Author: Robert Hyatt
Date: 20:47:26 03/08/03
Go up one level in this thread
On March 08, 2003 at 02:58:59, Matt Taylor wrote: >On March 06, 2003 at 11:18:22, Robert Hyatt wrote: > >>On March 06, 2003 at 00:19:54, Keith Evans wrote: >> >>>On March 05, 2003 at 22:41:02, Robert Hyatt wrote: >>> >>>>On March 05, 2003 at 20:10:56, Keith Evans wrote: >>>> >>>>>On March 05, 2003 at 15:09:38, Robert Hyatt wrote: >>>>> >>>>>>On March 05, 2003 at 14:46:18, Keith Evans wrote: >>>>>> >>>>>>>On March 05, 2003 at 11:34:39, Robert Hyatt wrote: >>>>>>> >>>>>>>>On March 05, 2003 at 10:21:32, Vincent Diepeveen wrote: >>>>>>>> >>>>>>>>Your data is simply wrong. The xeon core has _not_ changed from 2.8ghz to 3.06 >>>>>>>>ghz, >>>>>>>>and I have no idea why you want to supply your "disinformation" that it has. >>>>>>>> >>>>>>>>We have four of these on the way (dual 3.06 dell 650s) for faculty. They have >>>>>>>>shipped >>>>>>>>(2/28) so they should be here any time. I'll run the tests and post the results >>>>>>>>to further >>>>>>>>debunk this "myth" that 3.06's are different... >>>>>>>> >>>>>>>> >>>>>>> >>>>>>>What is the difference between a 3.06 GHz P4 and a 3.06 GHz Xeon? >>>>>>> >>>>>>>Thanks, >>>>>>>Keith >>>>>> >>>>>> >>>>>>Not a whole lot, other than the xeons have the necessary hardware for multiple >>>>>>processor >>>>>>configurations. IE the ability for each to invalidate pieces of the cache based >>>>>>on activity on >>>>>>the other processor. >>>>> >>>>>At Dell you can buy dual processor P4 systems. How are they handling the >>>>>multiprocessor configurations compared to a Xeon? Is there a performance >>>>>hit? Are they really Xeons and not P4s? >>>> >>>>What machine? We have bought several 650's and they are xeon-based. My >>>>2600 server platform is also xeon-based. >>>> >>>>Tell me what you are looking at and I'll poke around Dell... >>> >>>Now that I'm looking again I can't find them. Which means that it must have been >>>a figment of my imagination, and you're right about the multiple processor >>>stuff. I'm actually quite glad to be wrong about that because otherwise I would >>>be very confused... Sorry about that. >>>> >>>> >>>>> >>>>>>Also there are xeons and xeon-MP versions. The MP >>>>>>versions are >>>>>>supposedly the only ones capable of running in 4-way and up boxes. What the >>>>>>differences are >>>>>>I really have not taken the time to discover... >>>>>> >>>>>>Oh yes. There is one significant difference. Price. :) >>>>> >>>>>Has anyone run crafty on an MP system? Any idea how a 2 GHz MP would compare to >>>>>a 3 GHz Xeon? >>>> >>>> >>>>I have run it on a 4-way 2ghz box. It is faster than my dual 2.8, but then >>>>it is _way_ more expensive as well as the 2ghz xeons (MP certified) only come >>>>in 2mb L2 cache versions and each CPU is selling for $4K or so. Makes the >>>>machine _very_ expensive. >>> >>>I might be interested in such a box for running ASIC simulations so cost >>>wouldn't be an issue. I think that a dual MP box could be had for under $15k at >>>Dell which isn't bad compared to the cost of the software that I would be >>>running on it. The large cache might be a huge win - the 512k Xeon cache >>>definitely helped. >>> >>>Regards, >>>Keith >> >> >>When you say "dual MP" I assume you mean dual xeons but with huge L2? I haven't >>tried >>to check, but I would suspect it might not work. There is something in the back >>of my mind >>that the MP xeons have one extra pin, but that might be a faulty recollection. >>But if it is >>true, it would mean that you couldn't swap a couple of 2MB L2 xeons into a box >>like mine. >> >>I think this is mentioned somewhere on the Intel web site. I ran across it when >>trying to figure >>out what I wanted to buy when I got my dual. > >There are 503-pin Xeons and 504-pin Xeons. I don't know what the difference is, >but I know some have 1 extra pin. > >-Matt That's close to what I remembered. I _think_ the extra pin is the difference between MP and non-MP xeons. It may be that the socket has a filled-in space so that the 504's won't fit, but I'm not certain.
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.