Crafty 25.2

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

Tobber
Posts: 379
Joined: Fri Sep 28, 2012 5:53 pm
Location: Sweden

Crafty 25.2

Post by Tobber »

Hi

I have a problem running engine-engine matches with Crafty. Tried under Arena and Winboard but the behaviour is the same, when it comes out of the book it start to search but never moves, the time will run out. Are there some parameters needed in the setup?

Thanks
John
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Crafty 25.2

Post by MikeB »

Tobber wrote:Hi

I have a problem running engine-engine matches with Crafty. Tried under Arena and Winboard but the behaviour is the same, when it comes out of the book it start to search but never moves, the time will run out. Are there some parameters needed in the setup?

Thanks
John
If you are using crafty.rc file , please post what is in it. If you haven't already done so , type "log on" in the crafty.rc file and publish what it shows when you set up the engine match. Is it only engine vs engine or you play human vs engine does it play normal. Just to make sure the engine is compatible with your pc , open it in a dos window and type "bench" if that works , type "go". Any error messages ?
User avatar
Graham Banks
Posts: 41468
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Crafty 25.2

Post by Graham Banks »

Tobber wrote:Hi

I have a problem running engine-engine matches with Crafty. Tried under Arena and Winboard but the behaviour is the same, when it comes out of the book it start to search but never moves, the time will run out. Are there some parameters needed in the setup?

Thanks
John
Try this rc file.
No problems under ChessGUI.

mt 0
smpnice=1
egtb
tbpath=C:\SyzygyTablebases;
hash=128M
hashp=128M
ponder off
learn 0
book off
log=off
swindle off
exit
gbanksnz at gmail.com
adnoh
Posts: 72
Joined: Tue Jun 26, 2007 6:31 am
Full name: Charles Wong

Re: Crafty 25.2

Post by adnoh »

I experience almost the same thing for Crafty 25.2 in engine-engine matches except during the first game it does seem to search and play a few moves and then stop replying and loses. Then all subsequent games in the matche does not play coming out of the book. Crafty 25.1 works fine and I am using Peter Osterlund's builds for 25.1 and 25.2 and same crafty.rc file.

My crafty.rc is very minimal and here it is.

Code: Select all

egtb
learn 0
log on
ponder off
smpmt 0
smpnice 1
tbpath=c:\Users\Charles\Documents\Chess\syzygy\
exit
With log on it seems Crafty 25.2 rejects the "new" command.

The beginning of the log in a ga,e with 25.1

Code: Select all

pondering disabled.
parallel threads disabled.
SMP terminate extra threads when idle.


Crafty v25.1

machine has 4 processors

White(1): xboard

tellicsnoalias set 1 Crafty v25.1 (1 cpus)
tellicsnoalias kibitz Hello from Crafty v25.1! (1 cpus)
White(1): post
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): new
White(1): force
White(1): e2e4
feature done=1
Black(1): c7c5
White(2): g1f3
The beginning of a game with 25.2. Note it says

Code: Select all

pondering disabled.
parallel threads disabled.
SMP terminate extra threads when idle.


Crafty v25.2

machine has 4 processors

White(1): xboard

tellicsnoalias set 1 Crafty v25.2 (1 cpus)
tellicsnoalias kibitz Hello from Crafty v25.2! (1 cpus)
White(1): post
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): new
NOTICE:  new command not implemented, please exit and
restart crafty to re-initialize everything for a new game
White(1): force
White(1): d2d4
feature done=1
Black(1): g7g6
White(2): c2c4

During play with 25.2 I see the new command being rejected again and again like this.

