Author: Dann Corbit
Date: 10:02:35 06/20/01
Go up one level in this thread
On June 20, 2001 at 03:41:45, pete stein wrote: > >> >>An attempt was made to model Deep Blue. I suspect it lacks something besides 4 >>orders of magnitude in speed but it's an interesting experiment nonetheless. On >>the other hand, I don't look for it to give Kasparov a beating any time soon. >> >>This is always a good place to look for recent free downloads: >>http://home.t-online.de/home/thomas.j.mayer/Quark-Englisch/Winboard-Newsticker/Winboard-Update-List/winboard-update-list.html > >Hi Dann, > >An attempt to model Deep Blue ???? That´s more than interesting. >That must be either the best thing since sliced bread, >either a dead hoax. Where´s that baby available for download ? I have the source, but I have not bothered to compile it. Deep Blue did not use null move, so you might say, just strip out null move and there you have it. I imagine there were a few other small alterations. There are a few small problems. The "immitation" can't do one billion NPS peak with 200M NPS average throughput. The "immitation" does not have 4000 tunable paramters in the eval. Little things like that. > >I saw on Thomas Mayer´s site you have ready-2-run versions of Beowulf on your >site but where exactly ? Generally, and new chess engine work will be found here: ftp://cap.connx.com/pub/chess-engines/new-approach/
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.