Author: Robert Hyatt
Date: 06:06:39 05/27/98
Go up one level in this thread
On May 27, 1998 at 04:58:48, Ernst A. Heinz wrote: >On May 26, 1998 at 23:53:22, Robert Hyatt wrote: > >>On May 26, 1998 at 21:25:13, Mark Young wrote: >> >>>Anand vs Fritz in Frankfurt >>> >>>A spectacular event will be held on June 20, 1998 at 8 p.m. in the City >>>Hall of Frankfurt-Zeilsheim. A special multiprocessor verseion of >>>Fritz5, runing on a Siemens Primergy, will take on the world's number >>>two player in a rapid chess match. The match comes at the end of the >>>strongest tournament ever staged, the "Siemens Nixdorf Giants", with >>>Garry Kasparov, Vishy Anand, Vladimir Kramnik and Anatoly Karpov (June >>>17 to 19). The Siemens Nixdorf Primergy 460 is commercially available >>>server with a dual Pentium II board with two 333 Mhz CPUs. Each has 256 >>>Mbyte of RAM, which is useful, since at the expected speed Fritz will >>>fill 200 to 300 MB of transposition tables between its moves. There will >>>be live coverage of all events by the Lost Boys >>>(www.lostcity.nl/chess/ccs/frankfurt.html), with daily wrap-ups and >>>games on this web site (www.chessbase.com). >>> >>>From www.chessbase.com/news.htm >> >> >>would be interesting to see who/how/what they've done with the parallel >>search, since none of 'em have been "practicing" parallel search. I >>have >>an idea. :) > >Bob, > >You are obviously speculating that Frans Morsch does similar things >as you do in your "SMP-Crafty". This is certainly quite probable. > >During the 1997 WMCC in Paris I talked a good deal with Frans who is >a really interesting person. He openly admits to have learnt a lot >while studying the source code of "Crafty". Moreover, he was very >interested in faster machines like Alphas, x86-SMPs and future >SMPs-on-a-chip. He suspected other commercial authors (especially >Chrilly Donninger) to be actively working on multi-threaded (i.e. >SMP-ready) versions of their chess programs, too. Maybe he therefore >started with "SMP-Fritz" before your public "SMP-Crafty" became >available and has done something completely different? > >Unfortunately, Frans never tells much about the internal details of >his chess software. But both possibilities seem to be equally likely. > >=Ernst= I don't object at all. In fact, they've told me that they took "book learning" right from crafty, with some functions taken line for line. They added "result learning" before I did, as they needed that for the SSDF I assume. I only hate to see them "copy" what I've done... IE there is nothing that says what I'm doing is the "best" that can be done... and when something is copied, there's always the risk of being "blinded by someone else's brilliance"... :) But as far as copying the SMP stuff from Crafty goes, that's a non-issue, or I wouldn't have released source code in the first place... So it doesn't matter to me at all. And since we won't ever see any public revellations about what the commercial guys are doing, it's nothing to really spend any mental energy on anyway. :(
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.