Author: blass uri
Date: 23:59:05 05/07/00
Go up one level in this thread
On May 08, 2000 at 01:12:35, Harald Faber wrote: >On May 07, 2000 at 05:27:03, blass uri wrote: > >>On May 07, 2000 at 03:31:24, Harald Faber wrote: >> >>>On May 06, 2000 at 15:51:55, Marcus Kaestner wrote: >>> >>>>On the ChessBits homepage >>>> >>>>http://mitglied.tripod.de/ChessBits/index.html >>>> >>>>you can find a live game between Fritz experimental and Shredder experimental >>> >>>Fritz is no experimental version and the hardware and time control favour >>>Fritz6a very much. >> >>1)How do you know that Fritz is not experimental version? > >Because the operator told so. :-) > >>I looked at the times and Fritz seems to play without an opening book(otherwise >>Fritz would not use time for theory moves). > >The operator, CBBTEG = ChessBaseBetaTesterErdoganGünes, used his own book. > >>2)The hardware favours Fritz but if it is Fritz6a and not experimental Fritz >>then I guess that one detail favours shredder(the hash tables) >> >>Fritz is using too big hash tables for this time control so I guess that the not >>experimental version needs time to clear the hash tables between moves so the >>fact that Fritz is using 600 Mbytes is a disadvantage at 30 minutes/game. > >I disagree. AFAIK Fritz uses persistant hash tables without any clearing. I read about some blitz games that Fritz lost on time and I thought that clearing the hash tables was the reason but maybe the reason was different. I know that Junior and nimzo are the programs who like most to lose on time and after them Fritz likes to do it. > >>3)How do you know that fast time control favour Fritz against shredder >>experimental version? > >Short time controls with hardware advantage, come on Uri, even you must see >this. ;-) >Except for Hiarcs, only fast searchers are good rapid chess players. > >>The only way to know is by testing both versions. >>Do you have shredder experimental version? > >I am no beta-tester. > >>I think that it is wrong to assume without testing that the time control favours >>shredder only because of the fact that shredder is slower in nps. >> >>Uri > >See various tournament results, it is fact that Shredder profits much from >longer time controls. You are probably right about previous versions but You cannot know from it about last version. You can guess and you are probably right but you cannot know. 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.