Author: K. Burcham
Date: 12:01:40 02/28/02
Go up one level in this thread
On February 28, 2002 at 14:21:54, Dann Corbit wrote: >On February 28, 2002 at 05:39:56, Aaron Gordon wrote: > >>I know a few of you here are interested in the AMD vs Intel thing and just speed >>in general. Well, here you go. Crafty benchmarks of CPU's clocked beyond >>anything thats out currently. This should show us what we can expect in the near >>future as clock speeds ramp up even further. >> >>Just got a friend to test Crafty on his new Northwood P4-1.8A at >>2465MHz(137fsb). He's using a Pentium 4 optimized (and profiled) Crafty v18.11. >>I am using Crafty 18.11 profiled for the AthlonXP. The results were taken >>without a crafty.rc file. Here they are... >> >>Intel Pentium 4 (Northwood) 1800MHz overclocked to 2465MHz, 137fsb. >>Total nodes: 104161030 >>Raw nodes per second: 992009 >>Total elapsed time: 105 >>SMP time-to-ply measurement: 6.095238 >> >>AMD AthlonXP 1900+ 1600MHz overclocked to 1860MHz, 155fsb. >>Total nodes: 104161030 >>Raw nodes per second: 1197253 >>Total elapsed time: 87 >>SMP time-to-ply measurement: 7.356322 >> >>As you can see here... the Pentium 4 having twice the L2 and 605MHz more than >>the AthlonXP still gets beaten by over 20%. Not too shabby I figure. :) Can't >>wait for the AMD Hammer (x86-64)... > >And then another year for someone to write a good compiler for it (and even >then, you will end up using some compiler targeting Intel and hoping). That is >one area where Intel steps on AMD's nose. Intel provides excellent compilers >for their chips. AMD hopes that someone else will do it for them. Nobody ever >does, and so we are left with compiling for an Intel target and hoping that it >works well on AMD. > >What compiler is there (for instance) that has AMD Athlon specifically as a >target? >You have to pick some Intel chip target and various other compiler options and >hope for the best. since I have been buying amd, and intel chips, it seems amd is doing very well as far as chess software. with almost every mhz release, amd has beaten intel when comparing price/mhz/kns/performance. and Aaron shows this again. unless you say that behind the scenes, the compiling has been a real nightmare for the programmers, when considering amd. and after these software releases, there has not been anymore amd bugs then intel bugs with the chess software. kburcham kburcham
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.