Author: Robert Hyatt
Date: 19:07:38 07/16/04
Go up one level in this thread
On July 16, 2004 at 11:17:25, Peter Berger wrote: >On July 16, 2004 at 11:06:54, Robert Hyatt wrote: > >>On July 16, 2004 at 06:38:07, Peter Berger wrote: >> >>>On July 15, 2004 at 21:00:18, Robert Hyatt wrote: >>> >>>>On July 15, 2004 at 18:20:48, Peter Berger wrote: >>>> >>>>>On July 15, 2004 at 14:50:18, Robert Hyatt wrote: >>>>> >>>>>>No. It would be at least 10% faster. I have the actual comparison numbers from >>>>>>the same version, one run by Eugene using his compiler, one run by me on linux >>>>>>using 64 bit gcc. Eugene's compiler ate the gcc compiler for lunch. >>>>> >>>>>Before we go into all the other points I missed ;) - what operating system did >>>>>the computer which ran Crafty compiled by Eugene use ? If the answer is Windows >>>>>XP I stand corrected, but that would come as a surprise. >>>> >>>> >>>>Windows XP. :) >>>> >>>>The O/S has no influence on Crafty's speed. IE windows XP is no faster or >>>>slower than linux there. But Microsoft's compiler is far better than GCC. SO >>>>it wins hands down. >>> >>>Eugene's answer was different :). >> >>Maybe we read different posts. :) He mentioned XP64 because you have to have >>that to run the opteron in 64 bit mode, just as we used the Suse AMD64 linux >>release for the same reason... > >OK - and that is where my 4.0 is coming from. Had Crafty run on Falcon's PC it >would have been about 4 times slower than on the Quad Opteron. It was exactly >four times slower on Ingo's PC e.g. I tried to explain this in previous posts, >but maybe didn't make myself clear. > >Peter Right. But don't forget that you are looking at NPS. That is not parallel search efficiency. NPS might be 4x. Actual search will be maybe 3.1X faster, or something pretty close to that...
This page took 0.01 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.