1 A list of games to put in the regression suite.
3 This list can be processed automatically:
5 sed -ne 's/^\* //p' <TODO | while read r c; do echo wget -O $(echo "$r" | sed 's#.*games/##' | tr / - | sed 's/-\([^-]\)-/-0\1-/g;s/-\([^-]\)-/-0\1-/') "$r"; done
7 * http://files.gokgs.com/games/2011/6/5/pachi2-SimpleBot.sgf ladder
8 pachi shouldn't play the broken ladders up to move 53
10 * http://files.gokgs.com/games/2011/5/29/snowmen-pachi2.sgf
11 169 B5 seems to be the losing move. This move doesn't make sense.
12 Pachi has been wrongly evaluating both bottom group around S4 (it's just
13 hairy for simulations) and top group around A19 - that one is a bit more
14 interesting, the problem is that black is double likely to play E19
15 (since it's C19 selfatari cousin), but after E19, white will not notice
16 the need to connect at C19. Not sure how to solve this yet, perhaps some
17 way to add cousins to last_move logic...
19 * http://files.gokgs.com/games/2011/5/28/pachi2-Rubel.sgf
20 won game lost: 212 H13 should be at J12.
22 * http://files.gokgs.com/games/2011/5/27/ManyFaces1-pachi2.sgf
23 had won game, move 135 F1 very bad. at end doesn't see ko at t7. (b t10
26 * http://files.gokgs.com/games/2011/5/23/pachi2-ManyFaces1.sgf
27 leading up to 90% then lost.
28 bad evaluation of 3 areas according to pasky.
30 * http://files.gokgs.com/games/2011/5/25/pachi2-Zen19S.sgf
31 lost but was leading according to ajahuang and Zen itself
33 * http://files.gokgs.com/games/2011/2/20/ramesses2-pachi2.sgf
34 good example of losing a won game.
36 * http://files.gokgs.com/games/2011/2/18/pachi2-tf-3.sgf seki
37 lost because of unrecognized seki
39 * http://files.gokgs.com/games/2011/6/9/jinen-pachi30s-2.sgf ko
40 move 165 G5 again playing the ko threat before the ko.
41 move 187 R18 is not a ko threat
43 At this point pachi still considers the top right as
44 unsetled, although it is is clearly lost for B. So pachi's
45 estimate of winning by 61% is way too optimistic.
47 * http://files.gokgs.com/games/2011/6/14/pachi2-Regenbogen.sgf
48 pachi doesn't see that bent four in top left corner is dead.
49 Still thinks it is winning 79% at move 250 G4.
51 * http://files.gokgs.com/games/2011/6/14/pachi2-cdocdo-3.sgf semeai
52 pachi loses every semai by a large number of liberties
53 but still considers itself winning 57% up to move 222 G16.
55 * http://files.gokgs.com/games/2011/6/18/dorabon-pachi2-4.sgf semeai falseeye
56 pachi considers the bottom left semeai won by black (but black
57 loses it) and the bottom right as undecided (white is clearly
58 alive). At move 270 L4 it still thinks it is leading by 80%
59 although the game is lost.
60 The bottom right is the same problem as
61 http://files.gokgs.com/games/2011/6/10/pachi30s-samba-2.sgf
62 and http://files.gokgs.com/games/2011/6/9/botkiller2-pachi30s.sgf
63 This corner shape is very frequent and pachi always gets it wrong.
65 * http://files.gokgs.com/games/2011/6/18/pachi2-ultimakang-4.sgf
66 pachi was winning up to move 230 Q14 big mistake. Win ratio dropped
67 from 60% to 25% at next move.
69 * http://files.gokgs.com/games/2011/6/18/akkannbee-pachi2.sgf seki
70 pachi totally confused by very simple seki.
72 * http://files.gokgs.com/games/2011/7/6/xiaosugi-pachi2.sgf seki
73 pachi thinks until the end that it wins by 80% because it
74 doesn't understand the seki in lower right. It should have lost
75 by 42.5 points but white mistakenly agreed to treat the seki as dead.
77 * http://files.gokgs.com/games/2011/7/5/pachi2-azlan7-2.sgf
78 move 84 E10 & 86 F10 big mistakes, leave white group dead.
79 (amazingly pachi recovered and azlan7 escaped in the end).
81 * http://files.gokgs.com/games/2011/7/5/fafahhhh-pachi2.sgf
82 threw away won game, probably because of complex corner situations
83 won by time but game was lost on board
85 * http://files.gokgs.com/games/2011/7/3/pachi2-oasis7.sgf
86 Passed even though a dead opponent group was marked alive.
88 * http://files.gokgs.com/games/2011/7/28/pachi2-pkunzip-2.sgf ladder
89 W 132 C9, W 134 B8, W 136 A8 plays a broken ladder but win ratio 52%
90 so it should not play desperate moves.
91 Pachi thought it was winning 72% at W 198 M3 but lost the game.
93 * http://files.gokgs.com/games/2011/7/16/ihatebots-pachi2.sgf semeai
94 pachi threw away the game because it didn't evaluate the
95 semeai in bottom left correctly.
97 * http://files.gokgs.com/games/2011/7/31/pachi2-Drogba987.sgf
98 B 28 H4 should be at F1 or G1 to save E2 & E3 and the game.
99 Pachi usually finds this so I'm surprised it didn't get it here.
101 * http://files.gokgs.com/games/2011/8/9/pachi2-BlueSpark.sgf ladder
102 move 150 R19: pachi extends a losing ladder
104 * http://files.gokgs.com/games/2011/8/8/Dallas-pachi2-2.sgf ladder
105 move 121 L19: pachi extends a losing ladder
107 * http://files.gokgs.com/games/2011/8/9/somrak-pachi2-2.sgf ladder
108 move 143 N19: pachi extends a losing ladder
109 This ladder bug seems more frequent since e9df2b6 (even though
110 this change made Pachi 40 elo stronger).
112 * http://files.gokgs.com/games/2011/8/10/jerryfish-pachi2.sgf semeai
113 lost even game against 3k. Win ratio drops from 61% at 117 N1
114 to 49% at the next move. So pachi didn't evaluate the semai correctly.