Author: Frank Quisinsky
Date: 13:24:58 10/22/00
Go up one level in this thread
On October 22, 2000 at 15:59:43, Uri Blass wrote: >On October 22, 2000 at 15:01:28, Jorge Pichard wrote: > >>On October 22, 2000 at 14:41:57, Ulrich Tuerke wrote: >> >>>Congrats to Roland, too. Great performance shown by Patzer. >>> >>>Uli >> >>Patzer result would have been better if it was capable of using the Dual >>processors from the beginning. I understand that patzer had problem in the >>settings for which it was unable to recognize both processors. >> >>Pichard. > >We do not know that the result could be better. > >I understood that the problem was that it could not use the 2 proccesors in the >permanent brain and later the programmer fixed this problem but created another >problem when patzer could not use the 2 proccesors when it played and could use >it only in the permanent brain. > >I guess that the difference from these problem is only about 30 elo and it means >that the expected difference in the result is something like 0.4125 points out >of 11. > >0.4125 is an average number and the number can be also 0 points. > >Uri Hello, I come for minutes at home :-) Yes this is right Uri ! Roland played round 8-11 with the version from last weekend. Here 2 processor in normal mode and 1 processor in permanent brain. I think also that this is not very important. Only in round 7 against Chess Tiger this big problem from Patzer. Furthermore I forget to added the 5-piece directory in the configuration set variable from Patzer. Patzer played against Tiger with 4-piece nalimov TBs. I think very big luck for Chess Tiger. I think 1.3 - 1.4 for this dual Patzer version is very good. This is the power from a Pentium III ~ 1 GHz ! The latest game against Nimzo 8.0 is fantastic. Patzer see not a little combination after move 23 or 24 (Qg6+) ! Nimzo 8.0 see the combination and the evalutation from Nimzo was -8.0 ! Also not important because Patzer played an other move and the evalutation is +5.0 ! After move 20 or 21 is the game for Nimzo over. Chess Tiger played an very good tournament and I think Christophe make a very good work (he have a good team and a very good opening book). In all games stand Tiger with the opening book from Christophe better, OK not in the game against Patzer. Also good Tao and XiniX. Interesting is the game from XiniX against Ant. Xinix with ... KQQQ and Ant with KP ! Xinix will not made the mate move in 20-30 moves and losed later on time. Unfortunately, a little bug in XiniX ! Also interesting the game Diep - Goldbar. Goldbar with: KQPPP and Diep with KQ ! Goldbar make very bad moves and Diep won 2 pawns. The game is later remis ! Diep with luck ... For Roland Pfister is the result very very good. Against Quest not luck and against Kallisto big luck ! In the game against McTobber found Patzer in the last second with 5-piece TBs the right move ! But the Patzer games against Morphy, Nimzo and Diep are fantastic. I think 3 of the 5 best games in the complete tournament. A nice action from Gambit-Soft. All amateurs (10 programmers) got from Gambit-Soft a price (Gandalf 4.32f CD). Roland Troeger from Gambit-Soft visit the tournament and we (Roland Pfister, Roland Troeger, Guenter Rehburg and I) have a lot of fun in Leiden. The organistion is perfect, in all years in the organisation from the Dutch-ch perfect. Theo van der Storm made here a very good work. The last ... Congratulations to Christophe Theron, he created a monster ... But Christophe, I make later a test ... Gambit-Tiger 1.0 against Gandalf 4.32f ... (I will see) ! Regards Frank
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.