Author: blass uri
Date: 04:08:55 12/28/98
Go up one level in this thread
On December 26, 1998 at 19:33:01, Bertil Eklund wrote: >On December 25, 1998 at 11:03:25, blass uri wrote: > >> >>On December 24, 1998 at 16:13:09, Jim Paul wrote: >> >>>This is related to my previous CM6000 posting. >>>What would most people regard as the typical rating improvement in blitz >>>compared to standard time controls for most PC chess programs these days? >>> >>>I know it heavily depends on whether the program is mainly a "heavy calculator" >>>or a "knowledge based" program >> >> >>No, It is not truth >>Hiarcs7 is a "knowledge based" program and it is good at tactics and did very >>good results in blitz games >> >>, ie. I'd expect Rebel and Hiarcs to exhibit less >>>of a blitz rating increase (if any) compared to their standard 40/2 ratings (on > > > > > > > > > > > > >>>SSDF or Selective Search rating lists) for example. Whereas Fritz and probably >>>CM6000 would have higher ratings at blitz. >> >>CM6000 is not very strong at blitz and is better at longer time control. >>> >>>I'd expect CM6000 to play about 100 ELO points higher at blitz than it's SSDF >>>rating (when it gets one anyway !!). That would make it around 2550 at blitz ?! >> >>In other words you say that you expect chessmaster6000 to have 2450 at long time >>control. >> >>I expect chessmaster to have better results( more than 2550 ssdf rating with a >>chance to be number 1 in the ssdf list) >> >>Uri >The best programs can easily compete with the worlds best players in blitz. >In speed-chess on a fast pc they can compete with everyone maybe except from the >top five or so. In tournament-level it is still along way to go.... > >Bertil I did not mean to rating against humans but to rating in computer-computer games. Chessmaster is not strong at blitz relatively to other top programs at blitz. 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.