Author: Matt Frank
Date: 11:33:48 01/29/99
Go up one level in this thread
KarinsDad: Your points are well taken. However, I would suggest that comparing Hiracs 7 with Hiracs 6 on identical hardware, hashtables, etc. would produce a clear victory for 7. A good handle on this prospect can be derived from the published beta test of 7 on Hiracs' web site. Other comparisons are given on identically configured matches against 7. I do think that when you add in a faster processor (e.g., 400 MHZ instead of 200 MHZ) using the same software, a 65-70 elo jump should ensue. Doubling the hashtables would add 7 elo points to any, otherwise, similarly configured machine/software comparison. These assumptions are elaborated on at the SSDF web site FAQ. Added together, that 75 point elo jump plus 2576 gives you an elo on the SSDF, of approx. 2650 for Hiarcs. Therefore, I am saying that the Hiarcs 7 is a clear improvement over Hiarcs 6 on a similarly configured machine (between 45-75 elo points), and that the newer, faster hardware configs, will produce an even larger surge in SSDF ratings when the more powerful machines are tested. These assumptions are based on my own research using published data. I plan to buy Hiarcs 7 for my G3 233 Power Mac. My estimate of its power is 2610 based on extensive analyses, using the published data that I have noted. I am interested in testing my machine in match conditions with US players rated FIDE elo 2500-2640). I believe that a regulation tournament 40/2hrs. and cash for winning should provide a strong test of this software/machine config. References for my include Shep's computer web site which has nmatch results and engine tests on chess problems (e.g., LT11) and Enrique Irazoqui's match among others. Matt Frank
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.