Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Doesn't "faster" mean "stronger"(dif. versions of the same program)?

Author: Robert Hyatt

Date: 09:04:25 01/11/01

Go up one level in this thread


On January 11, 2001 at 08:57:53, Christopher R. Dorr wrote:

>Nonsense? Hardly. First off, what does it hurt? You seem to be acting as if the
>testing itself harms something. It clearly does not. It adds information to our
>'knowledge set', and informs a consumer that, if you are using a
>single-processor box, you probably should *not* get Deep Fritz. I'm glad I now
>know that. Had this testing revealed that DF was *better* on a single processor,
>you would not have complained, because the consumer would then be able to buy a
>better program, and this added knowledge would help him receive a stronger
>program. Had they tested DF and found it better on a single processor, would you
>have objected to the test then?
>
>
>So how can the *testing* be bad, when it's only a particular result that you
>don't like? The only way to find out is to test. The testing was done, and we
>now have more information than we did before. How can that *possibly* be bad?I'm
>glad it was done, and appreciate the time and efforts of the testers.
>
>Chris


The only question is why is DF worse on a single processor machine?  I don't
suffer from that problem, as my SMP code simply isn't used...  I think that the
SMP version runs about .1% slower on a single cpu than the non-SMP version.  I
don't think that difference can be measured in terms of Elo.



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.