Winboard Forum

by **Jose Carlos** » 08 Apr 2003, 09:34

Geschrieben von: / Posted by: Jose Carlos at 08 April 2003 10:34:14:

Hi, first I must say this is partially off-topic, since some of the engines of the test are UCI, and Arena was used. If the moderators consider I shouldn't post this stuff here, I'll stop. But I think it's interesting.
I'm testing the endgame eval of my new program, Anubis.
I called a friend of mine (>2400 FIDE) and asked him to choose a complex endgame position with chances for both sides. We're starting this test with a pawn ending. I have no access to the "secret" position so that I can't prepare. He ran a tournament at 3 1 in Arena with some free programs (some UCI and some WB) plus Deep Sjeng, which he bought and wanted to test, starting from the "secret" position. The results of the tournament are below.
Disclaimer: this is not intented to prove anything; this is not intended to be statistically significant; this is not even intended to give a clue on which program has better pawn endgame eval. I find it useful for me, but that's all.
Note that not all the programs have the same number of games. Some games were lost on time without making a single move; since several programs had this problem, maybe it's a bug in arena... ?!

Program Elo + - Games Score Av.Op. Draws
1 Little Goliath 2000 v3.9 : 2489 113 34 31 62.9 % 2397 74.2 %
2 Delfi 4.1 : 2485 113 49 31 62.9 % 2393 67.7 %
3 Anubis v0.00 : 2447 124 34 30 56.7 % 2400 80.0 %
4 Abrok 4.0 : 2441 120 31 32 56.2 % 2398 81.2 %
5 Tao 5.4 : 2440 122 44 31 56.5 % 2395 74.2 %
6 Ruffian 1.0.1 : 2434 125 29 31 54.8 % 2401 83.9 %
7 SOS.2 for Arena : 2434 125 29 31 54.8 % 2401 83.9 %
8 AnMon5.21 : 2431 123 56 32 54.7 % 2398 65.6 %
9 Yace Paderborn : 2431 123 12 32 54.7 % 2398 90.6 %
10 Deep Sjeng 1.0 : 2392 59 130 31 48.4 % 2403 64.5 %
11 Aristarch 4.4 : 2392 52 130 31 48.4 % 2403 71.0 %
12 KnightDreamer v3.1b : 2388 59 130 31 48.4 % 2399 64.5 %
13 Amyan 1.59 : 2380 60 125 32 46.9 % 2401 62.5 %
14 Hagrid 0.7.56 : 2317 57 111 32 37.5 % 2405 62.5 %
15 Green Light Chess v3.00 : 2312 65 121 28 37.5 % 2401 60.7 %
16 Amateur 2.4 : 2310 85 114 30 36.7 % 2405 46.7 %
17 Averno v0.51beta (051) : 2271 52 104 32 31.2 % 2408 62.5 %

We plan to extend this further, to rook endgames, queen endgames, etc. but I don't know when I'll be able to post more info.
José C.

by **Tim Foden** » 08 Apr 2003, 09:47

Geschrieben von: / Posted by: Tim Foden at 08 April 2003 10:47:26:
Als Antwort auf: / In reply to: Endgame test series geschrieben von: / posted by: Jose Carlos at 08 April 2003 10:34:14:

Disclaimer: this is not intented to prove anything; this is not intended to be statistically significant; this is not even intended to give a clue on which program has better pawn endgame eval. I find it useful for me, but that's all.

It looks interesting, and helps to confirm my feeling that GLC is not so good in pawn endgames.
It would help me if you could actually post the position that was used for these games, or even the games themselves?
Cheers, Tim.

Green Light Chess Site

by **Jose Carlos** » 08 Apr 2003, 09:53

Geschrieben von: / Posted by: Jose Carlos at 08 April 2003 10:53:59:
Als Antwort auf: / In reply to: Re: Endgame test series geschrieben von: / posted by: Tim Foden at 08 April 2003 10:47:26:

Disclaimer: this is not intented to prove anything; this is not intended to be statistically significant; this is not even intended to give a clue on which program has better pawn endgame eval. I find it useful for me, but that's all.
It looks interesting, and helps to confirm my feeling that GLC is not so good in pawn endgames.
It would help me if you could actually post the position that was used for these games, or even the games themselves?
Cheers, Tim.

