Author: blass uri
Date: 08:37:45 05/10/00
Go up one level in this thread
On May 10, 2000 at 10:41:56, Mogens Larsen wrote: >On May 10, 2000 at 09:06:55, blass uri wrote: > >>The question is what do you want to compare. > >Yes that's true, which wasn't entirely clear IMO. I only see a lot of blitz >results and a few standard games that can't be compared correctly. > >>I agree that they cannot be used for strength assessment in different positions >>but I think that comparing programs in the nunn match is interesting. > >Yes and I've never said otherwise. It's just wrong to assume that Nunn positions >are better or more fair than any other set available. > >>I do not think that learning had a big influence on the results. > >Maybe not, but you don't know. If you run successive blitz matches (20-80 games) >and forget to clear learning then you're in trouble, especially if you compare >them with results where you did remember to clear learning. That should be >_very_ obvious. I've asked twice if the learning was cleared consistently during >blitz matches, no answer has yet been offered. > >>The different results 9:0 for Fritz and 11:9 for crafty were because of the fact >>that one engine was slower and the usual result is about 15:5 > >That's is your opinion and not a fact at all. None of the unusual results were >explained properly. Due to statistics there's no such thing as an usual result. >You should know better. Most of the results were close to 15:5 so when something is significantly different it is natural to call it an unusual result and it is logical first to check if there is a problem with the games with these results. It was discovered that Crafty was slower in the games of 9:0 and that Frit was slower in the 11:9 games. This is probably the right explanation to the unusual results. Uri
This page took 0.01 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.