Author: Robert Hyatt
Date: 09:35:35 12/01/02
Go up one level in this thread
On December 01, 2002 at 09:32:40, Yen Art Tham wrote: >On November 30, 2002 at 23:40:30, Robert Hyatt wrote: > >>On November 30, 2002 at 20:41:28, Yen Art Tham wrote: >> >>>On November 29, 2002 at 22:56:21, Robert Hyatt wrote: >>> >>>>On November 29, 2002 at 17:27:30, David Rasmussen wrote: >>>> >>>>>On November 29, 2002 at 11:20:04, Robert Hyatt wrote: >>>>> >>>>>> >>>>>>You might have a better chipset or motherboard. I don't know. I do know that >>>>>>everyone that has tested (perhaps except yourself) dual AMDs have reported >>>>>>disappointing results, which makes me believe there is a memory bottleneck. >>>>>> >>>>> >>>>>This is simply not true, if you're talking in general. >>>> >>>> >>>>If you have read what I have written on this subject, it has _always_ been >>>>in the context of running Crafty. Nothing more. nothing less. And running >>>>crafty, AMD has problems in their dual configurations. Eugene and others have >>>>posted numbers showing this clearly... >>> >>> >>>Could this be the other way around (ie crafty has problems running on dual AMD) >>>Others have posted here that DF7 gets 1.8x on dual AMD. >>> >> >>And some have reported no speedup for DF as well, but I don't worry since I >>don't have anything to do with that program. But as far as "does crafty have >>trouble with AMD?" or "does AMD have trouble with Crafty?" the two questions >>are identical... > > >I said DF7 not DF. They are two different engines. >The question may sound identical but the conclusion is faulty. > A lot of posters here have been using the 1.4x number to suggest that AMD duals >are inferior to Intel's, which is pure baloney. It isn't just "a lot of folks here". AMD SMP performance has been under the magnifying glass for quite a while on several benchmarking sites, for the same reason. There _is_ a problem with AMD/SMP stuff. Perhaps not with every AMD chipset, as I don't follow AMD closely enough to know what they are doing. But some very recent benchmarks show the PIV clobbering the AMD in some common benchmarks (non-floatingpoint stuff mainly, which fits CC perfectly). > > >> >> >>> >>>> >>>> >>>> >>>>> The general word on >>>>>usenet about dual P4s vs. dual Athlons is that besides being cheaper, they're >>>>>also better. Many people who use it "seriously" as in heavy 3D work, rendering, >>>>>photoshop etc. report in favor of AMD. P4's might be better for Crafty, I don't >>>>>know. It would only be natural, since you have tested and developed it >>>>>extensively on P4s and not on Athlons. But P4s are certainly not better in >>>>>general. >>>>> >>>>>/David
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.