Author: Jonathan Parle
Date: 22:28:56 01/18/02
Go up one level in this thread
On January 18, 2002 at 10:00:55, K. Burcham wrote: > >http://www.xbitlabs.com/news/story.html?id=1011297004 > >notice, todays fastest amd is at 730 on this chart. >if you are going to have the fastest six months from now, get ready to >sell everything you now have, it seems these increases in performance >will be significant. > >i used to vacuum my fans every once in awhile. now i sell my systems so quick >that the dust never settles on my cooling fans. >kburcham The question is, for a given program, how much rating increase would one get from buying ever faster hardware???. There are diminishing returns. Back in the 6502 days, you could reliably state that a doubling of processor speed was worth 70 - 100 ELO points. Looking at the latest SSDF rating list, I see that Deep Fritz on an Athlon 1200 with 256MB = 2711. Deep Fritz on a mere K6-2 450 (not only somewhere around a 1/3 speed but totally different and much slower architecture) with 128MB = 2658. Only 53 points difference. A similar story for Gambit Tiger. And I would hazard a guess that it is the extra RAM which is responsible for a significant component of the strength increase reported. If you are mainly using your PC grunt for chess programs, I can't see the point in upgrading more often than every few years, as is the case for PC's used for business applications. Is a handful of ELO points (which is all you are going to get) really worth thousands of dollars? I'd probably just be buying RAM for existing machines and beefing up the hash tables.
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.