From f846b599f971e0d250cd0d817aa1b957ef8b8da1 Mon Sep 17 00:00:00 2001 From: Petr Baudis Date: Mon, 16 Jan 2012 01:43:03 +0100 Subject: [PATCH] t-regress: Add 2011-08-24-StoneGrid-pachi2.sgf --- t-regress/TODO | 12 - .../by-ladder/2011-08-24-StoneGrid-pachi2.sgf | 1 + t-regress/games/2011-08-24-StoneGrid-pachi2.sgf | 391 +++++++++++++++++++++ 3 files changed, 392 insertions(+), 12 deletions(-) create mode 120000 t-regress/by-ladder/2011-08-24-StoneGrid-pachi2.sgf create mode 100644 t-regress/games/2011-08-24-StoneGrid-pachi2.sgf diff --git a/t-regress/TODO b/t-regress/TODO index 6f3e8c7..3e163a5 100644 --- a/t-regress/TODO +++ b/t-regress/TODO @@ -125,15 +125,3 @@ kills above) * http://files.gokgs.com/games/2011/8/10/jerryfish-pachi2.sgf semeai lost even game against 3k. Win ratio drops from 61% at 117 N1 to 49% at the next move. So pachi didn't evaluate the semai correctly. - -* http://files.gokgs.com/games/2011/8/24/StoneGrid-pachi2.sgf ladder - Move 51 onwards: pachi plays a broken ladder, which almost cost the game. - This might be the same bug as - http://files.gokgs.com/games/2011/8/9/somrak-pachi2-2.sgf - http://files.gokgs.com/games/2011/8/9/pachi2-BlueSpark.sgf - http://files.gokgs.com/games/2011/8/8/Dallas-pachi2-2.sgf - http://files.gokgs.com/games/2011/7/28/pachi2-pkunzip-2.sgf - http://files.gokgs.com/games/2011/6/13/pachi30s-Jep.sgf - http://files.gokgs.com/games/2011/6/6/tyzef-pachi30s.sgf - http://files.gokgs.com/games/2011/6/5/pachi2-SimpleBot.sgf - http://files.gokgs.com/games/2011/1/11/llopl-pachi2.sgf diff --git a/t-regress/by-ladder/2011-08-24-StoneGrid-pachi2.sgf b/t-regress/by-ladder/2011-08-24-StoneGrid-pachi2.sgf new file mode 120000 index 0000000..8baae4b --- /dev/null +++ b/t-regress/by-ladder/2011-08-24-StoneGrid-pachi2.sgf @@ -0,0 +1 @@ +../games/2011-08-24-StoneGrid-pachi2.sgf \ No newline at end of file diff --git a/t-regress/games/2011-08-24-StoneGrid-pachi2.sgf b/t-regress/games/2011-08-24-StoneGrid-pachi2.sgf new file mode 100644 index 0000000..e2d8113 --- /dev/null +++ b/t-regress/games/2011-08-24-StoneGrid-pachi2.sgf @@ -0,0 +1,391 @@ +(;GM[1]FF[4]CA[UTF-8]AP[CGoban:3]ST[2] +RU[Chinese]SZ[19]KM[7.50]TM[14400] +PW[StoneGrid]PB[pachi2]BR[3d]DT[2011-08-24]GC[51 E14:bad!{ladder} pachi plays a broken ladder, which almost cost the game]PC[The KGS Go Server at http://www.gokgs.com/]C[StoneGrid [-\]: GTP Engine for StoneGrid (white): stonegrid version 0.3.4 48 +pachi2 [3d\]: GTP Engine for pachi2 (black): Pachi Distributed Engine version 8.99 (Hakugen-devel): I'm playing the distributed engine. When I'm losing, I will resign, if I think I win, I play until you pass. Anyone can send me 'winrate' in private chat to get my assessment of the position. +]RE[B+Resign] +;B[pp]BL[14399.379]C[maproomad [-\]: pachi2: In 2367711 playouts at 64 machines, white M10 can win with 96.89% probability. +maproomad [-\]: I guess that is left over from the previous round +maproomad [-\]: pachi2: In 10750 playouts at 64 machines, black Q4 can win with 100.00% probability. +maproomad [-\]: maybe not +] +;W[mg]WL[14246.492]C[BinX [-\]: black