For anybody who wants to fool around with ColChess

[Dann Corbit]

Geschrieben von:/Posted by: Dann Corbit am 20 Januar 2000 02:04:15:
This is an xboard program I just ported to Win32:
ftp://38.168.214.175/pub/COLCHESS.ZIP


My FTP site

[Pete Galati]

Geschrieben von:/Posted by: Pete Galati am 20 Januar 2000 04:24:17:
Als Antwort auf:/As an answer to: For anybody who wants to fool around with ColChess geschrieben von:/posted by: Dann Corbit am 20 Januar 2000 02:04:15:

This is an xboard program I just ported to Win32:
ftp://38.168.214.175/pub/COLCHESS.ZIP

Thanks Dann,
I can't seem to make it work, it doesn't find it's book or calculate a move, I've tried it with no command line parameter, I've also tried with xboard and -xboard, but for me, it won't go.
Also, it doesn't want to shut down when you shut off Winboard, I have to hit Ctrl-Alt-Delete and then end it's task from there.
Pete

[Dann Corbit]

Geschrieben von:/Posted by: Dann Corbit am 20 Januar 2000 09:37:59:
Als Antwort auf:/As an answer to: Re: For anybody who wants to fool around with ColChess geschrieben von:/posted by: Pete Galati am 20 Januar 2000 04:24:17:

This is an xboard program I just ported to Win32:
ftp://38.168.214.175/pub/COLCHESS.ZIP

Thanks Dann,
I can't seem to make it work, it doesn't find it's book or calculate a move, I've tried it with no command line parameter, I've also tried with xboard and -xboard, but for me, it won't go.
Also, it doesn't want to shut down when you shut off Winboard, I have to hit Ctrl-Alt-Delete and then end it's task from there.

Darn. I had to modify it and forgot to describe it. There is an environmental variable (which I forget the name of but if you grep the code for getenv() you will find it. That variable describes the home directory of the program. From NT you can set it from control-panel/system/environment or you could set it in a batch file that launches the program or something.
It is actually pretty snappy for some things. I let it run at one second per position, and it got 45% of wac correct even at that very fast time control.
The author says it is weak, so it might be just the (beatable) program that everyone is begging for!

[Dann Corbit]

Geschrieben von:/Posted by: Dann Corbit am 20 Januar 2000 10:17:00:
Als Antwort auf:/As an answer to: Re: For anybody who wants to fool around with ColChess geschrieben von:/posted by: Pete Galati am 20 Januar 2000 04:24:17:
SET COLCHESS_PATH=
Also, make subdirectories there called .\games and .\test to store games and EPD test files.
It gets just over 50% on WAC at two seconds per move. Here is the tail end of the analysis:
Current Position
----------------
. . . Q . . . .
p . . . b . k . pP=Pawn
. . p . . r P p rR=Rook
. . q . . . . . nN=Knight lower=Black
. . . . B . . . bB=Bishop CAPS=White
P . . P . . . . qQ=Queen
. . . . . . . P kK=King
. . . . . . R K
Thinking 2 -29 3 1591 d8c7 c5a3 2.
3 45 28 14496 d8e8 f6f8 2. d3d4
4 -35 82 56439 d8e8 c5a3 2. e4c6 a7a5 3.
5 -16 200 129807 d8e8 c5a3 2. e4c6 a7a5 3. h2h4
---Answer: Qh8+ ~Incorrect-??
Correct So Far: 149[2]/298
Test #299 : White to play
Current Position
----------------
. n . . r r . .
. p k . . . p p pP=Pawn
p N n . . p . . rR=Rook
. . N . p . . . nN=Knight lower=Black
. . . . . . . . bB=Bishop CAPS=White
. . . . . . P . qQ=Queen
P P . . P P K P kK=King
. . R R . . . .
Thinking 2 124 1 1195 b6d5 c7c8 2.
3 -25 12 4321 b6d5 c7c8 2. e2e4
4 310 67 48615 c5a4 b8d7 2.
5 361 200 91305 c5a4 b8d7 2. d1d7 c7b8 3. d7g7
---Answer: Nca4 **-Correct-**
Correct So Far: 150[2]/299
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 21 2 1764 g5g6 f7g6 2.
3 7 10 4805 g5g6 f7g6 2. f6g7
4 60 43 38076 g5g6 h7h6 2. f6g7 h8g7 3.
5 60 139 82632 g5g6 h7h6 2. f6g7 h8g7 3. d7f7
---Answer: g6 **-Correct-**
Correct So Far: 151[2]/300
Result: 151[2]/300 (50%)[0%]
Some high NPS problems:
Test #255 : White to play
Current Position
----------------
. . . r . . . r
p . . . . p k . pP=Pawn
. . . . . R p . rR=Rook
. . . q . . . . nN=Knight lower=Black
. p . P . . R Q bB=Bishop CAPS=White
. . . . . N . . qQ=Queen
P . P . . . . P kK=King
. . b . . . . K
Thinking 2 -424 2 1751 g4g6 f7g6 2.
3 -389 12 5729 f6g6 f7g6 2. h4g3
4 -360 38 41448 f6g6 f7g6 2. h4e7 g7h6 3.
5 -384 119 89383 f6g6 f7g6 2. h4e7 g7h6 3. g4h4
6 -372 200 216585 f6g6 f7g6 2. h4e7 d5f7 3. e7e5 f7f6 4.
---Answer: Rfxg6+ **-Correct-**
Correct So Far: 134[12]/255
Test #281 : White to play
Current Position
----------------
. . R . . . . .
. . R . . . . p pP=Pawn
. . . . . p . k rR=Rook
. . . . . . n . nN=Knight lower=Black
. . . . . . . . bB=Bishop CAPS=White
. P . . Q P P q qQ=Queen
r . . . . . . . kK=King
. . . . . . K .
Thinking 2 -181 1 597 c7c2 h3g3 2.
3 -306 3 2216 c7c2 h3g3 2. g1h1
4 -297 8 5487 c7c2 h3g3 2. g1f1 a2c2 3.
5 -305 26 28859 c7c2 a2a1 2. c2c1 3.
6 -305 58 58626 c7c2 a2a1 2. c2c1 h3g3 3. g1f1 a1c1 4.
7 -305 206 271089
---Answer: Rxh7+ ~Incorrect-??
Correct So Far: 143[0]/281
Both of these achieved well over 100K NPS.

