ColChess 5.5 steadily advances.

[Dann Corbit]

Geschrieben von:/Posted by: Dann Corbit at 18 February 2000 21:36:41:
Each version is clearly stronger than the previous iteration. Here is 5.5 compiled for Winboard:
ftp://38.168.214.175/pub/COLCHESS.ZIP
I now get 200/300 on WAC on a lowly PII 300 MHz machine:
Test #300 : White to play
Current Position
----------------
b . . b . r . k
. . . R . p p p pP=Pawn
. . . . q P . . rR=Rook
. . . . p . P Q nN=Knight lower=Black
. . . . P . . . bB=Bishop CAPS=White
. . . . . B . . qQ=Queen
. . . . N . K . kK=King
. . . . . . . .
Thinking 2 355 2 1761 g5g6 h7h6
3 28 12 5114 g5g6 h7h6 d7f7
4 180 61 41407 g5g6 h7h6 f6g7 h8g7
5 88 351 171985 d7a7 d8f6 g5f6 e6f6 a7d7
6 88 501 258579
---Answer: g6 ~Incorrect-??
Correct So Far: 200[49]/300
Result: 200[49]/300 (66%)[16%]


My FTP site

[Dann Corbit]

Geschrieben von:/Posted by: Dann Corbit at 19 February 2000 03:35:46:
Als Antwort auf:/As an answer to: ColChess 5.5 steadily advances. geschrieben von:/posted by: Dann Corbit at 18 February 2000 21:36:41:
On a 500 MHz PIII machine at 5 seconds per position, ColChess 5.5 gets 214/300.
Test #300 : White to play
Current Position
----------------
b . . b . r . k
. . . R . p p p pP=Pawn
. . . . q P . . rR=Rook
. . . . p . P Q nN=Knight lower=Black
. . . . P . . . bB=Bishop CAPS=White
. . . . . B . . qQ=Queen
. . . . N . K . kK=King
. . . . . . . .
Thinking 2 355 2 1761 g5g6 h7h6
3 28 8 5114 g5g6 h7h6 d7f7
4 180 37 41407 g5g6 h7h6 f6g7 h8g7
5 88 210 171985 d7a7 d8f6 g5f6 e6f6 a7d7
6 192 500 522498 g5g6 h7h6 f6g7 h8g7 d7a7 d8g5
---Answer: g6 **-Correct-**
Correct So Far: 214[52]/300
Result: 214[52]/300 (71%)[17%]

My FTP site

[Pete Galati]

Geschrieben von:/Posted by: Pete Galati at 19 February 2000 05:52:19:
Als Antwort auf:/As an answer to: Re: ColChess 5.5 steadily advances. geschrieben von:/posted by: Dann Corbit at 19 February 2000 03:35:46:

On a 500 MHz PIII machine at 5 seconds per position, ColChess 5.5 gets 214/300.
Test #300 : White to play
Current Position
Thinking 2 355 2 1761 g5g6 h7h6
3 28 8 5114 g5g6 h7h6 d7f7
4 180 37 41407 g5g6 h7h6 f6g7 h8g7
5 88 210 171985 d7a7 d8f6 g5f6 e6f6 a7d7
6 192 500 522498 g5g6 h7h6 f6g7 h8g7 d7a7 d8g5
---Answer: g6 **-Correct-**
Correct So Far: 214[52]/300
Result: 214[52]/300 (71%)[17%]

----------------
b . . b . r . k
. . . R . p p p pP=Pawn
. . . . q P . . rR=Rook
. . . . p . P Q nN=Knight lower=Black
. . . . P . . . bB=Bishop CAPS=White
. . . . . B . . qQ=Queen
. . . . N . K . kK=King
. . . . . . . .

