Author: Ulrich Tuerke
Date: 12:04:05 12/21/00
Go up one level in this thread
On December 21, 2000 at 13:45:38, Ed Panek wrote: > Hello all, > > For fun i have been playing a match between two computers using rs232 >both using Hiarcs7.32 > > One computer is an PIII 800 with 128 MBRAM > the other is an old 486 DX-2 100 MHZ with 32 mb EDO > > the PIII generates about 70-100k Nodes per second > the 486 does about 1-3k nodes per second > > The PIII is up on games 8.5-.5 but the interesting thing is that >the 486 actually holds its own until about move 30-40. > > I also played a few manual games with genius 4.0 and it lasted even longer Genius really does extremely well on 486 hardware. Recently, I'd made some changes to Comet and I just wanted to see if everything's still working ok. So I made a short test-match of 4 blitz games (5min), Comet running on a K6/233 and Genius-3 on a 486/50. I couldn't hardly believe to see that Genius did win this. Later, it took the K6/450 to give sufficient advantage to win against Genius/486. Though my program may be weak in blitz games, I don't think that any other opponent would be able to win a blitz match with such hardware disadvantage. Yes, the Lang programs had been something very special. Greetings, Uli > >I wonder if 7-8 years from now we can expect the same difference ie. >the computers running H732 will generate about 60X more nodes or will search >4200-6000knodes per second. > > >hmmm > >Ed
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.