[Colin Frayn]

Geschrieben von:/Posted by: Colin Frayn am 20 Januar 2000 16:35:55:
Als Antwort auf:/As an answer to: Re: For anybody who wants to fool around with ColChess geschrieben von:/posted by: Dann Corbit am 20 Januar 2000 10:17:00:

It gets just over 50% on WAC at two seconds per move. Here is the tail end of the analysis:
Current Position
----------------
. . . Q . . . .
p . . . b . k . pP=Pawn
. . p . . r P p rR=Rook
. . q . . . . . nN=Knight lower=Black
. . . . B . . . bB=Bishop CAPS=White
P . . P . . . . qQ=Queen
. . . . . . . P kK=King
. . . . . . R K
Thinking 2 -29 3 1591 d8c7 c5a3 2.
3 45 28 14496 d8e8 f6f8 2. d3d4
4 -35 82 56439 d8e8 c5a3 2. e4c6 a7a5 3.
5 -16 200 129807 d8e8 c5a3 2. e4c6 a7a5 3. h2h4
---Answer: Qh8+ ~Incorrect-??
Correct So Far: 149[2]/298

What system is that running on?
Try recompiling with xboard undefined (in xboard.h) and you'll find it gives slightly more readable output! The test function wasn't designed to be used in xboard mode. In fact I might even turn it off when in xboard mode, along with one or two other functions that aren't of any use then.
Cheers,
Colin

ColChess Homepage

[Colin Frayn]

Geschrieben von:/Posted by: Colin Frayn am 20 Januar 2000 16:40:48:
Als Antwort auf:/As an answer to: Re: For anybody who wants to fool around with ColChess geschrieben von:/posted by: Dann Corbit am 20 Januar 2000 09:37:59:

It is actually pretty snappy for some things. I let it run at one second per position, and it got 45% of wac correct even at that very fast time control.
The author says it is weak, so it might be just the (beatable) program that >everyone is begging for!

That's interesting. The annoying thing about computer programs is that they can spot almost instantaneously mates in 3 or less without even a slight effort!
Email me the results and I'll put them up on my web page.
Hehe glad to see I've been of assistance I can see this becoming a cult program *evil grin*
I'll just remove some of those options that are incompatible with xboard now, and post the new vesion on my web page.
Cheers,
Colin

ColChess Homepage

[Volker Pittlik]

Geschrieben von:/Posted by: Volker Pittlik am 20 Januar 2000 18:55:48:
Als Antwort auf:/As an answer to: Re: For anybody who wants to fool around with ColChess geschrieben von:/posted by: Colin Frayn am 20 Januar 2000 16:40:48:
What's my mistake?
set:
C:\>set colchess_path
COLCHESS_PATH=d:\winboard\colchess\
WB-ini:
/firstChessProgramNames="GNUChess
...
\"Colchess\" /fd=d:\\Winboard\\colchess
...
/secondChessProgramNames="GNUChess
...
\"Colchess\" /sd=d:\\Winboard\\colchess
...
wb-debug:
WinBoard 4.0.5 + Colchess
StartChildProcess (dir="d:\Winboard\colchess") Colchess
141 >first : xboard
141 >first : new
random
141 >first : level 0 2 0
141 >first : post
141 >first : hard
141 >first : easy
156

[Dann Corbit]

Geschrieben von:/Posted by: Dann Corbit am 20 Januar 2000 20:02:41:
Als Antwort auf:/As an answer to: Re: For anybody who wants to fool around with ColChess geschrieben von:/posted by: Colin Frayn am 20 Januar 2000 16:35:55:

It gets just over 50% on WAC at two seconds per move. Here is the tail end of the analysis:
Current Position
----------------
. . . Q . . . .
p . . . b . k . pP=Pawn
. . p . . r P p rR=Rook
. . q . . . . . nN=Knight lower=Black
. . . . B . . . bB=Bishop CAPS=White
P . . P . . . . qQ=Queen
. . . . . . . P kK=King
. . . . . . R K
Thinking 2 -29 3 1591 d8c7 c5a3 2.
3 45 28 14496 d8e8 f6f8 2. d3d4
4 -35 82 56439 d8e8 c5a3 2. e4c6 a7a5 3.
5 -16 200 129807 d8e8 c5a3 2. e4c6 a7a5 3. h2h4
---Answer: Qh8+ ~Incorrect-??
Correct So Far: 149[2]/298

What system is that running on?
Try recompiling with xboard undefined (in xboard.h) and you'll find it gives slightly more readable output! The test function wasn't designed to be used in xboard mode. In fact I might even turn it off when in xboard mode, along with one or two other functions that aren't of any use then.

The run was on a 350 MHz PIII running NT 4.0. The program was compiled with MS VC++ 6.0

I only use xboard programs. I don't have any interest at all in other formats.
I can read the output just fine. It seems pretty obvious to me what is going on.