Author: Marc D
Date: 09:42:26 06/24/05
Go up one level in this thread
On June 24, 2005 at 11:58:19, Eric Oldre wrote: >On June 24, 2005 at 06:48:49, Aaron Gordon wrote: > >>Binary: Default windows Fruit 2.1 binary that comes with the release version >>(compiled by Bryan) >> >>Test method: >>Run fruit21, type: >>uci >>isready >>ucinewgame >>go depth 12 >> >>991583 n/s = Athlon XP (Barton core) 2.7GHz: >>809906 n/s = Athlon 64 2GHz (3200+ 1mb): >>765542 n/s = Pentium 4 (Prescott core) 3.61GHz: >>742800 n/s = Athlon XP 2400+ (2.0GHz) >>714049 n/s = Pentium 4 (Northwood core) 3.0GHz: >>598093 n/s = Athlon XP 1900+ (1.6ghz): >>578746 n/s = Pentium 4 (NW) 2.4GHz: >>434772 n/s = Pentium 4 (NW) 1.8GHz: >>409346 n/s = Pentium 3 (Tualatin core) 1.33GHz: >> >>Here is a (poor) graph of the results: >>ftp://spdycpu.dyndns.org/fruit21-bench.jpg >> >>The (NW) means Northwood a core Pentium 4, the (P) is a Prescott Pentium 4 >>(latest core). As you can see, it would only take a Pentium 4 Northwood at >>3.2GHz to equal the 3.6GHz Pentium 4 Prescott. The Athlon 64 result is 32 bit >>under a 32 bit OS (Fruit doesn't take advantage of 64 bit processors yet). > >results on my laptop (Pentium M 735 @ 1.7ghz) > >info depth 12 seldepth 30 time 5348 nodes 4150768 nps 776135 >info time 5348 nodes 4150768 nps 776135 cpuload 1000 >bestmove g1f3 ponder b8c6 I got slightly lower results on my Pentium M 745 1.8 Ghz info depth 12 seldepth 30 time 5468 nodes 4150768 nps 759102 info time 5468 nodes 4150768 nps 759102 cpuload 1000 I am somewhat disappointet :-( Marc
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.