Author: Dann Corbit
Date: 15:37:21 01/30/02
Go up one level in this thread
On January 30, 2002 at 18:21:53, Tony Werten wrote: >On January 30, 2002 at 14:15:32, Dann Corbit wrote: > >>On January 30, 2002 at 14:00:32, David Rasmussen wrote: >> >>>On January 30, 2002 at 09:43:49, Vincent Diepeveen wrote: >>> >>>I don't know anything about what Bob has stated, but I agree that 18 ply >>>fullwidth was/is impossible, even with 200 Mn/s. >>> >>>Please email me how you do singular extensions, Vincent. :) >> >>I've done 18 ply searches with a PC. What makes you think it is impossible for >>a machine that can peak at one billion nodes per second? >> >>They did not use null move, but neither did they blindly search the whole tree. > >I'm very sure you didn't get 18 ply fullwidth. > >They had a branching factor of 4. You don't get 18 ply with that. Either you >don't have that BF (pruning) or you don't get 18 ply ( just some extensions do) Let's suppose (for a moment) that the hardware searches on the ends of the branches were considered the equivalent of hash table hits. Hence, they would be pretty useless for PV and things of that nature without extra work, but they would verify the value of the ending node. What will that mean for the throughput of the search? What does "fullwidth" mean in this context? Obviously, it cannot possibly mean that every branch of the tree is examined exhaustively to 18 ply (which would be absurd). Was 18 ply claimed for the opening? For the middle game? For the endgame? As a broad average? If (for instance) I take crafty or chess tiger or pepito and let it run overnight, and 18 ply search is not unusual in the least bit. How does that compare with deep blue at 10 minutes? I don't think it is certain (either for or against) and I don't think that the terms used are certain. Just like "ply" means 20 different things to 20 different programmers. Even "nodes" does not always mean the same thing. In computer chess, there really are no precise definitions, because nobody has done a standardization. Therefore, the terms are very general. I am not sure (either way) what the real search throughput was. At any rate, it was good enough to beat Kasparov.
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.