Author: Robert Hyatt
Date: 20:49:58 02/19/03
Go up one level in this thread
On February 19, 2003 at 18:41:47, Charles Worthington wrote: >Bob from my observation the 1310 kNs with df7 would almost have to be on dual >threads. The fastest I have seen with 3.06 was 1195 kNs (but I am comparing >apples and oranges because that was on fritz 8). Anyone running df7 on the >server with the SMT-enabled 3.06 would be placed on both threads by default so I >have not observed single thread performance on df7. I do know that the fritz 8 >engine is a bit slower to due the knowledge but trying to >compute_That_difference would involve too many variables to do it accurately. My >estimate based on 1195 kNs x 1.85 plus the conservative 10% SMT figure would put >me roughly in the ballpark of 2432 kNs with df7. Of course once the tweaking is >done to make the engine SMT friendly this figure will rise considerably. OK. I was interpreting that as a single cpu, no SMT. that means it is going to be very tough to break even 2600K, based on raw math... but it will still be very strong... Of course that 3.06 might be a 400mhz or 533mhz FSB, if it is the slower one, then you will get a bit more bang.
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.