Author: Jeremiah Penery
Date: 02:21:02 01/25/00
Go up one level in this thread
Whoops, didn't finish. :P On January 25, 2000 at 05:06:13, Jeremiah Penery wrote: >On January 25, 2000 at 04:35:33, David Blackman wrote: > >>On January 24, 2000 at 12:20:12, Robert Hyatt wrote: >> >>>On January 24, 2000 at 09:57:58, Amir Ban wrote: >>>>"I'll bet that I have several evaluation terms that are not practical for them >>>>to compute." >>>> >>>>Amir >>> >>> >>>Let me repeat also: "There is _nothing_ you can do in software that they can't >>>do in hardware in _far_ less time. _absolutely nothing_." That is the benefit >>>of doing what they did in hardware. Never a question of "can I afford this or >>>will it slow me down too much?" Only a question of "is this worth the time it >>>will take to design it?" >> >>I wonder if there is some stuff that you want to eval in about 1% of positions. >>You just have a quick if statement mostly, but in 1% of positions it triggers >>and you do 500 lines of code. In a software program, this costs you very little >>except RAM, which is cheap. > >If you do 200k NPS, 1% of that is still 2000. So you're running 500 extra lines >of code 2000x/sec. And if you put several of these things in, you'll probably >end up doing it at least 10-20K x/sec. This is going to slow you down quite a >bit. You'll no longer be doing 200k NPS, for sure. :) > >>In hardware, it costs you for chip area and partly for power, even for those 99% >>of positions that don't use it. >> >>If i was doing hardware, i'd avoid most of these. In software, i'd put them in >>if i had the time and the knowledge. >> >>I can't think of any good examples right now, but i'm sure the slow/smart >>brigade use plenty of them. > >If you were making a machine to play against arguably the best chessplayer >_ever_, would you want to risk _not_ putting them in, just to avoid some extra >chip area/heat, when you could mis-evaluate something and lose a game? I don't know whether DB had any of these things in it or not. I'm just guessing that if it could be done, they probably at least tried it. DB-1 supposedly had a bunch of knowledge, but we saw what happened to it when it played Kasparov. He rolled it up into a little ball, even though it was searching 100M NPS. DB-2, searching 200M NPS, probably wasn't searching effectively much deeper than DB-1, since DB's branching factor was 5-6. We saw how much better it played than the DB-1. Most of that improvement came from the evaluation. It's obvious that it had to have a lot of it, covering a lot of different areas, to even be able to come fairly close to Kasparov. If it didn't, it would've lost, as it almost did anyway in a couple games.
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.