I don't know the position myself. I've preferred not to know it in order to improve my program with independence. If I knew the position, I could fall in the temptation... Anyway, I'll ask my friend to join winboard forum and possibly post his positions after all tests are finished.
José C.

by **Jose Carlos** » 08 Apr 2003, 11:38

Geschrieben von: / Posted by: Jose Carlos at 08 April 2003 12:38:05:
Als Antwort auf: / In reply to: Re: Endgame test series geschrieben von: / posted by: Jose Carlos at 08 April 2003 10:53:59:

Disclaimer: this is not intented to prove anything; this is not intended to be statistically significant; this is not even intended to give a clue on which program has better pawn endgame eval. I find it useful for me, but that's all.
It looks interesting, and helps to confirm my feeling that GLC is not so good in pawn endgames.
It would help me if you could actually post the position that was used for these games, or even the games themselves?
Cheers, Tim.
I don't know the position myself. I've preferred not to know it in order to improve my program with independence. If I knew the position, I could fall in the temptation... Anyway, I'll ask my friend to join winboard forum and possibly post his positions after all tests are finished.
José C.

One more point: engines EGTB capable are using them. My friend's using the tb's included in the Deep Sjeng CD. I don't know for sure, but I guess 3-4 men. The idea is to test the programs under the conditions they'll actually face each other in real games.
Averno is EGTB capable. Anubis isn't.
José C.

by **Lyapko George** » 08 Apr 2003, 11:49

Geschrieben von: / Posted by: Lyapko George at 08 April 2003 12:49:35:
Als Antwort auf: / In reply to: Re: Endgame test series geschrieben von: / posted by: Jose Carlos at 08 April 2003 12:38:05:

Disclaimer: this is not intented to prove anything; this is not intended to be statistically significant; this is not even intended to give a clue on which program has better pawn endgame eval. I find it useful for me, but that's all.
It looks interesting, and helps to confirm my feeling that GLC is not so good in pawn endgames.
It would help me if you could actually post the position that was used for these games, or even the games themselves?
Cheers, Tim.
I don't know the position myself. I've preferred not to know it in order to improve my program with independence. If I knew the position, I could fall in the temptation... Anyway, I'll ask my friend to join winboard forum and possibly post his positions after all tests are finished.
José C.
One more point: engines EGTB capable are using them. My friend's using the tb's included in the Deep Sjeng CD. I don't know for sure, but I guess 3-4 men. The idea is to test the programs under the conditions they'll actually face each other in real games.
Averno is EGTB capable. Anubis isn't.
José C.

What about adding Bestia to your tests?
I know it is not that strong in middlegame, but in endgame it has a chances...
Best regards,
George

by **Jose Carlos** » 08 Apr 2003, 12:05

Geschrieben von: / Posted by: Jose Carlos at 08 April 2003 13:05:24:
Als Antwort auf: / In reply to: Re: Endgame test series geschrieben von: / posted by: Lyapko George at 08 April 2003 12:49:35:

Disclaimer: this is not intented to prove anything; this is not intended to be statistically significant; this is not even intended to give a clue on which program has better pawn endgame eval. I find it useful for me, but that's all.
It looks interesting, and helps to confirm my feeling that GLC is not so good in pawn endgames.
It would help me if you could actually post the position that was used for these games, or even the games themselves?
Cheers, Tim.
I don't know the position myself. I've preferred not to know it in order to improve my program with independence. If I knew the position, I could fall in the temptation... Anyway, I'll ask my friend to join winboard forum and possibly post his positions after all tests are finished.
José C.
One more point: engines EGTB capable are using them. My friend's using the tb's included in the Deep Sjeng CD. I don't know for sure, but I guess 3-4 men. The idea is to test the programs under the conditions they'll actually face each other in real games.
Averno is EGTB capable. Anubis isn't.
José C.
What about adding Bestia to your tests?
I know it is not that strong in middlegame, but in endgame it has a chances...
Best regards,
George

