ratio: Raise back to 2.0
Petr Baudis [Sun, 12 Aug 2012 12:24:52 +0000 (12 14:24 +0200)]
UCT max_maintime_ratio: Raise back to 2.0

Based on empirical observation during a 9x9 tournament.

2.0 is still dramatic decrease from the previous default 3.0 and should
avoid most troublesome situations, but my gut feeling is that Pachi is
just playing way too fast now. It's not based on hard data, but neither
was the reduction to 1.5, IIRC.

11 years agoUCT virtual_loss: Revert default change
Petr Baudis [Sun, 12 Aug 2012 12:19:17 +0000 (12 14:19 +0200)]
UCT virtual_loss: Revert default change

It seems all our virtual_loss tuning was bogus due to a mistake
in the option setting code - all tests were in fact done with
virtual_loss zero.

High virtual_loss has catastrophic effect on gameplay - it transforms
the tree to long narrow corridors with high difference between black
and white node values. I'm not 100% sure about the cause even though
I spent quite long time trying to debug this, but I think it's because
the noise introduced by updates lost due to conflicts is just too high.

Moreover, with virtual_loss raising with number of threads, both the
noise rate and noise magnitude goes up, amplifying the effect.

11 years agoUCT virtual_loss: Allow setting virtual_loss to arbitrary values
Petr Baudis [Sun, 12 Aug 2012 12:18:41 +0000 (12 14:18 +0200)]
UCT virtual_loss: Allow setting virtual_loss to arbitrary values

11 years agouct_leaf_node(): Fix node value debug print
Petr Baudis [Sun, 12 Aug 2012 12:18:15 +0000 (12 14:18 +0200)]
uct_leaf_node(): Fix node value debug print

11 years agofbook_init(): Remove bogus comment
Petr Baudis [Sun, 12 Aug 2012 10:37:24 +0000 (12 12:37 +0200)]
fbook_init(): Remove bogus comment

11 years agoNew parameter defaults, obtained by tuning on large configurations
Jean-loup Gailly [Sun, 12 Aug 2012 00:21:25 +0000 (12 02:21 +0200)]
New parameter defaults, obtained by tuning on large configurations

Kept default debug_level=3 and pondering=0 but for optimal results
use: -d 0 pondering=1

11 years agochat_init(): Fix problematic ptrdiff_t format string handling
Petr Baudis [Thu, 9 Aug 2012 22:09:42 +0000 (10 00:09 +0200)]
chat_init(): Fix problematic ptrdiff_t format string handling

Introduced in commit 2e8729f.

11 years agoAmmend to static inline, for no gcc warning.
Neil Mclean [Wed, 8 Aug 2012 19:32:09 +0000 (8 21:32 +0200)]
Ammend to static inline, for no gcc warning.

11 years agoGCC: -Wextra warning: ‘static’ is not at beginning of declaration.
Neil Mclean [Wed, 8 Aug 2012 19:31:52 +0000 (8 21:31 +0200)]
GCC: -Wextra warning: ‘static’ is not at beginning of declaration.

11 years agoChange type of mercymin to 'int' to avoid gcc sign-compare warning.
Neil Mclean [Wed, 8 Aug 2012 19:31:10 +0000 (8 21:31 +0200)]
Change type of mercymin to 'int' to avoid gcc sign-compare warning.

11 years agoWin32 gcc compiler warning fix. Added util.h prototype for stpcpy().
Neil Mclean [Wed, 18 Jul 2012 00:06:29 +0000 (18 01:06 +0100)]
Win32 gcc compiler warning fix. Added util.h prototype for stpcpy().

11 years agoFix stack overflows and crashes when reading very long ladders
Petr Baudis [Mon, 6 Aug 2012 00:32:43 +0000 (6 02:32 +0200)]
Fix stack overflows and crashes when reading very long ladders

11 years agoUCT initial_extra_komi: Way to seed adaptive dynkomi, also preserve it across state...
Petr Baudis [Sun, 5 Aug 2012 16:24:04 +0000 (5 18:24 +0200)]
UCT initial_extra_komi: Way to seed adaptive dynkomi, also preserve it across state resets

This means that unexpected opponent move (promotion failure) will not reset the dynamic komi. This is fairly important esp. for the maximize_score option.

11 years agoUCT spawn_thread_manager(): Make sure each thread gets 1M stack
Petr Baudis [Wed, 1 Aug 2012 20:56:03 +0000 (1 22:56 +0200)]
UCT spawn_thread_manager(): Make sure each thread gets 1M stack

11 years agomiddle_ladder_walk(): Use alloca() for board copy allocation
Petr Baudis [Wed, 1 Aug 2012 20:55:13 +0000 (1 22:55 +0200)]
middle_ladder_walk(): Use alloca() for board copy allocation

This should somewhat reduce stack usage which seems to give us some trouble with deep ladders.

11 years agoREADME: Mention large-scale board patterns
Petr Baudis [Wed, 1 Aug 2012 16:40:20 +0000 (1 18:40 +0200)]
README: Mention large-scale board patterns

11 years agoREADME: Update strength estimates
Petr Baudis [Wed, 1 Aug 2012 16:37:50 +0000 (1 18:37 +0200)]
README: Update strength estimates

11 years agoREADME Greedy Pachi: Mention xkomi/extra komi messages
Petr Baudis [Wed, 1 Aug 2012 15:22:16 +0000 (1 17:22 +0200)]
README Greedy Pachi: Mention xkomi/extra komi messages

11 years agouct_progress_text(): komi -> xkomi
Petr Baudis [Wed, 1 Aug 2012 15:21:27 +0000 (1 17:21 +0200)]
uct_progress_text(): komi -> xkomi

11 years agoUCT adaptive dynkomi: Less verbose
Petr Baudis [Tue, 31 Jul 2012 07:41:51 +0000 (31 09:41 +0200)]
UCT adaptive dynkomi: Less verbose

11 years agoUCT: Introduce 'maximize_score' preset that turns on adaptive dynkomi, val_scale...
Petr Baudis [Mon, 30 Jul 2012 20:52:51 +0000 (30 22:52 +0200)]
UCT: Introduce 'maximize_score' preset that turns on adaptive dynkomi, val_scale and allow_losing_pass.

11 years agoRULES_PASS_STONES -> RULES_SIMING, also count handicap stones as points
Petr Baudis [Mon, 30 Jul 2012 20:37:10 +0000 (30 22:37 +0200)]
RULES_PASS_STONES -> RULES_SIMING, also count handicap stones as points

11 years agoRULES_PASS_STONES: Correctly support undo
Petr Baudis [Mon, 30 Jul 2012 20:04:53 +0000 (30 22:04 +0200)]
RULES_PASS_STONES: Correctly support undo

11 years agoPreserve rules over board_clear()
Petr Baudis [Mon, 30 Jul 2012 20:03:38 +0000 (30 22:03 +0200)]
Preserve rules over board_clear()

11 years agoAdd support for RULES_PASS_STONES - player hands a stone to the opponent on pass
Petr Baudis [Mon, 30 Jul 2012 12:27:13 +0000 (30 14:27 +0200)]
Add support for RULES_PASS_STONES - player hands a stone to the opponent on pass

Requested by Robert Jasiek for the 13x13 EGC2012 tournament.