Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Opteron > 970, Good article

Author: Andreas Guettinger

Date: 12:04:22 06/24/03

Go up one level in this thread


On June 24, 2003 at 14:34:12, Dan Andersson wrote:

> I think you missed the point. There was a heap of fudging. Apple did a
>substitute of libraries for the G5. And the library is for single thread
>execution.  They applied oprimizations for the G5. They chose Fortran compiler.
>And by some miracle all the canges favoured the G5 by a whopping 30%. That is
>machitecture for you.
> Their behaviour is even documented. So they stacked the deck big time. No
>amount of arguing will ever change their underhandendness in this case.
>
>MvH Dan Andersson

If you read the documenation then yo see that VeriTest performed this changes,
not Apple. It's up to them how they to their test, I don't see the reason while
suddenly everybody turns to Apple and points the finger at them "You faked it".

cu
Andy



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.