Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Vincent, will you comment on Diep hardware for Maastricht

Author: Robert Henry Durrett

Date: 09:57:35 06/29/02

Go up one level in this thread


On June 29, 2002 at 10:54:09, Vincent Diepeveen wrote:

>On June 29, 2002 at 10:49:55, Vincent Diepeveen wrote:
>
>>On June 29, 2002 at 10:31:21, Robert Henry Durrett wrote:
>>
>>even worse, they will even manage within a year to run on a supercomputer,
>                       not even manage obviously.
>>simply because fritz is written in assembly.
>
>rewriting a pc assembly program to a cc-NUMA supercomputer is not easy.

Who's afraid of a little work?  Many people like challenges.

Bob D.

>
>
>>>On June 28, 2002 at 23:26:47, Robert Hyatt wrote:
>>>
>>>>On June 28, 2002 at 20:33:04, Robert Henry Durrett wrote:
>>>>
>>>>>On June 28, 2002 at 20:25:18, K. Burcham wrote:
>>>>>
>>>>>>
>>>>>>
>>>>>>http://www.sara.nl/hpc.www/teras/description/
>>>>>>
>>>>>>Chessbase is saying this:
>>>>>>DIEP is expected to run on a 1024-CPU TERAS system. This machine consists of 45
>>>>>>racks and has a peak performance of 1 TeraFlops (10^12 floating point
>>>>>>operations) per second.
>>>>>>
>>>>>>
>>>>>>First of all, is this the system Diep will be running on in tournament?
>>>>>>
>>>>>>If so, what kns do you expect? What is the total mhz you will actually
>>>>>>be using in this system?
>>>>>>
>>>>>>kburcham
>>>>>
>>>>>That's the one they should use for the Kramnik - Fritz match in October!
>>>>>
>>>>>Bob D.
>>>>
>>>>
>>>>Impossible.  It is unix, not windows.
>>>
>>>Wait a minute!  Are you "really" saying that "the Fritz people" are incapable of
>>>making a UNIX version of Fritz?  Are you saying the're incompetent?
>>>
>>>Bob D.



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.