Computer Chess Club Archives

Search

Messages

Subject: Re: OT: P4- 3 GHz with hyper-threading

Author: Gian-Carlo Pascutto

Date: 12:23:48 11/02/02

Go up one level in this thread

On November 01, 2002 at 14:22:10, Eugene Nalimov wrote:

>Sorry, after I wrote previous message, I remembered about parallel
>non-determinism, so I re-run the tests 5 times on the same system and calculated
>average.
>
>mt=2: times are 42, 43, 39, 40, 38, average is 40.
>mt=2: times are 36, 31, 37, 29, 38, average is 34.
>
>So average speedup is even greater than I estimated earlier, something like
>~15%.

According to those numbers, the speedup is 18% with a (2sd standard error)
margin of 15%.

With this error margin, it is somewhat premature to talk about 'double
digit' figures.

--
GCP

• Re: OT: P4- 3 GHz with hyper-threading Robert Hyatt 14:49:43 11/02/02
  • Re: OT: P4- 3 GHz with hyper-threading Gian-Carlo Pascutto 15:08:49 11/02/02
    • Re: OT: P4- 3 GHz with hyper-threading Vincent Diepeveen 07:09:19 11/03/02
      • Re: OT: P4- 3 GHz with hyper-threading Robert Hyatt 09:49:51 11/03/02
    • Re: OT: P4- 3 GHz with hyper-threading Robert Hyatt 20:45:39 11/02/02
      • Re: OT: P4- 3 GHz with hyper-threading Anthony Cozzie 23:00:41 11/02/02
        • Re: OT: P4- 3 GHz with hyper-threading Robert Hyatt 09:53:39 11/03/02
        • Re: OT: P4- 3 GHz with hyper-threading Gian-Carlo Pascutto 00:29:09 11/03/02
          • Re: OT: P4- 3 GHz with hyper-threading Robert Hyatt 09:56:06 11/03/02
          • Re: OT: P4- 3 GHz with hyper-threading Vincent Diepeveen 07:01:15 11/03/02
            • Re: OT: P4- 3 GHz with hyper-threading Robert Hyatt 09:58:51 11/03/02

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.