Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Participants WCCM 2000

Author: Vincent Diepeveen

Date: 05:56:49 08/15/00

Go up one level in this thread


On August 15, 2000 at 01:00:36, Uri Blass wrote:

>On August 14, 2000 at 23:19:17, pavel wrote:
>
>>On August 14, 2000 at 22:16:24, Uri Blass wrote:
>>
>>>On August 14, 2000 at 21:16:55, Martin Grabriel wrote:
>>>
>>>>Fritz
>>>>Tiger
>>>>Nimzo
>>>>Rebel
>>>>Junior
>>>>
>>>>My pick of top 5 (in that order)...
>>>
>>>Did you consider the fact that Crafty has a better hardware(probably at least
>>>twice faster than the opponents)?
>>>
>>>I guess that Crafty will be in the top 5 but I do not guess which program will
>>>be number 1.
>>>
>>>Uri
>>
>>
>>how do you know crafty has a better hardware?
>>do you know what hardware fritz, junior, rebel, nimzo tiger are using?
>>or are you just assuming.
>
>I know that Crafty will use an alpha and I know that other programs cannot use
>an alpha.
>

Diep can use alpha, but regrettably i can't get one :)

Basically any program that's in ANSI-C and non-chessbase
can use a program.

>>crafy's hardware is better comparing to "other hardwares' it uses  eg, its
>>deffinately better than AMDs since crafty doesnt run *good* on them. because of
>>the memory cache sizes and e.t.c
>>
>>on the other hand if crafty wins this tourny (or even does a good result), I
>>know what "some" people over here is going to say  "oh but crafty had a better
>>hardware, thats why" ........duh!! who stoped you to get a better one? :) *grin*
>>
>>pavel :)
>
>
>There are no sponsors to give every computer the hardware it wants and it is
>possible that you could see more alphas if there was a sponsor to pay for the
>alphas.

if there are 10 alphas 21264s there i think some people might consider to
an alpharob to live quietly the rest of their life :)


>I guess that programmers also did not care about the fact that their program
>cannot run on an alpha because most of them have no alpha at home.

I could never test the 21264 so far, only the 21164.
633Mhz 21164 is as fast as a PII at 380Mhz for DIEP.
that was with the gcc compiler however which is inferior to the DEC
compiler by a large margin (i have no idea how much, 20%?).

However where 21164 had only 8kb L1 cache,
PII has 32kb L1,
21264 has 128kb L1 and a new and very advanced branch prediction
strategy and not 3 but 4 instructions a clock.

So obviously 21264 is already 33% faster in advance because of the
extra instruction. Add to that more for better branch prediction etcetera.

Perhaps 2.0x as fast as a PII. Note that PIII is 17% faster for DIEP as a
PII, so also there are big advances.

>Uri



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.