Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: New processorgenaration and chessprograms

Author: Robert Hyatt

Date: 20:16:56 02/06/99

Go up one level in this thread


On February 05, 1999 at 18:51:15, Eugene Nalimov wrote:

>On February 05, 1999 at 18:27:42, Dann Corbit wrote:
>
>>On February 05, 1999 at 17:53:57, Eugene Nalimov wrote:
>>[snip]
>>>Here I have Dell dual PII/400 and Digital dual Alpha 21164A/500.
>>>Based on my experience, those machines have almost identical
>>>performance - even on Crafty, which does a lot of 64-bit operations,
>>>Alpha is only marginally faster. My feelings are confirmed by
>>>SpecInt95.
>>>
>>>I know that 21164A is slower than 21264, but Pentium/400 is not
>>>the last processor from Intel, too.
>>>
>>>The real Alpha advantage lays in 64-bit pointers - it's ideal for
>>>huge databases. But not for chess, and not with 32-bit NT.
>>What about 6 piece tablebase files and 100 million position opening books?
>>The greatest advances will come from things of that nature, unless some
>>fundamentally new algorithm is invented.
>>
>>So the Alpha should be good for chess, given a 64 bit OS.
>
>I don't beleive *any* current machine (let alone Cray or some
>1000-CPU animals) will handle 6-man tables in a near future.
>Even if you'll write generator in a better way than mine was
>written (I made some design decisions that simplified it, but
>slowed it down and/or increased its RAM usage), it'll still
>work for a month to generate simple pawnless TB. Size of the
>TB will be ~55 time larger than for average 5-man TB. 10Gb
>for single TB (or 100Gb for complete set - including one TB
>with pawns and all promotion cases) will be too much for a
>reasonable machine in the next few years.


some 6 piece files have already been done on a Cray by Lewis Stiller.
But they are _very_ big as you point out.  But on such a  machine, you
use different algorithms that vectorize better... where the cray can
easily be 1,000 times faster than the fastest PC ever built...



>
>I agree that after those years Alpha will be better suited for
>handling of those huge resources than x86. But I still beleive
>that Intel will resolve its problems with IA-64, as it resolved
>problems with CISC that many beleived will forever harm x86. Or
>that something else will happen. Five years is a very long term
>when you are talking about computers.
>
>Eugene



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.