Page 1 of 1

New UCI engine - Critter Chess 0.34

PostPosted: 17 Apr 2009, 02:10
by Tony Mokonen
http://sourceforge.net/projects/critterchess/

By Richard Vida, Slovak Republic

Re: New UCI engine - Critter Chess 0.34

PostPosted: 17 Apr 2009, 02:31
by Graham Banks
Hi Tony,

any idea as to Critter's strength?

Cheers,
Graham.

Re: New UCI engine - Critter Chess 0.34

PostPosted: 17 Apr 2009, 03:17
by Tony Mokonen
Quick reply Graham, I barely had time to try it out :)

This critter seems pretty scrappy, hmm in the 2300's on the UEL scale, from the few quick 3/0 tests I tried:

Critter 0.34 - Glass 1.1 1-0
Critter 0.34 - Gibbon 2.52a 1-0
Philou 2.7.0 - Critter 0.34 0-1
GreKo 6.30 - Critter 0.34 0-1
Critter 0.34 - Arasan 11.3 1-0
Yace 0.99.57 DC - Critter 0.34 1-0

Re: New UCI engine - Critter Chess 0.34

PostPosted: 17 Apr 2009, 03:41
by Graham Banks
Tony Mokonen wrote:Quick reply Graham, I barely had time to try it out :)

This critter seems pretty scrappy, hmm in the 2300's on the UEL scale, from the few quick 3/0 tests I tried:

Critter 0.34 - Glass 1.1 1-0
Critter 0.34 - Gibbon 2.52a 1-0
Philou 2.7.0 - Critter 0.34 0-1
GreKo 6.30 - Critter 0.34 0-1
Critter 0.34 - Arasan 11.3 1-0
Yace 0.99.57 DC - Critter 0.34 1-0


Hmm - not too bad if those early results can be considered a good indication. Thanks Tony. :)

Re: New UCI engine - Critter Chess 0.34

PostPosted: 17 Apr 2009, 05:53
by Olivier Deville
Nice find Tony !

Olivier

Re: New UCI engine - Critter Chess 0.34

PostPosted: 19 Apr 2009, 11:34
by Graham Banks
Image

Re: New UCI engine - Critter Chess 0.34

PostPosted: 22 Apr 2009, 18:14
by Guenther Simon
Olivier Deville wrote:Nice find Tony !

Olivier


I have it losing on time a lot here in move 40 in x/y time control. It seems it often tries
to finish the last ply when failing high or low and ignores the time then?

Guenther

Re: New UCI engine - Critter Chess 0.34

PostPosted: 26 Apr 2009, 11:50
by rvida
Guenther Simon wrote:
I have it losing on time a lot here in move 40 in x/y time control. It seems it often tries
to finish the last ply when failing high or low and ignores the time then?

Guenther


It doesnt try to finish the last ply even when failing low. The problem is probably that search termination has a latency about 40-60 ms (at least on my machine). Arena seems to tolerate that. I'll try to fix it in the next release.

Re: New UCI engine - Critter Chess 0.34

PostPosted: 26 Apr 2009, 12:36
by Olivier Deville
Welcome to the Winboard Forum Richard, and congratulations for this very promising new engine :)

Critter got a wildcard for my E group that will start in a few days.

Olivier

Re: New UCI engine - Critter Chess 0.34

PostPosted: 26 Apr 2009, 20:10
by Guenther Simon
rvida wrote:
Guenther Simon wrote:
I have it losing on time a lot here in move 40 in x/y time control. It seems it often tries
to finish the last ply when failing high or low and ignores the time then?

Guenther


It doesnt try to finish the last ply even when failing low. The problem is probably that search termination has a latency about 40-60 ms (at least on my machine). Arena seems to tolerate that. I'll try to fix it in the next release.


Hi Richard,

That could be the wellknown problem in Arena, which even allows to reach the cut with negative time
on the clock(which is a bug)! WB doesn't forgive any 10th of milliseconds. I have a lot of logs here with Polyglot
timestamps which show that CritterChess doesn't move at the given time in move 40.
I will happily await your next version :)

Guenther

Re: New UCI engine - Critter Chess 0.34

PostPosted: 27 Apr 2009, 16:33
by rvida
Critter chess 0.36 released, timing issues hopefully fixed