Author: Yen Art Tham
Date: 10:25:44 09/07/02
Go up one level in this thread
On September 06, 2002 at 18:46:31, Slater Wold wrote: >Yet Another Speedup Post! > > >Overnight I ran the BS2830 suite on my computer (I only used the first 26 >positions). I ran each position for 15 minutes using 1 and then 2 CPUs. > >I will supply more "facts" later, but here are some interesting tid-bits: > >Number positions not solved by 1 CPU: 10 >Number positions not solved by 2 CPU: 9 > > >Average NPS speedup over 26 positions: 1.48x >Average Solve Time speedup over 14 positions: 1.42x > > >Number of positions with a negative NPS speedup: 0 >Number of positions with a negative Solve Time speedup: 3 > >Slowest NPS speedup: 1.30x (#4) >Fastest NPS speedup: 1.65x (#10 & #19) > >Slowest Solve Time speedup: 1.03x (#7) >Fastest Solve Time speedup: 2.03x (#24) > > > >More to come........ Try it with DF7. Maybe a stronger program will have better speedup. Try it with AMD and Intel boxes.
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.