Author: Robert Hyatt
Date: 15:24:08 09/03/02
Go up one level in this thread
On September 03, 2002 at 17:41:21, George Sobala wrote: >Seems a reasonable explanation. I must say that the speedup table does look >perfectly genuine (i.e. no tell-tale fingerprints of having been "made up".) > As I said, that is _the_ part of the data that we are certain was totally correct. Trying to extrapolate nodes searched was, to me, totally pointless. But referees are referees. And I gave 'em the data, which I might add, really was very accurate. But it was based on the time each cpu was busy, which means 100% correlation to speedups and times, since they are all related. >On September 03, 2002 at 17:30:35, Robert Hyatt wrote: > >snip > >> >>So, perhaps the data has some questionable aspects to it. The only part that >>I am _certain_ is "raw data" is the individual speedup values, because that is >>what we were looking at specifically. I had not remembered the node count >>problem until this email came in and then I remembered a case where Vincent >>was trying to prove something about crafty and got node counts suggesting that >>it should have gotten a > 2.0 speedup. I had pointed out that the way I do >>nodes, it is impossible to produce them anywhere except when all processors are >>idle, if you want an accurate number. I _should_ have remembered that we had >>the same problem back then. I am therefore afraid that the times might have >>been computed in the same way since it would have been quite natural to do >>so... >> > >snip > >>But the bottom line is "trust the speedup numbers explicitly". And if you >>trust them, the others can be directly derived from them. >>Bob
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.