Author: Tord Romstad
Date: 04:27:06 09/20/03
Go up one level in this thread
On September 19, 2003 at 14:33:57, Andreas Guettinger wrote: >>> >>>I can not share your scepticism. I now from personal sources that my engine has >>>very high NPS on a G5 1.6 Ghz machine. >> >>Thanks! This is good news! And this is a non-bitboard program? >> >>Tord > >It's basically not much more than a move-generator in 0x88 with a simple eval >based on material balance. > >Here's the results a friend send me: > >G4 400Mhz: > >perft 5: 416kNPS >search 10s: nodes = 1474004 NPS = 146k > >G5 1.8Ghz: > >perft 5: 1900kNPS >search 10s: nodes = 6281000 NPS = 627k > >Hm, seem to be a pretty linear scale up 1800/400 = 4.5 times Thanks for the data. If my own program behaves similarly, I can expect the 2 GHz G5 to be about as fast as a 2.4 GHz Pentium IV, and about 30% slower than an Athlon XP 2.4 GHz. This is of course not very impressive, but it is certainly a lot better than I feared. Tord > >regards >Andy
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.