Ok, I'll add Bestia. Do you want me to download the public version or do you want to test some other version? Is it stable under Arena?
José C.

by **Lyapko George** » 08 Apr 2003, 12:09

Geschrieben von: / Posted by: Lyapko George at 08 April 2003 13:09:58:
Als Antwort auf: / In reply to: Re: Endgame test series geschrieben von: / posted by: Jose Carlos at 08 April 2003 13:05:24:

Disclaimer: this is not intented to prove anything; this is not intended to be statistically significant; this is not even intended to give a clue on which program has better pawn endgame eval. I find it useful for me, but that's all.
It looks interesting, and helps to confirm my feeling that GLC is not so good in pawn endgames.
It would help me if you could actually post the position that was used for these games, or even the games themselves?
Cheers, Tim.
I don't know the position myself. I've preferred not to know it in order to improve my program with independence. If I knew the position, I could fall in the temptation... Anyway, I'll ask my friend to join winboard forum and possibly post his positions after all tests are finished.
José C.
One more point: engines EGTB capable are using them. My friend's using the tb's included in the Deep Sjeng CD. I don't know for sure, but I guess 3-4 men. The idea is to test the programs under the conditions they'll actually face each other in real games.
Averno is EGTB capable. Anubis isn't.
José C.
What about adding Bestia to your tests?
I know it is not that strong in middlegame, but in endgame it has a chances...
Best regards,
George
Ok, I'll add Bestia. Do you want me to download the public version or do you want to test some other version? Is it stable under Arena?
José C.

Latest public version (0.90) is ok. I have never heard about problems with Bestia (since 0.8) under any GUI.
LG

by **Jose Carlos** » 08 Apr 2003, 15:47

Geschrieben von: / Posted by: Jose Carlos at 08 April 2003 16:47:10:
Als Antwort auf: / In reply to: Re: Endgame test series geschrieben von: / posted by: Lyapko George at 08 April 2003 13:09:58:

Disclaimer: this is not intented to prove anything; this is not intended to be statistically significant; this is not even intended to give a clue on which program has better pawn endgame eval. I find it useful for me, but that's all.
It looks interesting, and helps to confirm my feeling that GLC is not so good in pawn endgames.
It would help me if you could actually post the position that was used for these games, or even the games themselves?
Cheers, Tim.
I don't know the position myself. I've preferred not to know it in order to improve my program with independence. If I knew the position, I could fall in the temptation... Anyway, I'll ask my friend to join winboard forum and possibly post his positions after all tests are finished.
José C.
One more point: engines EGTB capable are using them. My friend's using the tb's included in the Deep Sjeng CD. I don't know for sure, but I guess 3-4 men. The idea is to test the programs under the conditions they'll actually face each other in real games.
Averno is EGTB capable. Anubis isn't.
José C.
What about adding Bestia to your tests?
I know it is not that strong in middlegame, but in endgame it has a chances...
Best regards,
George
Ok, I'll add Bestia. Do you want me to download the public version or do you want to test some other version? Is it stable under Arena?
José C.
Latest public version (0.90) is ok. I have never heard about problems with Bestia (since 0.8) under any GUI.
LG