Code: Select all

White(11): level 0 1:01 0
 game/61 seconds primary time control
White(11): time 6000
time remaining:   1:00 (Crafty).
White(11): otim 6000
time remaining:   1:00 (opponent).
White(11): white
White(11): go
        time limit 2.30 (11.50)
        depth     time       score   variation (1)
         17     1.44/1.65     0.66   11. Bf4 Bf8 12. b3 Nh5 13. Be3 Nd7 14. Qd2
                                     Ne5 15. Rad1 Be7 16. Na4 Bd7 17. Bb6 Qc8
                                     18. Nc3 g5 19. Nd4
         17     1.57/2.76       ++   11. Be3! (>+0.84)                   
         17     3.21/2.76     0.72   11. Be3 c5 12. Bf4 Re6 13. Qd2 Nc6
                                     14. Rad1 Nd7 15. a3 Nde5 16. Nd5 Bd7
                                     17. Bg5 f6 18. Bf4 Kh8 19. Nce3
         17->   3.30/2.76     0.72   11. Be3 c5 12. Bf4 Re6 13. Qd2 Nc6
                                     14. Rad1 Nd7 15. a3 Nde5 16. Nd5 Bd7
                                     17. Bg5 f6 18. Bf4 Kh8 19. Nce3
        time=3.30(100%)  nodes=20717104(20.7M)  fh1=91%  pred=0  nps=6.3M
        chk=94.5K  qchk=137.9K  fp=7.3M  mcp=4.2M  50move=1
        LMReductions:  1/484.9K  2/311.8K  3/230.3K  4/119.1K  5/10.7K  6/63
        null-move (R):  3/671.7K  4/37.2K  5/356
White(11): Be3
              time used:   3.30
Black(11): new
NOTICE:  new command not implemented, please exit and
restart crafty to re-initialize everything for a new game
Black(11): force
Black(11): d2d4
Illegal move: d2d4
Black(11): g7g6
Illegal move: g7g6
Does that help answer why 25.2 is behaving this way? Any further things I can try to help debug?
bob
Posts: 20943
Joined: Mon Feb 27, 2006 7:30 pm
Location: Birmingham, AL

Re: Crafty 25.2

Post by bob »

adnoh wrote:I experience almost the same thing for Crafty 25.2 in engine-engine matches except during the first game it does seem to search and play a few moves and then stop replying and loses. Then all subsequent games in the matche does not play coming out of the book. Crafty 25.1 works fine and I am using Peter Osterlund's builds for 25.1 and 25.2 and same crafty.rc file.

My crafty.rc is very minimal and here it is.

Code: Select all

egtb
learn 0
log on
ponder off
smpmt 0
smpnice 1
tbpath=c:\Users\Charles\Documents\Chess\syzygy\
exit
With log on it seems Crafty 25.2 rejects the "new" command.

The beginning of the log in a ga,e with 25.1

Code: Select all

pondering disabled.
parallel threads disabled.
SMP terminate extra threads when idle.


Crafty v25.1

machine has 4 processors

White(1): xboard

tellicsnoalias set 1 Crafty v25.1 (1 cpus)
tellicsnoalias kibitz Hello from Crafty v25.1! (1 cpus)
White(1): post
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): new
White(1): force
White(1): e2e4
feature done=1
Black(1): c7c5
White(2): g1f3
The beginning of a game with 25.2. Note it says

Code: Select all

pondering disabled.
parallel threads disabled.
SMP terminate extra threads when idle.


Crafty v25.2

machine has 4 processors

White(1): xboard

tellicsnoalias set 1 Crafty v25.2 (1 cpus)
tellicsnoalias kibitz Hello from Crafty v25.2! (1 cpus)
White(1): post
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): hard
pondering enabled.
White(1): easy
pondering disabled.
White(1): new
NOTICE:  new command not implemented, please exit and
restart crafty to re-initialize everything for a new game
White(1): force
White(1): d2d4
feature done=1
Black(1): g7g6
White(2): c2c4

During play with 25.2 I see the new command being rejected again and again like this.

