Author: Robert Hyatt
Date: 08:10:22 09/05/02
Go up one level in this thread
On September 05, 2002 at 11:02:28, Vincent Diepeveen wrote: >On September 05, 2002 at 10:11:32, Robert Hyatt wrote: > >Try again i'd say. Just for a thought experiment also try >without asymmetry. Perhaps it matters when the tree of >what you search matters for speed too. Not for NPS. Do I need to run that to show you? eval affects the shape of the tree. It doesn't affect the size of the tree, for a given time limit. > >If there is little overlap when running dual, the number >of nodes a second obviously drops with tens of percents too. Not with my program. Can't and won't say about yours. But NPS is a pretty constant thing the way I do things. As I have shown with data I have sent you and posted here, of course... > >Could be that 'analysis' to 13 ply shows a different number >as well. Certainly. It could show a faster SMP speedup. Or a slower SMP speedup. Because it changes the shape of the tree. It won't change the NPS much (it might go up a bit as some parts of the eval are turned off in analysis mode)... > >>On September 05, 2002 at 10:08:59, Vincent Diepeveen wrote: >> >>>On September 05, 2002 at 02:20:22, Slater Wold wrote: >>> >>>>On September 04, 2002 at 22:26:13, Robert Hyatt wrote: >>>> >>>>>On September 04, 2002 at 22:03:24, martin fierz wrote: >>>>> >>>>>>...or in the short version, 1.69 more nps for the 2-processor box. >>>>> >>>>> >>>>Totals so far: >>>> >>>>dual xeon 1700mhz 1.84X raw NPS >>>>quad xeon 700mhz 1.90X raw NPS (using only two processors) >>>>dual AMD 1730mhz 1.67X raw NPS >>>>dual cele 433mhz 1.86X raw NPS >>>>dual P3 450mhz 1.92X raw NPS >>> >>>different things get compared here. outputs without cleaning >>>hashtables are in this. That sucks ass. >> >> >>I didn't see any of those. I specifically asked them to make two different >>runs, which everyone I looked at did. So, as usual, I don't know what you >>are talking about. But you sure do jump to conclusions without looking at >>each post here where you would have noticed that they ran crafty two times, >>as can be seen for the ones I have checked...
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.