Author: Thorsten Czub
Date: 14:20:03 02/28/98
Go up one level in this thread
>>If we want to measure something we should make sure that anything is at >>its best. > >So why not give all programs machines with 64 MB RAM and see who gains >most from it? You know the answer ... > > >Giving all machines 16 MB RAM and see who suffers most could be another >experiment ... > >Moritz RIGHT !!! This is the key to the whole discussion ! By giving the slow programs less ram and slow machines advantages the slow ones !!! Exactly this was my point ! It is not the same !!! It is another shoe ! In my experiments and maybe with the 16 MB hash and p90 machines the knowledge programs win. And maybe also with 50 ELO ahead or even more (as we have seen in the elo-lists before). When I now change the parameters with advantages for the fast-searchers the slow-ones cannot compensate.
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.