q4 can win with a 100% probability ? kind of a divine first move ? +] +;B[cj]BL[14189.181]C[BinX [-\]: oh :) +BinX [-\]: fun game. +] +;W[no]WL[14092.666] +;B[pm]BL[14083.262] +;W[op]WL[13940.553] +;B[pq]BL[13977.422]C[maproomad [-\]: pachi2: In 125164015 playouts at 64 machines, black Q3 can win with 48.70% probability. +] +;W[cq]WL[13788.393] +;B[co]BL[13775.243]C[uurtamo [-\]: but pachi2 could have checked each of the 6-ply inbetween then. :) +] +;W[dn]WL[13637.551]C[goIngo [-\]: is this avantgarde go? +uurtamo [-\]: it's curious. +BinX [-\]: oh, pachi lose confidence with q3 already :o +] +;B[cn]BL[13669.876]C[uurtamo [-\]: yes, to go from 100% to 48.7% in a sequence of moves all of which could be explored, means something isn't right. :) +blob [5k\]: sure, n13, to hell with joseki +uurtamo [-\]: i don't understand n13 well enough to say whether it's joseki or not. +] +;W[do]WL[13486.901]C[blob [5k\]: n13 of course, is not joseki +blob [5k\]: not yet anyway....... +uurtamo [-\]: joseki isn't about what is in a book, it's about what makes the board equal. +blob [5k\]: uh, no +blob [5k\]: it's what has the historical concensus of being a solid move +BinX [-\]: ? +] +;B[dm]BL[13564.821]C[uurtamo [-\]: it's what has the historical concensus of giving a fair outcome for both players. +blob [5k\]: which n13 hasn't +uurtamo [-\]: fair enough. +uurtamo [-\]: however, it might give a fair outcome for both players. all i'm saying. +blob [5k\]: "might".... which is why i said "not yet anyway" +] +;W[cm]WL[13336.94] +;B[em]BL[13457.786]C[uurtamo [-\]: you're fixated on the "historical concensus" and i'm fixated on the "fair outcome". no biggie. +blob [5k\]: sure, but you just misunderstand the meaning of joseki apparently +uurtamo [-\]: could very well be the case. +] +;W[cp]WL[13186.961]C[blob [5k\]: the jo in joseki means "established" (among other things) +blob [5k\]: i.o.w., an established move +] +;B[bm]BL[13363.95]C[blob [5k\]: ....why do these bots get so much time? +uurtamo [-\]: because it's the "slow play tournament" +blob [5k\]: it sort of rewards brute force approaches to machine go, which i think isn't really desirable +] +;W[cc]WL[13039.78]C[uurtamo [-\]: isn't desirable in what sense? +blob [5k\]: is there a fast tournament? +uurtamo [-\]: have you ever tried to write a machine go program? +blob [5k\]: i've started one yes +uurtamo [-\]: there are tournaments every month, and a server where programs are continuously playing one another +blob [5k\]: by not desirable i mean that i find efficiency in computation intuitively more "human" +blob [5k\]: than monte carlo-esque stuff anyway +uurtamo [-\]: http://www.weddslist.com/kgs +blob [5k\]: thanks for that +] +;B[kq]BL[13166.445]C[uurtamo [-\]: i think that you want computation to be a certain way that it might not be. +blob [5k\]: look, the status quo is that humans are stronger than machines at go, yet use less resources (energy wise) +blob [5k\]: i think it's natural to learn by that example somehow +uurtamo [-\]: that idea was very popular in the 70's. +blob [5k\]: so? +uurtamo [-\]: i'm