Author: Uri Blass
Date: 22:19:25 06/12/01
Go up one level in this thread
On June 12, 2001 at 22:52:46, Chessfun wrote: >On June 12, 2001 at 21:29:36, Jonas Cohonas wrote: > >>On June 12, 2001 at 18:02:34, Chessfun wrote: >> >>>On June 12, 2001 at 15:06:43, Dann Corbit wrote: >>> >>>>I guess that Deep Junior has not been tested yet? Probably too new. >>> >>>Deep Junior is over 18 months old and is known to be slower on a single cpu than >>>Junior 6. Whether that is true of the next release of Deep Junior I don't know >>>but it still has not been released by Chessbase AFAIK. >>> >>>Sarah. >> >>Deep Junior in my experience is just as good on a single cpu as Depp Fritz is in >>terms of speed if run on longer timecontrols! >>I guess that when DJ was released the low processor speed avaiable to the >>average joe did not give DJ the boost of todays hardware, thus the statement "DJ >>only runs on dual" >> >>Regards >>Jonas > >Two things. First from what I saw of a previous post of yours The version you >are using is the original DJ not patched. I did post a question about it which >went unanswered. IMHO that version cannot compete against Deep Fritz on >identical single (as the SSDF use) cpu hardware. > >The statement never was "DJ only runs on dual". AFAIK the statement was simply >that Junior 6 and Deep Junior are in effect the same program with one optimized >to run on a dual processor. Therefore Junior 6 would produce about 20-30% more >NPS and therefore would be stronger on a single cpu. I remember that I understood from a previous post of Amir that the difference in nodes is only 5%-10% and that he was surprised that the difference was so small. He explained in that post that the reason for the fact that it is not the case for Crafty is the fact that Crafty is not optimized for one processor. The difference for Crafty is only about 0.1% 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.