Computer Chess Club Archives




Subject: Re: LCT II Fin4, Deep Thought, and Deep Blue (was Re: LCT II results...)

Author: Jay Scott

Date: 11:14:51 01/07/98

Go up one level in this thread

On January 07, 1998 at 02:35:51, Don Dailey wrote:
>On January 06, 1998 at 18:35:32, Jay Scott wrote:
>>On January 06, 1998 at 17:30:40, Don Dailey wrote:
>>>every way (except raw speed) the Deep Blue team is handicapped so you
>>>can not expect them to compete with the highly tuned micro programs.
>>This is not true. The hardware has one big advantage besides
>>outrageous speed: it's parallel at the transistor level. You can
>>get more evaluation function terms at each node by adding
>>transistors, and adding transistors does not slow you down.
>Hi Jay,
>I consider this a speed issue too.   I concede the quantity issue,
>they do more things and do them all faster.

Well, of course it's a speed issue, but it's a different speed
issue. Even if the hardware were slower than software in nodes
per second, you'd still design a more knowledgeable program for
hardware because the tradeoff is so happy. One speed
issue is that everything's faster in hardware, another
speed issue is that there's more chance for fine-grain
parallelism in hardware. I don't think we're disagreeing here.

Designing hardware rather than software loosens up all the
rules and lets you reach new design points (in return for the
headache of dealing with silicon rather than bits). So I'm
not sure of anything much about Deep Blue until they tell us.


PS If you want to see something really funny, give Chris
Whittington some fab access. :-)

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.