Author: blass uri
Date: 01:32:11 10/11/98
Go up one level in this thread
On October 11, 1998 at 04:01:51, Keith Ian Price wrote: >On October 09, 1998 at 03:17:11, Harald Faber wrote: > >>On October 07, 1998 at 19:01:38, Keith Ian Price wrote: >> >>>>>>>Nunn-Match 40/120, 20/60 >>>>>>> 1 2 3 4 5 6 7 8 9 10 >>>>>>>1 HIARCS 6,P233MMX 2540 1½1111½1111½0½½11½11 16.0/20 >>>>>>>2 CST Black,PII-400 2450 0½0000½0000½1½½00½00 4.0/20 >>>>>>> >>>>>>>ELO performance difference Hiarcs P233MMX - CST PII400: 240 >>>> >>>>>>Performance on ICC was: >>>>>> >>>>>>Opponents avg. rating: 2393 >>>>>>CSTWin/BETA +19 =9 -38 total: 36% >>>>>>ELO-Performance 2284 >>>> >>>>>I cannot believe that Cstalwin95 has only 2300 in serious games after seeing the >>>>>results of Thorsten's tournament(Cstal has 6.5 out of 9) >>>>>Uri >>>> >>>>Thorstens tourney results is like flipping a coin. EVERYTHING is possible. >>> >>>I think that the ICC games were blitz or rapid chess games, and CSTal does much >>>better in games at tournament time controls than at blitz. Thorsten's tournament >>>is at 40/120, and therefore his results are quite possible, although any single >>>tournament should not be used as a measure of the strength of a program. >>>kp >> >>Start reading above. The Nunn-Test, onem ay call it none-test, was played in >>tournament mode. SO no excuses. > >Start reading above. "although any single tournament should not be used as a >measure of the strength of a program": SO no unsubstantiated conclusions. The >Nunn-Test you refer to was conducted between CSTBlack and Hiarcs. Hiarcs was not >even in mclane's tournament Hiarcs is in mclane's tournament It is the only program that did not lose a game(4 wins and 6 draws) , and mclane is using the Win95 version, so my >statement that his results are quite possible still stands. I agree that his results are quite possible because Win95 version is not Cstalblack and the nunn match is about games without opening preperation and it is not the same as tournament games. The games In ICC were close to touenament time control 45 minutes per game+45 seconds per move or 30 minutes per game+30 seconds per move but: 1) I am not sure if Thorsten use the same win95 version that were used in ICC maybe they used a beta version that is worse than the version that thorsten uses. 2)Thorsten said that he used version n+x when chris in ICC used version n+x+y that is worse than n+x 3)Many games were against the same opponent crafty so maybe tal has problems against crafty. Uri Your conclusion that >my statement was a defense for CST was incorrect. It was more a defense of >Thorsten's reputation from an unwarranted attack in a forum in which he is no >longer allowed to defend himself. > >kp
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.