just suggesting that other approaches have been quite a bit more successful, not just in computer go, but machine learning in general and that you ought not discount their success. +uurtamo [-\]: there's a mistake in being fixated on what "seems natural" +uurtamo [-\]: computers aren't all that natural and neither is computation. +blob [5k\]: i am familiar with ai techniques +] +;W[ko]WL[12892.264]C[uurtamo [-\]: there's a mailing list about computer go if you're interested in seeing what people think about. +blob [5k\]: i'm saying that it seems that human level of higher abstraction can lead to significant gains in efficiency +blob [5k\]: i know what's out there already..... +] +;B[hq]BL[12958.564]C[blob [5k\]: there's a huge gain to be had over mere monte carlo, by clustering and classifying patterns dynamically +uurtamo [-\]: well, the easiest way to test your ideas is on CGOS. +uurtamo [-\]: http://cgos.boardspace.net +blob [5k\]: yeah, as soon as i get some more time on my hands ): +blob [5k\]: i will :) +uurtamo [-\]: because otherwise you are simply fantasizing. +uurtamo [-\]: also, "mere monte carlo" has developed quite a bit. +blob [5k\]: i already have a working ML go prototype +uurtamo [-\]: great. you can probably hook it into cgos with very little trouble, then. +] +;W[jp]WL[12747.122]C[blob [5k\]: probably, but i want to work on it more first +blob [5k\]: it's still weak +uurtamo [-\]: cgos will let you see its level exactly compared with other programs. +blob [5k\]: i'll keep that in mind, thanks +] +;B[jq]BL[12854.702]C[maproomad [-\]: pachi2: In 76190780 playouts at 64 machines, black K3 can win with 50.28% probability. +blob [5k\]: btw, any existing go bots that use machine learning out there? +uurtamo [-\]: yes, and they all get crushed by the monte carlo bots. which is why everyone is studying that now. +uurtamo [-\]: just teasing. +blob [5k\]: hah +] +;W[hp]WL[12601.969]C[uurtamo [-\]: there are bots that implement some kinds of machine learning, but they usually also have a very strong monte-carlo component. +blob [5k\]: that's good news for me +uurtamo [-\]: because otherwise they get absolutely crushed. +uurtamo [-\]: but you should ask on the computer go mailing list. there are lots of helpful ideas there and people there. +blob [5k\]: they probably rely on monte carlo so much because their board evaluation sucks +blob [5k\]: and have to play out very far +uurtamo [-\]: be careful with your estimate of other programs you haven't written. +] +;B[gp]BL[12750.975]C[uurtamo [-\]: the only thing you can be sure of is what you've already written. +blob [5k\]: ok :p +blob [5k\]: have you written a bot? +uurtamo [-\]: for instance, many faces of go, which is very strong, not uses some MC, but didn't before. +uurtamo [-\]: nope, i just started writing one. +uurtamo [-\]: *now uses +uurtamo [-\]: i've written a board evaluator, but never turned it into a bot. +blob [5k\]: evaluation is crucial though, so that should take you a long way +uurtamo [-\]: or not. +uurtamo [-\]: words like "should" require deep experience. +] +;W[iq]WL[12457.681]C[blob [5k\]: whatever, either agree or don't +] +;B[ir]BL[12647.293]C[goIngo [-\]: nick, can you give pachi %? +maproomad [-\]: The more people know about computer Go, the less they pontificate about it :) +maproomad [-\]: ok +maproomad [-\]: pachi2: In 130351958 playouts at 64 machines, black J2 can win with 51.04% probability. +goIngo [-\]: thx +klausP [3d?