Author: enrico carrisco
Date: 21:02:12 10/12/04
Go up one level in this thread
On October 12, 2004 at 21:08:30, Russell Reagan wrote: >On October 12, 2004 at 19:32:44, stuart taylor wrote: > >>I thought we were all waiting for 64-bit programming which might have made it 4 >>times as fast, wouldn't it? Or at least a bit better, and I just can't >>understand why it is not designed to take advantage of the most recent hardware >>cpu advances, of about a year ago. > >I wouldn't expect Tiger to be released as an AMD64 program at this time. I doubt >it is significantly faster on AMD64, and it doesn't make sense from a business >point of view at this time. > >Tiger is not a bitboard program. Non-bitboard programs are very hit and miss >when it comes to the "64-bit bonus". Some get a modest speedup and some get no >speedup, but I don't know of any that gets a significant speedup. Some get NO speed-up? Which? I would venture to guess that NO program (chess or otherwise) would see less than 15-20%, MHz for MHz vs. the Athlon XP. -elc. >It may well be >that 32-bit hardware is the fastest for Tiger at this point in time, especially >considering that the best compilers for AMD64 are either still in beta form (ex. >Microsoft) or something like gcc. I also don't know of any chess program which >runs four times faster because of AMD64 hardware. Crafty, for instance, runs >about 60% faster on the Opteron compared to an equivalently clocked 32-bit >Athlon. It is highly improbable that Tiger would gain more from 64-bit hardware >than Crafty, unless Tiger has become a bitboard program, but I *seriously* doubt >that :) > >From a business point of view it doesn't make sense either. Not many people have >AMD64 machines, and even fewer have Windows for AMD64 (and most of them have a >beta version).
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.