First test. I used a random position. It plays fine with white, but fails with black. Here's Arena's log (1 is Bestia, 2 is Anubis):
1676180)1:level 0 3 1
1676180)1:new
1676210)1:force
1676210)1:setboard 4k3/8/p1p1p1p1/1p1p1p1p/1P1P1P1P/P1P1P1P1/8/4K3 w - -
1676210)2:level 0 3 1
1676210)2:new
1676260)2:force
1676260)2:setboard 4k3/8/p1p1p1p1/1p1p1p1p/1P1P1P1P/P1P1P1P1/8/4K3 w - -
1683921*2*Start calc, move no: 0
1684191)2:time 18000
1684191)2:otim 18000
1684191)2:go
1684362(2:1 -265 0 2 a3a4? (0 knps)
1684362(2:1 -265 0 3 a3a4?? (0 knps)
1684362(2:1 -265 0 4 a3a4 b5a4 (0 knps)
1684372(2:1 -130 0 7 c3c4 d5c4 (0 knps)
1684372(2:1 3 0 12 e1d2 (1 knps)
1684372(2:1 3 0 16 e1d2 (2 knps)
1684372(2:2 0 0 30 e1d2 e8d7 (3 knps)
1684372(2:2 0 0 55 e1d2 e8d7 (6 knps)
1684382(2:3 4 0 95 e1d2 e8d7 d2d3 (5 knps)
1684382(2:3 4 0 144 e1d2 e8d7 d2d3 (7 knps)
1684382(2:4 0 0 296 e1d2 e8d7 d2d3 d7d6 (15 knps)
1684392(2:4 0 0 452 e1d2 e8d7 d2d3 d7d6 (15 knps)
1684392(2:5 0 0 929 e1d2 e8d7 d2c2 d7d6 c2d3 (31 knps)
1684402(2:5 0 0 1346 e1d2 e8d7 d2c2 d7d6 c2d3 (45 knps)
1684402(2:6 0 0 2796 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (70 knps)
1684412(2:6 0 0 3634 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (73 knps)
1684422(2:7 0 0 6025 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (100 knps)
1684432(2:7 0 0 8094 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (116 knps)
1684452(2:8 0 0 13953 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (155 knps)
1684472(2:8 0 0 17906 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (163 knps)
1684522(2:9 0 0 32482 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (203 knps)
1684552(2:9 0 0 43083 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (227 knps)
1684682(2:10 0 0 82245 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (257 knps)
1684752(2:10 0 0 102481 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (270 knps)
1685083(2:11 0 0 195604 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 a3a4 (271 knps)
1685263(2:11 0 0 250689 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 a3a4 (281 knps)
1685974(2:12 0 100 456735 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (283 knps)
1686415(2:12 0 200 577305 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (281 knps)
1688097(2:13 0 300 1054279 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (282 knps)
1689209(2:13 0 400 1368019 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (282 knps)
1691572(2:move e1d2
1691572*2*Found move:Ke1-d2
1694136*1*Start calc, move no: 1
1694406)1:e1d2
1694426(1:fifty=0
1694426(1:Error (unknown command): e1d2
1694446)1:time 18100
1694446)1:otim 17363
1694446)1:black
1694446)1:go
1694486(1:inbook
1694486(1:ponder=FALSE
1694486(1:Static eval=0
1694486(1:-Failing low
1694486(1: 2 27 0 50 e8d7 e1d2
1694486(1: 2 0 0 77 e8d7 e1d2
1694486(1:+Failing high
1694516(1: 3 25 0 112 e8d7
