Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Sorry about ignorance. Are 64-bit comps X2 speed for chess?

Author: Robert Hyatt

Date: 09:45:59 12/09/03

Go up one level in this thread


On December 09, 2003 at 11:49:19, Mathieu Pagé wrote:

>On December 09, 2003 at 11:16:02, stuart taylor wrote:
>
>>Is 2.2 Ghz. of a 64-bit computer a similar speed for chess as is 4.4 is it were
>>a 32-bit one?
>>If not, what?
>>S.Taylor
>
>Hi!
>
>No, it is not.
>
>64-bit computer are not twice as fast as 32-bit ones. The number of bit
>represent the natural lenght of an number on a cpu. Since chess engines use lot
>of 64 bits numbers they will run faster on 64 bits machines because on 32-bit
>machines they have to do some trick to do 64 bits maths that are natural on a 64
>bit cpu.
>
>I dont think the improvement will be in the range of 2x speed up. Anyway it will
>vary from diffrents engines.
>
>Mathieu


This is a tough one to call.  IE write a test program that shifts 64
bit values over and over.  The 64 bit machine will be _more_ than twice
as fast.  Think about how you would shift a 64 bit value broken down into
two 32-bit chunks.  Ditto for comparing 64 bit values.  So some ops are
more than 2x faster, some are exactly 2x faster, and then there are the
ops that don't need 64 bits, like "loopz loc" for example.

For that reason, it is likely that different programs would speed up
different amounts moving from 32 to 64 bit cpus.  Other instructions
are not 64 bits at all, other than in their addressing.  IE jnz loc; for
example.  The opteron I am using is running with a 48 bit virtual address
space, a 40 bit real address space.  That gives a physical address space
of one terabyte (256 times 4 gigabytes).  A lot of memory.  Why the 48 bit
virtual address space is beyond me, as that lets me address 256 terabytes
and probably burn up a few disk drives paging.  :)





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.