Code: Select all

White(11): level 0 1:01 0
 game/61 seconds primary time control
White(11): time 6000
time remaining:   1:00 (Crafty).
White(11): otim 6000
time remaining:   1:00 (opponent).
White(11): white
White(11): go
        time limit 2.30 (11.50)
        depth     time       score   variation (1)
         17     1.44/1.65     0.66   11. Bf4 Bf8 12. b3 Nh5 13. Be3 Nd7 14. Qd2
                                     Ne5 15. Rad1 Be7 16. Na4 Bd7 17. Bb6 Qc8
                                     18. Nc3 g5 19. Nd4
         17     1.57/2.76       ++   11. Be3! (>+0.84)                   
         17     3.21/2.76     0.72   11. Be3 c5 12. Bf4 Re6 13. Qd2 Nc6
                                     14. Rad1 Nd7 15. a3 Nde5 16. Nd5 Bd7
                                     17. Bg5 f6 18. Bf4 Kh8 19. Nce3
         17->   3.30/2.76     0.72   11. Be3 c5 12. Bf4 Re6 13. Qd2 Nc6
                                     14. Rad1 Nd7 15. a3 Nde5 16. Nd5 Bd7
                                     17. Bg5 f6 18. Bf4 Kh8 19. Nce3
        time=3.30(100%)  nodes=20717104(20.7M)  fh1=91%  pred=0  nps=6.3M
        chk=94.5K  qchk=137.9K  fp=7.3M  mcp=4.2M  50move=1
        LMReductions:  1/484.9K  2/311.8K  3/230.3K  4/119.1K  5/10.7K  6/63
        null-move (R):  3/671.7K  4/37.2K  5/356
White(11): Be3
              time used:   3.30
Black(11): new
NOTICE:  new command not implemented, please exit and
restart crafty to re-initialize everything for a new game
Black(11): force
Black(11): d2d4
Illegal move: d2d4
Black(11): g7g6
Illegal move: g7g6
Does that help answer why 25.2 is behaving this way? Any further things I can try to help debug?
Yes it does reject the new command. It sends the string "reuse=0" which says it must be restarted for each new game. Ignoring the "new" doesn't hurt a thing in my testing. It does NOT send an error message to the GUI, just a message indicating that "new" is not implemented.

Based on the above log, the GUI you are using is not honoring the "feature reuse=0" option that is part of the winboard protocol. In the log a game was in progress and the GUI sent a "new" which Crafty ignored. What GUI is this? Perhaps one that has not implemented the full winboard version 2 protocol?
adnoh
Posts: 72
Joined: Tue Jun 26, 2007 6:31 am
Full name: Charles Wong

Re: Crafty 25.2

Post by adnoh »

I am using Shredder 13 as my GUI. Same thing happened with Shredder 12.
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Crafty 25.2

Post by MikeB »

adnoh wrote:I am using Shredder 13 as my GUI. Same thing happened with Shredder 12.
Crafty is an xBoard (Winboard) engine , Shredder a UCI interface. Try using Arena - it's free and handles both xboard and uci engines I believe.

If you're really adventurous, you can try to follow these instructions , but I would not be able to help you if you get stuck. I'm on macOS and xBoard handles both engines relatively easily - can even run Windows 32 bit engines on the macOS using vine.

https://sourceforge.net/p/craftychessin ... 20engines/
User avatar
Mike S.
Posts: 1480
Joined: Thu Mar 09, 2006 5:33 am

Re: Crafty 25.2

Post by Mike S. »

Graham Banks wrote:swindle off
As a sidenote, I don't see a reason to switch swindle off. If it can be maintained without risk, one should never reduce the opponent's chances to fail :mrgreen:

My current crafty.rc file is

Code: Select all

egtb
tbpath=N:\Nalimovs-345
cache=16M
ponder off
# mode=tournament  # (for use use with ponder on)
swindle on         # (Allows Crafty to try to win drawn games (according to Endgame Tables))
learn 2
book off            # (default = book on)
log=off
#mt=4              # (Increases Crafty's MaxThreads to 2 for a dual CPU computer )
#smpnice=1         # (Make Crafty not use cpu on opponents time) // not needed anymore
timebook 120 3     # (Makes crafty use a lot more time on the first 8 moves out of book)
hash=384M
hashp=64M
exit
...but I need to admit I havent run any matches or tournaments yet, with the new 25.2 version. If upcoming, it will be in Arena 3.5.
Regards, Mike
User avatar
Graham Banks
Posts: 41468
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Crafty 25.2

Post by Graham Banks »

Mike S. wrote:
Graham Banks wrote:swindle off
As a sidenote, I don't see a reason to switch swindle off. If it can be maintained without risk, one should never reduce the opponent's chances to fail :mrgreen:
Bob can correct me if I'm wrong, but it looked to be off by default.
gbanksnz at gmail.com
Dirt
Posts: 2851
Joined: Wed Mar 08, 2006 10:01 pm
Location: Irvine, CA, USA

Re: Crafty 25.2

Post by Dirt »

Mike S. wrote:
Graham Banks wrote:swindle off
As a sidenote, I don't see a reason to switch swindle off.
If all the engines support tablebases it won't help - and it can waste quite a bit of time.
Deasil is the right way to go.