Author: Graham Banks
Date: 02:37:53 08/15/04
Go up one level in this thread
On August 15, 2004 at 03:56:35, Kurt Utzinger wrote: >On August 15, 2004 at 01:13:20, Graham Banks wrote: > >>Athlon XP2000 >>256mb hash each (or closest allowed by each program) >>3,4,5 piece tablebases >>All using Fritz Powerbook tournament book (maximum variety, minimum 5 games for >>a move to be played, no learning) >>40 moves in 40 minutes repeating. >>Ponder off >>Tournament run under Deep Fritz 8 GUI >> >> >> >>STANDINGS (after 42 of 76 rounds) >> >>32.0 - CM9000 Judge II >>26.0 - List 512 >>25.5 - CM 10th Edition (default) >>25.5 - SOS 4 >>25.0 - Gandalf 5.1 >>24.5 - Pharaon 2.62 >>24.5 - Thinker 4.6c >>23.5 - Delfi 4.5 >>23.5 - Pro Deo 1.0 >>21.0 - Anaconda 1.6.2 >>20.5 - Tao 5.6 >>20.0 - Deep Sjeng 1.6 >>20.0 - Gothmog 1.0b10 >>18.5 - Yace 0.99.87 >>16.5 - Fruit 1.5 >>16.0 - El Chinito 3.25 >>15.0 - Little Goliath Revival >>15.0 - Crafty 19.15 >>14.0 - Comet B68 >>13.5 - Slow Chess 2.93a >> >> >> >>The games and crosstable should be up on Kurt’s website sometime soon. >>http://www.utzingerk.com/banks/st_03_qual.htm >> >>Comments welcome. >>Graham. > > Hi Graham > Games and crosstable are updated on my website. > Some remarks to your fine qualifying tournament. > CM9_Judge_II: to test this setting is a must as > soon as I find time to do this. List512 second > is no surprise for this very strong engine. However, > the third place for SOS 4 for Arena could not expected > in advance in my opinion. And 4th rank for CM10k shows > that the default settings of the new engine could indeed > be better than those of CM9k. Gandalf51 a surprise to > me but I have to admit that I have no experience with > the Danish engine. Pharao 2.62 still too good in your > tourney and I think ranking will drop later. Very well > doing is Thinker 4.6c but the same thing can be said > in our AT 2004 [http://www.utzingerk.com/at_2004.htm] > Delfi 4.5 and Pro Deo 1.0 have a score of more than > 50 % and it's hard to say if these engines will climb > or fall in the coming rounds. A great difference between > our tournaments is the behaviour of Crafty 19.15: 56 % > to 35 %. The remaining engines will surely try to improve > their situation, in particular Deep Sjeng 1.6 can do > better. > Kind regards > Kurt Thanks for your thoughts Kurt and also thanks for hosting my tournaments on your website. I wonder how much influence the 256mb hash is having on some of these programs and whether or not this could explain the unexpectedly strong performances of some. SOS 4 and Pharaon 2.62 come to mind here. I agree with you that the default settings of CM 10th Edition are more balanced and stronger than the CM9000 ones. I'll be upping the selectivity setting for CM 10th Edition in the final stage of Super Tournament III. I guess 12 would be best? Graham. Graham.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.