1694516(1: 3 52 0 232 e8d7 e1d2 d7d6
1694516(1:-Failing low
1694516(1: 4 27 0 412 e8d7 e1d2
1694516(1: 4 0 1 684 e8d7 e1d2 d7d6 d2d3
1694526(1: 5 0 1 1034 e8d7 e1d2 d7c7 d2d3 c7d6
1694526(1: 5 0 1 1481 e8d7 e1d2 d7c7 d2d3 c7d6
1694556(1: 6 -18 2 2995 e8d7 e1d2 d7e7 d2e2 e7f6 e2d3
1694556(1: 6 -18 5 4758 e8d7 e1d2 d7e7 d2e2 e7f6 e2d3
1694556(1: 7 7 9 7373 e8d7
1694556(1:+Failing high
1694566(1: 7 7 9 7373 e8d7
1694566(1: 7 16 10 9228 e8d7 e1d2 d7c7 d2e2 c7b6 e2d3 a6a5
1694576(1: 7 16 13 11372 e8d7 e1d2 d7c7 d2e2 c7b6 e2d3 a6a5
1694696(1: 8 0 25 16352 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4
1694727(1: 8 0 28 19576 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4
1694807(1: 9 0 36 28889 e8d7 e1d2 d7c7 d2c2 c7b7 c2b3 b7b6 a3a4 a6a5
1694847(1: 9 0 40 35195 e8d7 e1d2 d7c7 d2c2 c7b7 c2b3 b7b6 a3a4 a6a5
1695057(1: 10 12 61 60953 e8d7 e1d2 d7c7 d2c2 c7b7 c2b3 b7b6 a3a4 a6a5 a4b5
1695127(1: 10 12 68 70081 e8d7 e1d2 d7c7 d2c2 c7b7 c2b3 b7b6 a3a4 a6a5 a4b5
1695458(1: 11 10 101 117813 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 a4b5 a6b5 b4a5
1695588(1: 11 10 113 135645 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 a4b5 a6b5 b4a5
1696109(1: 12 0 166 213284 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 b3a3 a6b6 a3b3
1696339(1: 12 0 189 243605 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 b3a3 a6b6 a3b3
1697040(1: 13 0 259 349575 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 b3a3 a6b6 a3b3
1697360(1: 13 0 291 397721 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 b3a3 a6b6 a3b3
1698662(1: 14 25 422 592001 e8d7
1698672(1:+Failing high
1698672(1: 14 25 423 592001 e8d7
1700224(1: 14 28 578 827004 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 a5b4 a4b5 b4c3 b5c6 c3c2 b3c2
1700715(1: 14 28 627 901021 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 a5b4 a4b5 b4c3 b5c6 c3c2 b3c2
1703509(1: 15 42 906 1321146 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3c2 a5b6 c2b2 c6c5 b2b3 b6c6 a3a4
1704290(1: 15 42 984 1439047 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3c2 a5b6 c2b2 c6c5 b2b3 b6c6 a3a4
1708707(1: 16 28 1426 2125635 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3b2 a5b6 b2a2 b6a7 a3a4 a7a6 a4b5 a6b5 a2b3
1710179(1: 16 28 1573 2347564 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3b2 a5b6 b2a2 b6a7 a3a4 a7a6 a4b5 a6b5 a2b3
1716768(1: 17 28 2232 3321725 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3b2 a5b6 b2a2 b6a7 a2b3
1719142(1: 17 28 2470 3677191 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3b2 a5b6 b2a2 b6a7 a2b3
1746080(1: 18 4 5163 7592499 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 a5b4 a4b5 b6b5 c3b4 b5b6 b3c2 b6c7 c2d2 c7d6 d2d3
1794009)1:?
1794019(1:move e8d7
1794019*1*Found move:Ke8-d7
1794109(1:Error (unknown command): ?
I replaced < for ( and > for )
As you see, it seems that it doesn't recognize the opponent's move. Any clue?
José C.

