Author: José Carlos
Date: 08:24:09 06/03/99
Go up one level in this thread
On June 03, 1999 at 10:43:26, walter irvin wrote: > i have tested lots of programs on equal and unequal hardware ,in computer >chess speed is king .if you give a program as weak as cm2100 200,000,000 nodes a >sec it would defeat all of the best programs today on reg. pcs k-6 2 450 >ect.plus you could take a program with no evaluation other than material,give it >say 900,000,000,000,000 and it could beat even grandmasters.with this in mind >think how useless it is to rate a program on a p90 against k6 2 450 ,the p90 >will lose so why even play the game,just record loss for p90.now against people >it would be alot more meaningfull a p90 or what ever . I think this discussion could last forever, since there are too static points of view. I'll explain what I mean. Some people is trying to say that uneven competition is unfair to test the programs. That's mostly correct, but not completely. Some programs are more sensible to changes in speed. These programs will do better in a fast machine than others that don't improve their strenght so much with speed. So a World Campionship with PIII-550 would be favorable for some programs and a WC with P90 would be for others. Who can say what is exactly a fair test for programs? I think there's no one. The only fair test for someone is doing the games himself on his/her machine. But a WC is much more than a test. It's a great "circus" where you can find the newst (software and hardware, and not only that). If Deep Blue beats Hiarcs 7, nobody will say that Deep Blue's program is better than Hiarcs' program. The combination of hard, soft, opening book, endig tables and many more things would be (in that case) better in DB than in Hiarcs. We dont want a program test, we want a spectacle. José C.
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.