\]: i prefer black here +blob [5k\]: very few ppl "know" about computer go +] +;W[ip]WL[12313.346] +;B[gq]BL[12553.873]C[blob [5k\]: too slow for me.... good nigh +uurtamo [-\]: zen has a bye? +] +;W[lq]WL[12220.093]C[maproomad [-\]: yes - Zen has a bye at last. gghideki will use the time to isntall its new hardware +] +;B[lr]BL[12449.955]C[maproomad [-\]: pachi2: In 115171212 playouts at 64 machines, black M2 can win with 52.80% probability. +uurtamo [-\]: cool, new hardware. +] +;W[jr]WL[12076.655] +;B[kr]BL[12346.161] +;W[qh]WL[11934.434] +;B[lm]BL[12148.375] +;W[ln]WL[11793.194] +;B[mm]BL[11950.846] +;W[go]WL[11653.706]C[uurtamo [-\]: do you know what the hardware is? +] +;B[fo]BL[11785.708] +;W[km]WL[11513.825] +;B[kl]BL[11660.557] +;W[jm]WL[11373.958] +;B[ce]BL[11471.174]C[GoRoGoRo [9k\]: Zen still playing in its old clothes +] +;W[de]WL[11235.322] +;B[df]BL[11370.351] +;W[cf]WL[11098.126]C[goIngo [-\]: why to buy new ones, when the old fit? +] +;B[dd]BL[11269.657] +;W[ee]WL[10960.807] +;B[cd]BL[11184.929]C[maproomad [-\]: pachi2: In 178944910 playouts at 64 machines, black C16 can win with 56.76% probability. +] +;W[dg]WL[10855.87] +;B[ef]BL[11084.087] +;W[ff]WL[10719.349] +;B[eg]BL[10983.134]C[maproomad [-\]: e13 looks like a losing ladder +uurtamo [-\]: everyone wants to look stylin'. +maproomad [-\]: pachi2: In 126328940 playouts at 64 machines, black E13 can win with 57.39% probability. +] +;W[eh]WL[10584.286]C[uurtamo [-\]: oh, this could be a game ender. +uurtamo [-\]: i wonder how far into the ladder pachi2 will notice and tenuki. +] +;B[fg]BL[10862.716]C[maproomad [-\]: pachi2: In 148112137 playouts at 64 machines, black F13 can win with 55.81% probability. +uurtamo [-\]: i guess not yet, then. :) +] +;W[gg]WL[10448.687]C[ARPAnet [?\]: cool, lee sedol's ladder game :) +] +;B[fh]BL[10727.682]C[maproomad [-\]: pachi2: In 164712506 playouts at 64 machines, black F12 can win with 51.06% probability. +] +;W[fi]WL[10313.704] +;B[mo]BL[10544.165]C[maproomad [-\]: pachi2: In 170439444 playouts at 64 machines, black N5 can win with 44.51% probability. +uurtamo [-\]: doh! +] +;W[mn]WL[10180.199] +;B[nm]BL[10357.779] +;W[dc]WL[10057.33] +;B[gh]BL[10162.79]C[maproomad [-\]: pachi2: In 82988479 playouts at 64 machines, black G12 can win with 40.76% probability. +goIngo [-\]: hmm, pachi in downfall? +] +;W[hh]WL[9923.536]C[klausP [4d?\]: jup +] +;B[gn]BL[10004.306]C[maproomad [-\]: pachi2: In 122189851 playouts at 64 machines, black G6 can win with 38.62% probability. +] +;W[gi]WL[9793.004] +;B[ed]BL[9835.3] +;W[ec]WL[9667.99] +;B[fc]BL[9656.224] +;W[fd]WL[9543.048] +;B[gd]BL[9484.417]C[maproomad [-\]: pachi2: In 71631949 playouts at 64 machines, black G16 can win with 38.29% probability. +] +;W[fe]WL[9418.831] +;B[pg]BL[9335.446]C[klausP [3d?