by **Lyapko George** » 09 Apr 2003, 06:24

Geschrieben von: / Posted by: Lyapko George at 09 April 2003 07:24:01:
Als Antwort auf: / In reply to: Re: Endgame test series geschrieben von: / posted by: Jose Carlos at 08 April 2003 16:47:10:

Disclaimer: this is not intented to prove anything; this is not intended to be statistically significant; this is not even intended to give a clue on which program has better pawn endgame eval. I find it useful for me, but that's all.
It looks interesting, and helps to confirm my feeling that GLC is not so good in pawn endgames.
It would help me if you could actually post the position that was used for these games, or even the games themselves?
Cheers, Tim.
I don't know the position myself. I've preferred not to know it in order to improve my program with independence. If I knew the position, I could fall in the temptation... Anyway, I'll ask my friend to join winboard forum and possibly post his positions after all tests are finished.
José C.
One more point: engines EGTB capable are using them. My friend's using the tb's included in the Deep Sjeng CD. I don't know for sure, but I guess 3-4 men. The idea is to test the programs under the conditions they'll actually face each other in real games.
Averno is EGTB capable. Anubis isn't.
José C.
What about adding Bestia to your tests?
I know it is not that strong in middlegame, but in endgame it has a chances...
Best regards,
George
Ok, I'll add Bestia. Do you want me to download the public version or do you want to test some other version? Is it stable under Arena?
José C.
Latest public version (0.90) is ok. I have never heard about problems with Bestia (since 0.8) under any GUI.
LG
First test. I used a random position. It plays fine with white, but fails with black. Here's Arena's log (1 is Bestia, 2 is Anubis):
1676180)1:level 0 3 1
1676180)1:new
1676210)1:force
1676210)1:setboard 4k3/8/p1p1p1p1/1p1p1p1p/1P1P1P1P/P1P1P1P1/8/4K3 w - -
1676210)2:level 0 3 1
1676210)2:new
1676260)2:force
1676260)2:setboard 4k3/8/p1p1p1p1/1p1p1p1p/1P1P1P1P/P1P1P1P1/8/4K3 w - -
1683921*2*Start calc, move no: 0
1684191)2:time 18000
1684191)2:otim 18000
1684191)2:go
1684362(2:1 -265 0 2 a3a4? (0 knps)
1684362(2:1 -265 0 3 a3a4?? (0 knps)
1684362(2:1 -265 0 4 a3a4 b5a4 (0 knps)
1684372(2:1 -130 0 7 c3c4 d5c4 (0 knps)
1684372(2:1 3 0 12 e1d2 (1 knps)
1684372(2:1 3 0 16 e1d2 (2 knps)
1684372(2:2 0 0 30 e1d2 e8d7 (3 knps)
1684372(2:2 0 0 55 e1d2 e8d7 (6 knps)
1684382(2:3 4 0 95 e1d2 e8d7 d2d3 (5 knps)
1684382(2:3 4 0 144 e1d2 e8d7 d2d3 (7 knps)
1684382(2:4 0 0 296 e1d2 e8d7 d2d3 d7d6 (15 knps)
1684392(2:4 0 0 452 e1d2 e8d7 d2d3 d7d6 (15 knps)
1684392(2:5 0 0 929 e1d2 e8d7 d2c2 d7d6 c2d3 (31 knps)
1684402(2:5 0 0 1346 e1d2 e8d7 d2c2 d7d6 c2d3 (45 knps)
1684402(2:6 0 0 2796 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (70 knps)
1684412(2:6 0 0 3634 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (73 knps)
1684422(2:7 0 0 6025 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (100 knps)
1684432(2:7 0 0 8094 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (116 knps)
1684452(2:8 0 0 13953 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (155 knps)
1684472(2:8 0 0 17906 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (163 knps)
1684522(2:9 0 0 32482 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (203 knps)
1684552(2:9 0 0 43083 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (227 knps)
1684682(2:10 0 0 82245 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (257 knps)
1684752(2:10 0 0 102481 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (270 knps)
1685083(2:11 0 0 195604 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 a3a4 (271 knps)
1685263(2:11 0 0 250689 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 a3a4 (281 knps)
1685974(2:12 0 100 456735 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (283 knps)
1686415(2:12 0 200 577305 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (281 knps)
1688097(2:13 0 300 1054279 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (282 knps)
