Author: Albert Bertilsson
Date: 04:10:37 12/10/03
Go up one level in this thread
On December 10, 2003 at 04:42:10, Russell Reagan wrote: >On December 10, 2003 at 02:50:45, Tony Werten wrote: > >>On December 09, 2003 at 10:29:00, Albert Bertilsson wrote: >> >>>Hi! >>> >>>I've finally got around verifying perft(10) from the starting position and it >>>is: 69352859712417, that is 69*10^12 nodes. >>> >>>It took 85 hours for my four computers to calculate it. If some people has a >>>couple of spare cpu cycles perft(11) would be no mayor problem to calculate. >> >>That's over 226 Mn/s. Quite high. > > >Wow. It sure is. Even for 4 computers working on it. The fastest public perft >calculator that I know of is Movei, which calculates at about 15 Mnps on my >Athlon 2400+ (which is faster than any of the hardware mentioned). I have heard >that a private version of Yace is faster. Even if Yace is twice as fast, running >on four Athlon 2400+'s (which would be faster than the mentioned hardware), we >are still only at 120 Mnps. Only two of the computers were running 24/7 too. >With a generous estimate of about 7GHz of the total number of cycles executed >between all four machines per second (and assuming they were running 24/7), >we're looking at about 40 cycles per node. That is a lot of asumptions, and we >are still 100 Mnps short. > >So that's either very impressive, or something is wrong :) > On the 2GHz P4 (not the fastest machine, but the one I use right now) Sharper get 18Mnps. What makes the numbers look so good is of course the usage of hash tables, without them it would be a whole diffrent story. /Regards Albert > >>Tony >> >>> >>>Hardware used: >>>XP2100+ (256MB hash) >>>XP1700+ (128MB hash) (not working on the problem 24/7) >>>Celeron 800 (128MB hash) (not contributing very much =)) >>>P4 2GHz (256MB hash) (not working on the problem 24/7) >>> >>>/Regards Albert
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.