\]: poor pachi +] +;W[qe]WL[9293.787] +;B[qg]BL[9194.786] +;W[ph]WL[9171.43] +;B[oh]BL[8952.401] +;W[og]WL[9048.793] +;B[rh]BL[8814.474] +;W[ri]WL[8927.703] +;B[rf]BL[8681.749] +;W[oi]WL[8805.411] +;B[nh]BL[8551.044] +;W[ng]WL[8685.1] +;B[qi]BL[8422.3] +;W[pi]WL[8564.827] +;B[qj]BL[8323.697] +;W[sh]WL[8533.921] +;B[rg]BL[8198.341]C[maproomad [-\]: pachi2: In 169314007 playouts at 64 machines, black S13 can win with 42.44% probability. +] +;W[qk]WL[8451.774] +;B[pj]BL[8074.903]C[klausP [3d?\]: pachi fighting for a comeback, so is gomorra +maproomad [-\]: pachi2: In 142840643 playouts at 64 machines, black Q10 can win with 43.64% probability. +] +;W[rj]WL[8383.772] +;B[pk]BL[7955.361] +;W[hr]WL[8268.369] +;B[js]BL[7835.596]C[maproomad [-\]: pachi2: In 157538848 playouts at 64 machines, black K1 can win with 44.47% probability. +maproomad [-\]: creeping back towards 50% +] +;W[ni]WL[8150.466] +;B[mh]BL[7719.925]C[goIngo [-\]: unbelievable +] +;W[mi]WL[8035.124] +;B[lh]BL[7613.401] +;W[lg]WL[7944.056]C[goIngo [-\]: pachi %, please +] +;B[li]BL[7500.636]C[maproomad [-\]: pachi2: In 141731142 playouts at 64 machines, black M11 can win with 47.00% probability. +goIngo [-\]: thx +maproomad [-\]: if I don't res[pond, it's probably becasue I am not here +goIngo [-\]: ok +] +;W[ok]WL[7830.385]C[HYamashita [3k\]: pachi ran from ladder... if StoneGrid lose, 66th G11 is maybe lose move :-) +] +;B[pl]BL[7314.172]C[goIngo [-\]: KlausP also thought so +maproomad [-\]: pachi2: In 175825257 playouts at 64 machines, black Q8 can win with 48.13% probability. +] +;W[lj]WL[7716.828] +;B[kj]BL[7205.826] +;W[lk]WL[7604.161] +;B[oj]BL[7099.234]C[maproomad [-\]: pachi2: In 156331002 playouts at 64 machines, black P10 can win with 50.68% probability. +] +;W[mj]WL[7490.738] +;B[kk]BL[6994.124] +;W[re]WL[7408.672] +;B[pf]BL[6816.718] +;W[pe]WL[7298.225] +;B[of]BL[6715.798] +;W[ne]WL[7188.132] +;B[cl]BL[6616.812]C[maproomad [-\]: pachi2: In 87049581 playouts at 64 machines, black C8 can win with 54.53% probability. +] +;W[po]WL[7078.102]C[goIngo [-\]: wow, turning the table +maproomad [-\]: seems optimistic - W has a huge top, and B's central group is weak +] +;B[qo]BL[6432.123]C[maproomad [-\]: pachi2: In 136842147 playouts at 64 machines, black R5 can win with 54.94% probability. +] +;W[fr]WL[6970.775] +;B[gr]BL[6337.334] +;W[qp]WL[6863.532] +;B[pn]BL[6166.407] +;W[qq]WL[6758.457] +;B[oo]BL[6075.173] +;W[pr]WL[6653.682] +;B[oq]BL[5988.859] +;W[ro]WL[6548.985] +;B[jh]BL[5901.625] +;W[kg]WL[6446.203] +;B[rn]BL[5808.541] +;W[jg]WL[6343.276] +;B[ji]BL[5656.431] +;W[ik]WL[6242.372] +;B[jc]BL[5510.951] +;W[lc]WL[6140.836] +;B[fb]BL[5431.115] +;W[kh]WL[6041.661] +;B[ki]BL[5351.976] +;W[gc]WL[5942.788] +;B[hc]BL[5274.008] +;W[eb]WL[5844.746] +;B[gb]BL[5197.281] +;W[ib]WL[5747.462] +;B[jb]BL[5121.653] +;W[kb]WL[5651.079] +;B[kc]BL[4999.996] +;W[hd]WL[5553.721] +;B[he]BL[4882.043] +;W[kd]WL[5458.66] +;B[id]BL[4768.85] +;W[jd]WL[5365.841] +;B[lb]BL[4697.354] +;W[ic]WL[5273.326] +;B[ka]BL[4625.948] +;W[ia]WL[5180.722] +;B[mc]BL[4543.569] +;W[ld]WL[5089.09] +;B[hb]BL[4472.629] +;W[nb]WL[4999.019] +;B[nc]BL[4374.154] +;W[oc]WL[4909.481] +;B[od]BL[4276.269]C[maproomad [-\]: pachi2: In 45121745 playouts at 64 machines, black P16 can win with 81.12% probability. +maproomad [-\]: maybe it thinks the bottom left is dead +] +;W[ll]WL[4819.899] +;B[mk]BL[4182.18] +;W[oe]WL[4731.432]C[GoRoGoRo [9k\]: isn't it? +GoRoGoRo [9k\]: well ... W d2 may be a start? +maproomad [-\]: W d2? +GoRoGoRo [9k\]: thx +] +;B[ob]BL[4090.913] +;W[is]WL[4645.672] +;B[hs]BL[4021.198] +;W[pc]WL[4558.392] +;B[eq]BL[3931.121]C[maproomad [-\]: pachi2: In 66792603 playouts at 64 machines, black E3 can win with 83.32% probability. +] +;W[so]WL[4473.459] +;B[rq]BL[3841.046]C[maproomad [-\]: pachi2: In 53617646 playouts at 64 machines, black S3 can win with 84.08% probability. +] +;W[qr]WL[4389.52] +;B[rp]BL[3751.411] +;W[mp]WL[4306.146] +;B[rr]BL[3662.657]C[maproomad [-\]: pachi2: In 53837927 playouts at 64 machines, black S2 can win with 84.37% probability. +] +;W[ie]WL[4224.887] +;B[hd]BL[3595.694]C[maproomad [-\]: pachi2: In 112272634 playouts at 64 machines, black H16 can win with 82.43% probability. +] +;W[nq]WL[4143.842] +;B[np]BL[3529.095] +;W[qm]WL[4062.431] +;B[nn]BL[3441.868] +;W[qn]WL[3983.645] +;B[po]BL[3439.797] +;W[rm]WL[3905.712] +;B[rl]BL[3368.682] +;W[rs]WL[3828.818] +;B[bg]BL[3281.343] +;W[sr]WL[3751.145] +;B[hj]BL[3194.22] +;W[bh]WL[3673.258] +;B[bf]BL[3128.416] +;W[nk]WL[3597.054] +;B[ch]BL[3042.792] +;W[ml]WL[3523.668] +;B[nl]BL[2960.247]C[maproomad [-\]: pachi2: In 93360296 playouts at 64 machines, black O8 can win with 85.51% probability. +jlg [-\]: I should also report the dynamic komi. pachi is currently giving 26 points komi +] +;W[ci]WL[3448.2] +;B[bi]BL[2895.557] +;W[ai]WL[3375.882] +;B[di]BL[2814.737] +;W[bd]WL[3305.694] +;B[il]BL[2734.352] +;W[fl]WL[3232.07] +;B[or]BL[2655.689] +;W[fm]WL[3159.724]C[Winnetou [?\]: The moves are very strange. After pachi lost the ladder the game was terrible. +] +;B[ps]BL[2579.106]C[jlg [-\]: yes this is a known ladder bug indeed awful. But now pachi is trying to maximize the score (komi 35 points) +] +;W[jk]WL[3088.298]C[Winnetou [?\]: It seems that it is not pachi who maximizes the win but Stonegrid who maximizes the loss ;) +] +;B[be]BL[2503.785]C[maproomad [-\]: pachi2: In 93549867 playouts at 64 machines, black B15 can win with 89.42% probability. +] +;W[na]WL[3018.609]C[reticent [5k\]: go seigen would either love or shake his head at the fuseki in this game +reticent [5k\]: not nearly strong enough to know which +] +;B[ho]BL[2432.144]C[reticent [5k\]: and what the hell was the ladder pachi2 played out? +jlg [-\]: known ladder bug +reticent [5k\]: right +reticent [5k\]: anyway, looks like pachi2 spotted stonegrid fifty points and went on to win anyway +] +;W[dp]WL[2948.142] +;B[md]BL[2362.29] +;W[nd]WL[2879.078] +;B[bc]BL[2293.487]C[jlg [-\]: 92%, komi 54 +] +;W[bb]WL[2813.878] +;B[nj]BL[2291.855] +;W[mb]WL[2747.406] +;B[ql]BL[2222.578] +;W[sn]WL[2682.776] +;B[mk]BL[2221.186]C[salty [2d\]: i hope w isn`t planing to win that ^ +]) -- 2.11.4.GIT