Crafty 23.2 released.

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

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
User avatar
Aser Huerga
Posts: 812
Joined: Tue Jun 16, 2009 8:09 am
Location: Spain

Re: the crafty-crafty games

Post by Aser Huerga » Fri Mar 05, 2010 12:25 pm

After setting egtb on and indicate tb path on the command line parameters, the debug window shows:

9.703<--1:EGTB access enabled
389.735<--1:using tbpath=F:\Ajedrez\TB
389.797<--1:5 piece tablebase files found
389.828<--1:22274kb of RAM used for TB indices and decompression tables
389.860<--1:EGTB cache memory = 32 bytes.


but no ponder hits anyway.

User avatar
Aser Huerga
Posts: 812
Joined: Tue Jun 16, 2009 8:09 am
Location: Spain

Re: the crafty-crafty games

Post by Aser Huerga » Fri Mar 05, 2010 1:30 pm

I note that Crafty shows <TB> and <EGTB> in the evaluation output, so I asumme it's using EGTBs.

Robert Flesher
Posts: 1206
Joined: Tue Aug 18, 2009 1:06 am

Re: Crafty 23.2 released.

Post by Robert Flesher » Fri Mar 05, 2010 2:01 pm

bob wrote:
yanquis1972 wrote:20 elo is actually pretty impressive, imo. bob, do you have a guesstimate as to what crafty's human elo would be? i know come CC fans don't even think in such terms but i'd think as an author it would be one of the most interesting aspects to me. i guess computer chess has in some ways (certainly in strength) surpassed human chess though.
Too many different rating pools. USCF? CCF? FIDE? Each would be different. And since there are not many games between Crafty and FIDE players (except for blitz where humans have real problems anyway) it is difficult to produce a decent guess...

Some of us remember when you believed, and quite openly stated engines were not GM strength (not even close). :shock: However, that was then and this is now. Clearly Crafty is of strong GM strength, and seems alot harder to beat than many of the ELO giants. Personally, I like the direction you have taken your engine. True to its roots, and a S.O.B to beat in blitz.

Jouni
Posts: 1802
Joined: Wed Mar 08, 2006 7:15 pm

Re: Crafty 23.2 released.

Post by Jouni » Fri Mar 05, 2010 4:52 pm

With quite similar hardware (raw nodes per second: 6 846 344) I got only 67/100 solved (time 48:00).

Jouni

BBauer
Posts: 657
Joined: Wed Mar 08, 2006 7:58 pm

Re: Crafty 23.2 released.

Post by BBauer » Fri Mar 05, 2010 9:12 pm

I used an AMD X2 5200 computer, 2.66GHz, 2 procs.
Crafty is compiled with gcc.
Did you use tablebases? That may hurt.
Kind regards
Bernhard

bob
Posts: 20354
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Crafty 23.2 released.

Post by bob » Fri Mar 05, 2010 11:07 pm

Robert Flesher wrote:
bob wrote:
yanquis1972 wrote:20 elo is actually pretty impressive, imo. bob, do you have a guesstimate as to what crafty's human elo would be? i know come CC fans don't even think in such terms but i'd think as an author it would be one of the most interesting aspects to me. i guess computer chess has in some ways (certainly in strength) surpassed human chess though.
Too many different rating pools. USCF? CCF? FIDE? Each would be different. And since there are not many games between Crafty and FIDE players (except for blitz where humans have real problems anyway) it is difficult to produce a decent guess...

Some of us remember when you believed, and quite openly stated engines were not GM strength (not even close). :shock: However, that was then and this is now. Clearly Crafty is of strong GM strength, and seems alot harder to beat than many of the ELO giants. Personally, I like the direction you have taken your engine. True to its roots, and a S.O.B to beat in blitz.
That was 10-15 years ago, when it was correct. Hardware has changed a bunch since then.

bob
Posts: 20354
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: the crafty-crafty games

Post by bob » Fri Mar 05, 2010 11:08 pm

Aser Huerga wrote:Is my first time I try to test Crafty, and I'm finding problems with endgame tablebases, I can't manage Crafty to use it ... any help?

My rc file include:

egtb
#adaptive 750K 24M 192M 24M 48M
tbpath=c:\ajedrez\tb
Do you have a version with EGTB support compiled in? If so, it should work just fine, it does here...

bob
Posts: 20354
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: the crafty-crafty games

Post by bob » Fri Mar 05, 2010 11:09 pm

Aser Huerga wrote:No subfolder after TB folder... Testing a typical TB position, Crafty shows 0 TB Hits and a conventional progressive evaluation.
Not very convincing. If you pick a drawn EGTB position, swindle-mode can cause this to happen.

User avatar
Aser Huerga
Posts: 812
Joined: Tue Jun 16, 2009 8:09 am
Location: Spain

Re: the crafty-crafty games

Post by Aser Huerga » Fri Mar 05, 2010 11:52 pm

bob wrote:
Do you have a version with EGTB support compiled in? If so, it should work just fine, it does here...
Hello Bob,

I'm using the JA compile and Arena 2.0.1
Aser Huerga wrote:After setting egtb on and indicate tb path on the command line parameters, the debug window shows:

9.703<--1:EGTB access enabled
389.735<--1:using tbpath=F:\Ajedrez\TB
389.797<--1:5 piece tablebase files found
389.828<--1:22274kb of RAM used for TB indices and decompression tables
389.860<--1:EGTB cache memory = 32 bytes.


but no ponder hits anyway.

... take a look:

Image

bob
Posts: 20354
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: the crafty-crafty games

Post by bob » Sat Mar 06, 2010 2:44 am

Aser Huerga wrote:
bob wrote:
Do you have a version with EGTB support compiled in? If so, it should work just fine, it does here...
Hello Bob,

I'm using the JA compile and Arena 2.0.1
Aser Huerga wrote:After setting egtb on and indicate tb path on the command line parameters, the debug window shows:

9.703<--1:EGTB access enabled
389.735<--1:using tbpath=F:\Ajedrez\TB
389.797<--1:5 piece tablebase files found
389.828<--1:22274kb of RAM used for TB indices and decompression tables
389.860<--1:EGTB cache memory = 32 bytes.


but no ponder hits anyway.

... take a look:

Image
Something's broken. The program is definitely hitting the egtb's because the PV's end in <EGTB>. I am not sure how it determines the TB hits and it might be that Crafty doesn't send the necessary data for that to show.

Crafty's logfile would show egtb hits however, so you could look there to verify things are working.

Post Reply