I don't understand what this is, sorry, I hope this straigtens the formatting so I can see it better.
Pete

[Dann Corbit]

Geschrieben von:/Posted by: Dann Corbit at 19 February 2000 06:07:15:
Als Antwort auf:/As an answer to: Re: ColChess 5.5 steadily advances. geschrieben von:/posted by: Pete Galati at 19 February 2000 05:52:19:

On a 500 MHz PIII machine at 5 seconds per position, ColChess 5.5 gets 214/300.
Test #300 : White to play
Current Position
Thinking 2 355 2 1761 g5g6 h7h6
3 28 8 5114 g5g6 h7h6 d7f7
4 180 37 41407 g5g6 h7h6 f6g7 h8g7
5 88 210 171985 d7a7 d8f6 g5f6 e6f6 a7d7
6 192 500 522498 g5g6 h7h6 f6g7 h8g7 d7a7 d8g5
---Answer: g6 **-Correct-**
Correct So Far: 214[52]/300
Result: 214[52]/300 (71%)[17%]

----------------
b . . b . r . k
. . . R . p p p pP=Pawn
. . . . q P . . rR=Rook
. . . . p . P Q nN=Knight lower=Black
. . . . P . . . bB=Bishop CAPS=White
. . . . . B . . qQ=Queen
. . . . N . K . kK=King
. . . . . . . .

I don't understand what this is, sorry, I hope this straigtens the formatting so I can see it better.

Here is the same position in EPD:
b2b1r1k/3R1ppp/4qP2/4p1PQ/4P3/5B2/4N1K1/8 w - - bm g6; id "WAC.300";
The picture is an ASCII representation of a chess board, which I simply copied from the screen when the test finished. The position is #300 in WAC (Win At Chess). The suggested move {Answer:) is g6. The program suggested g5g6, which is correct. Altogether, the program got 214 problems out of 300 correct in a 5 second per move challenge. Of those, 52 were checkmates. That is 71% correct and 17% incorrect (roughly).
HTH.

My FTP site

[Dann Corbit's Chess FTP s]

Geschrieben von:/Posted by: Dann Corbit's Chess FTP site at 19 February 2000 06:51:24:
Als Antwort auf:/As an answer to: Re: ColChess 5.5 steadily advances. geschrieben von:/posted by: Dann Corbit at 19 February 2000 06:07:15:
[snip]

Correct So Far: 214[52]/300
Result: 214[52]/300 (71%)[17%]

Altogether, the program got 214 problems out of 300 correct in a 5 second per
move challenge. Of those, 52 were checkmates. That is 71% correct and 17%
incorrect (roughly).

[snip]
Err..
Make that 17% checkmates [cough]


My FTP site

[Andrew Tanner]

Geschrieben von:/Posted by: Andrew Tanner at 19 February 2000 10:11:59:
Als Antwort auf:/As an answer to: Re: ColChess 5.5 steadily advances. geschrieben von:/posted by: Dann Corbit's Chess FTP site at 19 February 2000 06:51:24:
Hello Dan,
I am having trouble getting your colchess.exe (152k) to work under winboard. I have it running in the same folder as my own compile chess.exe (413k compiled using Cygnus) which works fine. I have the env variable set which points to the proper directory with the executables. When I run both engines from dos I get "using path c:\programme\winboard\col\ colchess xboard mode". Before I compiled my own version I had to uncomment this line in chess.h as directed. "#define xboard // Uncomment the beginning of this line to compile in xboard mode"
Did you uncomment this line in chess.h also in making your binary?
-Andrew

[Colin Frayn]

Geschrieben von:/Posted by: Colin Frayn at 19 February 2000 12:20:06:
Als Antwort auf:/As an answer to: Re: ColChess 5.5 steadily advances. geschrieben von:/posted by: Dann Corbit at 19 February 2000 03:35:46:

On a 500 MHz PIII machine at 5 seconds per position,
Result: 214[52]/300 (71%)[17%]

Cheers Dann, that's an improvement of +6 on V5.4, isn't it? Hopefully I'll be able to improve it some more, though I'm running out of ideas at the moment.
I'd be interested to see if anyone runs a tournament with the new version. I'm not sure how much it has actually improved in real game play.
Cheers,
Colin

ColChess Homepage