Author: Anthony Cozzie
Date: 17:52:43 02/19/03
Go up one level in this thread
On February 19, 2003 at 01:14:13, Matt Taylor wrote: >On February 19, 2003 at 01:11:17, Aaron Gordon wrote: > >>On February 19, 2003 at 01:06:24, Matt Taylor wrote: >> >>>On February 19, 2003 at 00:52:09, enrico carrisco wrote: >>> >>>>On February 19, 2003 at 00:19:41, Robert Hyatt wrote: >>>> >>>>>On February 19, 2003 at 00:11:08, Aaron Gordon wrote: >>>>> >>>>>>I just downloaded Crafty 16.19 and ran a bench for you guys. No single cpu Intel >>>>>>box could ever touch this without sub-zero cooling. Just plain not going to >>>>>>happen. >>>>>> >>>>>>Crafty v16.19 >>>>>> >>>>>>White(1): bench >>>>>>Running benchmark. . . >>>>>>...... >>>>>>Total nodes: 67136136 >>>>>>Raw nodes per second: 1766740 >>>>>>Total elapsed time: 38 >>>>>>SMP time-to-ply measurement: 16.842105 >>>>> >>>>> >>>>>This is not a great test since that is a very old version. I'm not sure how >>>>>1.7M compares to version 19.3 in nps... >>>>> >>>>>However, while on the question, what is an XP 2.44ghz machine, since I am not >>>>>an AMD expert. Overclocked? If so, I consider that a worthless number, because >>>>>of obvious reasons... >>>> >>>> >>>>If done properly and tested for reliability -- what reasons do you speak of? >>>>Most CPUs are purposely locked from higher than marked performance from the >>>>manufacturer for marketing and other reasons -- both Intel and AMD. This, in no >>>>way, means the CPU is incapable of such performance. >>> >>>Propigation delays do. Intel and AMD release chips at a given speed for a >>>reason. Yes, much of it is about money. It is very profittable to allow >>>consumers to upgrade through every iteration of a chip. Not always. Intel >>>delayed the 1.13 GHz Pentium 3 for a while. I've heard that they could not mass >>>produce them reliably at the time. >>> >>>>In the case of AMD, chips with the same stepping are identical no matter what >>>>they're marked. So if a 1500+ AthlonXP has an AIUHB 0301 core and an AthlonXP >>>>2800+ has an AIUHB 0301 then they'll be able to run identical speeds. >>>>(Obviously there are slight variations in peak performance, if you're going for >>>>higher than XP3000+ level.) >>>> >>>>Are you suggesting that "unlocking" performance that is already included in the >>>>core simply because the marking on the top of the cpu says otherwise makes such >>>>results worthless? >>> >>>No, he is suggesting that comparing unguaranteed performance is worthless and >>>silly. You might stick a peltier on your chip, tweak the voltage, and manage to >>>run 2.8 GHz or something similarly fast. That doesn't mean I can. That doesn't >>>mean anyone else can. >>> >>>>Wouldn't that be the same as saying stronger results I may find with Crafty if I >>>>modified the settings are completely worthless if you did not include the >>>>settings in your official release? >>> >>>Modifying Crafty compile settings doesn't cause it to crash all of a sudden. >>>Furthermore, anyone can apply those same settings and get the same results. >>> >>>-Matt >> >>Thats what he was saying about the chip. Anyone can slap an air-cooler on an >>AIUHB chip and get at least 2.3ghz (up to 2.6ghz). If you can compile crafty, >>you can surely push a few keys to raise your bus and voltage. Even my fiance >>overclocks her computer (and did it by herself). I'm not magical, you can get >>the same hardware and run the same settings I do. :) > >AMD doesn't guarantee 2.3 GHz out of any of their chips. As I recall, you had a >particularly bad one recently... > >Anyway, if AMD doesn't define the limit of stability, who does? Some P4s >overclock to 4 GHz now, don't they? > >-Matt no one knows what the limits are. The problem has to do with false paths in the circuit. however, I *do* know that no .13u P4 will never make it to 11GHZ. Intel has published the setup time for the pipelining flipflops. anthony
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.