Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Go Brutus!!

Author: Robert Hyatt

Date: 13:25:46 11/24/03

Go up one level in this thread


On November 24, 2003 at 16:19:20, Slater Wold wrote:

>On November 24, 2003 at 16:14:12, Robert Hyatt wrote:
>
>>On November 24, 2003 at 15:05:53, Pete Rihaczek wrote:
>>
>>>I for one am very excited to see Brutus in the lead. This is an exciting advance
>>>in chess computing and FPGA computing in general. With the ability to add
>>>knowledge without the usual penalty, some version of this is the odds-on
>>>favorite to be the world's strongest chess machine. Such a system was a logical
>>>step after Deep Blue II had shown the advantages of computing in hardware. Can a
>>>Kasparov-Brutus match be far away? Well done Dr. Donninger!
>>
>>
>>I don't think it is _that_ revolutionary.  IE a single FPGA board and
>>computer together search about 2.5M nodes per second, according to comments
>>by them when we have played a few skittles games on ICC.  A dual-CPU opteron
>>is faster than that, as a reference point.
>
>Bob, Bob, Bob...
>
>A PC can get 2.5M nps, sure.  But with what program?  HIARCS?
>
>A FPGA can get 2.5M nps, and then you can stuff 100,000 lines of chess knowledge
>in it, and still get 2.5M nps.  That's a big difference.

The problem is this.  While you are "stuffing those 100,000 lines of chess
knowledge in it" the PCs are _still_ getting faster.  The FPGAs are simply
not _that_ fast.  See my comparison with Belle and Deep Thought.  Those
represented _quantum_ leaps in performance.  The FPGA does not, yet, if it
ever will.


>
>>yes, I know that he is running with four machines, two FPGA cards per machine
>>in Graz.  But then again, 8-way opterons are around as well.  I'm hardly
>>"anti-hardware" but the benefits to using hardware normally far-surpass
>>readily available general-purpose computers.  IE belle did 160K when the
>>fastest competitor was 20K (Cray Blitz).  Deep thought went to 1.5M when we
>>were at 200K with the fastest hardware Cray had at the time.  The FPGA
>>approach doesn't have that significant speed advantage.  IE a single card
>>at 2.5 M nodes per second is within reach of a single processor machine
>>today...



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.