Author: Slater Wold
Date: 23:30:03 04/17/02
Go up one level in this thread
On April 17, 2002 at 19:58:09, Jeremiah Penery wrote: >On April 17, 2002 at 14:48:46, Slater Wold wrote: > >>On April 17, 2002 at 12:21:26, Shep wrote: >> >>>On April 16, 2002 at 19:40:53, Slater Wold wrote: >>> >>>>AMD 2x1.53Ghz >>>> >>>>Dhrystone ALU: 5664 MIPS >>> >>>>AMD 2x1.73Ghz >>>> >>>>Dhrystone ALU: 9651 MIPS >>> >>>Can this be correct? After all, you pumped up the MHz by only 2x16% or so, yet >>>get a 80% speedup. >>> >>>--- >>>Shep >> >>It is using the *same* exact program. (Sandra) And I promise, these are the >>figures reported. > >Did you use a different version of Sandra? No. It is the *EXACT* same version, on the exact same HD. >>Also, that's the ALU, not the FPU. Chess performance comes from the FPU. And > >Chess programs do not use floating point - they do almost all work in the ALU. I've always understood they did. Perhaps I have been told wrong. Well, I've been told "Excellent FPU Performance == Greater Chess Performance".
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.