Author: James T. Walker
Date: 04:00:47 12/08/99
Go up one level in this thread
On December 08, 1999 at 04:28:43, Rajendran Ramachandran wrote: >Hi > >I have Pentium III 500MHz 128 Mb computer and my son has tested at various time >control the following engines. > >Hiarcs 7.32 >Fritz 5.32 >Fritz6 >Crafty16.6 > >As far as I have seen nothing comes close to Hiarcs in any sort of time control. > >I have copied the general book for the rest of engines to my hard drive and >enabled the book learning as well. The rest are showing some sort of resistance >now. > >Hiarcs prefers to play on PIII computer with ample RAM. There is no point >testing it in AMD or other processors. Knowing this very well, many try to still >evaluate Hiarcs in such computers, just to put him down! > >raj Hello raj, In my opinion you are wrong in your assessment. First of all, Hiarcs would prefer to play on the fastest computer available as would any chess program. Why do you believe the AMD processors are unfavorable to Hiarcs. When I compare the NPS of the PII-333 to the AMD K6-3-450 the AMD comes out on top. On the PII-333 it get 34Kn/s and on the AMD K6-3-450 it gets 55Kn/s. The Mhz ratio is about 35% faster but the NPS of Hiarcs goes up by 61%. This tells me that Hiarcs loves the AMD. Also after about 2000 games I have found that Hiarcs does not like max ram at all time controls. Your testing on one computer will favor Hiarcs because of it's hash table set up. My testing shows at Blitz and also at G/25 Hiarcs comes in behind Fritz 5.32 and also Fritz 6 so far. But they are all within a few points after more than 7000 total games. I have found Hiarcs is superior to all programs at G/1 or G/2. Again I believe it has something to do with the way it uses hash tables. Jim Walker
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.