Author: Robert Hyatt
Date: 11:21:12 04/10/03
Go up one level in this thread
On April 10, 2003 at 12:34:13, Keith Evans wrote: >On April 10, 2003 at 11:11:52, Vincent Diepeveen wrote: > >>On April 10, 2003 at 11:07:06, Robert Hyatt wrote: >> >>>On April 10, 2003 at 08:44:18, Vincent Diepeveen wrote: >>> >>>>On April 09, 2003 at 17:58:21, Robert Hyatt wrote: >>>> >>>>as usual you were asleep when replying. i did math for a single cpu. that >>>>extrapolates to more cpu's as well. >>> >>>I did math that extrapolates to _everything_. >>> >>>If I get 1.7X speedup for two cpus, I will get _some_ speedup no matter how slow >>>the >>>second processor is. >>> >>>Which was my point. >> >>with SMT that is not the case. the second cpu in SMT delivers somewhere between >>0% and 20%. >> >>If it is 10% like it is for most programs then: >> >>1.1 speed is what you get out of single P4 with smt. >> >>1.7 / 2 * 1.1 = 0.935 which is slower than single cpu. >> >>Which is my point. > >Why are you dividing 1.7 by 2? > >I thought that Bob already measured this on a single Xeon with SMT and got a >speedup of 1.1 or so. He won't let any real data get in the way of his hand-waving. It is called "inventing math that will prove a point". See my other post in this thread for a _real_ set of data for SMT off vs SMT on. It debunks this stuff quite clearly and anybody can run the test if they have a machine with SMT.
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.