Author: Uri Blass
Date: 13:48:14 08/09/03
Go up one level in this thread
On August 09, 2003 at 13:49:42, Gian-Carlo Pascutto wrote: >On August 09, 2003 at 11:59:36, Jorge Pichard wrote: > >>http://wbec-ridderkerk.nl/ > >1) I don't see Ruffian 1.0.1 in there. > >2) It's 44 games. All you can say is that neither program is a few hundreds >ELO stronger or weaker than the other. Is this surprising? > >I have a nice 50 game match with Deep Sjeng 1.5 beating Shredder 7.04. >This makes Deep Sjeng stronger? I don't think so. a match against one opponent may give misleading results even if you have 1000 games because it is possible that A can get 600-400 against B,B can get 600-400 against C and C can get 600-400 against A. I think that games against many opponents give a better picture. > >3) Uri: take a look at Yace's performance in both tournaments. >You still think the same? I do not know. Not the same yace played in both tournaments but it does not change my opinion that Ruffian has a better book. I know that Kurt also found that Ruffian performs better relative to the commercial programs when every program use its book and not when they use the same book(I do not mean that Ruffian is better than the commercial programs but thatRuffian is relatively better when it use its book and not when a common book is used for both programs). The fact that it is possible to explain Ruffian's superior result by luck does not mean that it is the only possible explanation. I also compare Deep Sjeng with Ruffian1.0.1 and not with Ruffian1.0.5 Ruffian1.0.5 is not free and the question if it is better than 1.0.1 is an open question. 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.