Author: Dann Corbit
Date: 09:47:13 06/15/05
Go up one level in this thread
On June 14, 2005 at 04:31:51, Uri Blass wrote: >On June 14, 2005 at 03:59:08, Fabien Letouzey wrote: > >>On June 13, 2005 at 04:45:35, Madhavan wrote: >> >>>>Fruit lost to AnMon (tactical blunder, but AnMon was a pawn up already) and >>>>Chess Wizard (the result was never in doubt since the opening, Pirc by Fruit). >>>>Fruit should have only drawn the game against Pharaon! >> >>>I didnt expect this,if only Fruit had been run on a faster hardware,it would >>>have won atleast 2 of these 3 games. >> >>It's easy to overestimate the impact of hardware. Fruit was maybe twice as slow >>as it could have, which translates to -50 Elo from what I have read. > >I think that it is 70 elo at blitz time control and there is diminishing returns >so if you use the ssdf you may see only 50 elo. Here is output on two of my systems for Fruit Massy (with the compile that is fastest for the machines in question): MSVC compile on 2.2 GHz AMD Athlon 64 bit with default hash: info depth 14 seldepth 40 time 32406 nodes 23264935 nps 717921 Intel compile (with PGO performed on this machine) on 950 MHz AMD 32 bit (very similar to the Massy machine): info depth 14 seldepth 40 time 69920 nodes 23264935 nps 332736 46% of 64 bit machine's speed (looks about commensurate with CPU speed difference).
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.