Author: Robert Hyatt
Date: 16:56:05 01/22/00
Go up one level in this thread
On January 22, 2000 at 19:11:40, Tom Kerrigan wrote: >On January 22, 2000 at 17:44:44, Robert Hyatt wrote: >>No. Because deep blue can't play "without its main speed hardware". It is >>based totally on hardware. > >You know, you say this a lot, and it makes perfect sense at first glance. > >However, if you think a little more about it, it's actually stupid. > >Making software is easy. Designing hardware is hard. Trying out software is >easy. Trying out hardware takes millions of dollars. If FHH didn't try his >algorithms in software before comitting them to silicon, well, that's a huge >waste of time and money. > >I think it's fairly safe to assume that DB can play without the DB chips... >otherwise somebody was being horribly irresponsible. > >-Tom Stupid is a good choice of words here. Because of _course_ it can run without the chips. Just like Intel runs the next-generation X86 without the chip. Of course the fact that it runs 1 billion times slower doesn't matter when testing, but it would certainly matter if you try to _use_ the thing... And the parts of the thing he worked on after the chip was finished was the tuning... all the weights and patterns are software, but the stuff to run the code is hardware. How would you _play_ against something that is running 1 billion times slower than it normally would, and using 1/480th of the number of chess processors, and 1/32nd of the number of SP processors? Sounds hopeless to me... for anything other than verification that it does what it is supposed to do hardware cycle-wise...
This page took 0.01 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.