Author: Mark Young
Date: 16:58:28 12/20/03
Go up one level in this thread
On December 20, 2003 at 08:50:41, Uri Blass wrote: >I assume for the discussion that the GM's do not prepare specially against the >program and the program also does not do it against the GM's. > >1)let talk about blitz time control > >Fritz3 on a pentium90 or even pentium66 (I am not sure about the facts) >already won a blitz tournament as first place with kasparov(probably in 1994) so >it was clearly super GM strength in blitz. > >I believe that more than hundred of free engines on a fast hardware are better >than Fritz3 on a hardware that is at least 20 times slower. > >Does it mean that they all are GM strength against humans or maybe they are >better only against computers and not against humans. > >2)Let talk about 25 minutes per game > >Genius3 on p90 got super GM performance >I guess that more than 80 engines on fast hardware are better than Genius3 on >p90 at 25 minutes per game. > >Does it mean that they are GM strength at that level? > >3)let talk about 120/40 time control. > >Tiger14 on hardware that is about 3 times slower than the best hardware of today >got super gm result in argentina so it is probably GM strength. > >How many free engines can beat tiger14 at 120/40 with 3:1 hardware advantage? > >Are they also GM strength against humans? > >Uri Hello Uri, I was wondering if it has ever been shown that Comp vs Comp results don't correlate well with games played against humans? I don't think it has, so I would be of the opinion that a program that does well against a known GM strength program(s) would itself be a GM class program based on those results alone when playing humans. I think the opposite is also true from what I have seen. I don't think there is a program including crafty that is better against humans, but plays worse against computers.
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.