1689209(2:13 0 400 1368019 e1d2 e8d7 d2d3 d7d6 d3d2 d6d7 (282 knps)
1691572(2:move e1d2
1691572*2*Found move:Ke1-d2
1694136*1*Start calc, move no: 1
1694406)1:e1d2
1694426(1:fifty=0
1694426(1:Error (unknown command): e1d2
1694446)1:time 18100
1694446)1:otim 17363
1694446)1:black
1694446)1:go
1694486(1:inbook
1694486(1:ponder=FALSE
1694486(1:Static eval=0
1694486(1:-Failing low
1694486(1: 2 27 0 50 e8d7 e1d2
1694486(1: 2 0 0 77 e8d7 e1d2
1694486(1:+Failing high
1694516(1: 3 25 0 112 e8d7
1694516(1: 3 52 0 232 e8d7 e1d2 d7d6
1694516(1:-Failing low
1694516(1: 4 27 0 412 e8d7 e1d2
1694516(1: 4 0 1 684 e8d7 e1d2 d7d6 d2d3
1694526(1: 5 0 1 1034 e8d7 e1d2 d7c7 d2d3 c7d6
1694526(1: 5 0 1 1481 e8d7 e1d2 d7c7 d2d3 c7d6
1694556(1: 6 -18 2 2995 e8d7 e1d2 d7e7 d2e2 e7f6 e2d3
1694556(1: 6 -18 5 4758 e8d7 e1d2 d7e7 d2e2 e7f6 e2d3
1694556(1: 7 7 9 7373 e8d7
1694556(1:+Failing high
1694566(1: 7 7 9 7373 e8d7
1694566(1: 7 16 10 9228 e8d7 e1d2 d7c7 d2e2 c7b6 e2d3 a6a5
1694576(1: 7 16 13 11372 e8d7 e1d2 d7c7 d2e2 c7b6 e2d3 a6a5
1694696(1: 8 0 25 16352 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4
1694727(1: 8 0 28 19576 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4
1694807(1: 9 0 36 28889 e8d7 e1d2 d7c7 d2c2 c7b7 c2b3 b7b6 a3a4 a6a5
1694847(1: 9 0 40 35195 e8d7 e1d2 d7c7 d2c2 c7b7 c2b3 b7b6 a3a4 a6a5
1695057(1: 10 12 61 60953 e8d7 e1d2 d7c7 d2c2 c7b7 c2b3 b7b6 a3a4 a6a5 a4b5
1695127(1: 10 12 68 70081 e8d7 e1d2 d7c7 d2c2 c7b7 c2b3 b7b6 a3a4 a6a5 a4b5
1695458(1: 11 10 101 117813 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 a4b5 a6b5 b4a5
1695588(1: 11 10 113 135645 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 a4b5 a6b5 b4a5
1696109(1: 12 0 166 213284 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 b3a3 a6b6 a3b3
1696339(1: 12 0 189 243605 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 b3a3 a6b6 a3b3
1697040(1: 13 0 259 349575 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 b3a3 a6b6 a3b3
1697360(1: 13 0 291 397721 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 b6a6 b3a3 a6b6 a3b3
1698662(1: 14 25 422 592001 e8d7
1698672(1:+Failing high
1698672(1: 14 25 423 592001 e8d7
1700224(1: 14 28 578 827004 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 a5b4 a4b5 b4c3 b5c6 c3c2 b3c2
1700715(1: 14 28 627 901021 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 a5b4 a4b5 b4c3 b5c6 c3c2 b3c2
1703509(1: 15 42 906 1321146 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3c2 a5b6 c2b2 c6c5 b2b3 b6c6 a3a4
1704290(1: 15 42 984 1439047 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3c2 a5b6 c2b2 c6c5 b2b3 b6c6 a3a4
1708707(1: 16 28 1426 2125635 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3b2 a5b6 b2a2 b6a7 a3a4 a7a6 a4b5 a6b5 a2b3
1710179(1: 16 28 1573 2347564 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3b2 a5b6 b2a2 b6a7 a3a4 a7a6 a4b5 a6b5 a2b3
1716768(1: 17 28 2232 3321725 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3b2 a5b6 b2a2 b6a7 a2b3
1719142(1: 17 28 2470 3677191 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 b4a5 b6a5 b3b2 a5b6 b2a2 b6a7 a2b3
1746080(1: 18 4 5163 7592499 e8d7 e1d2 d7c7 d2c2 c7b6 c2b3 a6a5 a3a4 a5b4 a4b5 b6b5 c3b4 b5b6 b3c2 b6c7 c2d2 c7d6 d2d3
1794009)1:?
1794019(1:move e8d7
1794019*1*Found move:Ke8-d7
1794109(1:Error (unknown command): ?
I replaced < for ( and > for )
As you see, it seems that it doesn't recognize the opponent's move. Any clue?
José C.

Hi Jose,
it is a bug(Bestia don't generate moves after 'setboard').
I could send you a fix by mail.
Best regards,
George

by **Jose Carlos** » 09 Apr 2003, 07:48

Geschrieben von: / Posted by: Jose Carlos at 09 April 2003 08:48:32:
Als Antwort auf: / In reply to: Re: Endgame test series geschrieben von: / posted by: Lyapko George at 09 April 2003 07:24:01:

Hi Jose,
it is a bug(Bestia don't generate moves after 'setboard').
I could send you a fix by mail.
Best regards,
George

If you send that mail, I'll inmclude Bestia in the test.
smyslov[a]hotmail.com
José C.

Winboard Forum

Who is online