Author: Vincent Diepeveen
Date: 05:14:16 02/25/99
Go up one level in this thread
On February 24, 1999 at 13:49:03, blass uri wrote: > >On February 24, 1999 at 12:35:27, Vincent Diepeveen wrote: > >>On February 24, 1999 at 09:05:12, blass uri wrote: >> >>> >>>On February 23, 1999 at 11:55:42, blass uri wrote: >>> >>>> >>>>On February 23, 1999 at 10:21:53, Dan Kiski wrote: >>>> >>>>>I read a number of posts on the subject of Rebel allegedly cooking some set >>>>>tests a while back and wondered if Fritz had done something similar. >>>>> >>>>>When I play engine 5.15 v 5.32 tests using the Nunn openings I find 5.32 will >>>>>win by Impressive margins 15 - 5, 15.5 - 4.5, 14.5 - 5.5. But when I play engine >>>>>v engine normal matches they seem relatively even. >>>> >>>>It seems that they cooked Fritz5.32 to win the nunn test games. >>>>This proves that the nunn test does not prove nothing about the strength of >>>>chess engines because it is a known test. >>>> >>>>I suggest another test based on the positions after the 10 move of black in the >>>>last category 18+ tournament. >>>> >>>>Uri >>> >>>There is a different explanation. >>> >>>Maybe Fritz5.32 is better at the opening stage of the game and it is the reason >>>that Fritz5.32 is getting better results when both sides do not use the opening >>>book in the nunn positions. >>> >>>If this is the case then I can understand the fact that Fritz5.32 is not better >>>in practical games because usually both sides are out of book only near move 20 >>>and Fritz5.32 is not better after move 20 >>> >>>I suggest to do a match of 40 games based on all the 20 positions after 1 ply: >>>after 1.a3,1.a4,1.b3,1.b4,..... >>> >>>I am sure that Chessbase did not tune Fritz for this match so if Fritz5.32 wins >>>convincingly then it will be a proof that Fritz5.32 is clearly better in the >>>opening stage then Fritz5.16 >>> >>>Uri >> >>this doesn't explain the difference between 5.32 and 5.16 > >If Fritz5.32 is better than Fritz5.16 in finding good moves in the opening stage >then it can explain the difference between Fritz5.32 and Fritz5.16 in the nunn >match. > >I do not know if this is the case. >> >>Further this is not true. Of course fritz is no good in opening, >>it is not good at all, > >How do you know? > >Did you test Fritz5.32? >> >>but if both programs don't know what to think from the position, then >>the deeper searching program wins, and at equal hardware, this >>happens to be fritz or ferret, and as fritz is the only commercial one >>of the 2, it is fritz in this case. > >I do not think that Fritz is the deeper searcher at equal hardware relative to >other programs. >I know that Chessmaster6000 can sometimes solve positions that other programs >(including Fritz) cannot solve. > >I read here that Chessmaster6000 won Fritz5.16 in the nunn match. Please don't confuse tactical search depth with positional search depth. One can see tactical everything, but still miss positional everything. See the game Diep-The King in dutch open. Every move i was outsearched tactical, yet positional i was just as deeply searching and see what happened. cm6000 was released after nunn test set, so no big deal that it plays it better. >Uri
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.