Author: Vincent Diepeveen
Date: 09:19:35 11/15/02
Go up one level in this thread
On November 15, 2002 at 11:54:48, Jorge Pichard wrote: the problem of the xeon 2.8 Ghz is that the P4 2.8Ghz is not even getting delivered by the shops here, despite that it was ordered for example by Jan Louwman. Not to mention the Xeon. Could be months before it is here in Xeon form. Right now fastest Xeon for example www.alternate.nl delivers is: Xeon 2.0 Ghz whereas the 2.8 Ghz P4 is on the list as 'deliverable' and the 3.0 Ghz P4 is on the list as 'coming soon'. But that 'deliverable' of the 2.8Ghz P4 we must take with a graint of salt. It's ordered before dutch open computerchess champs even by Jan Louwman (weeks before it). So that was like end of september. It's november now. As a replacement he got a bunch of 2.53Ghz P4s delivered until they could deliver 2.8 Ghz and they didn't yet. So i expect those 2.8Ghz to arrive soon. But that's P4s. That's not even near what Xeons do. www.comptechworld.nl has the Xeon 2.4Ghz on their list but doesn't want to do statements on when they want to deliver them, especially considering they get delivered for only 367 euro, that'll take a while i bet. Knowing how fast these cpu's have gone down in price, it will be no surprise to you that *no one* here has those CPUs in stock. They need to get from abroad always. Same story is also true for latest AMD cpu's, but their jumps are smaller. That P4 is clocked to 3.0 Ghz already, quite a big difference from the 2.53Ghz which can get delivered now. The Xeon is officially at 2.8Ghz now but all what you could get within a few weeks is a 2.4Ghz. And that is perhaps *not* the northwood yet. I do not know. It's the more recent core that's a lot faster than the old P4 core. Johan de Gelas is a very good tester. His results, without posting it, he repeats times and again in a very objective way. I was amazed by his accurate testing, this was around july 2001. The article took a month to place or so. In july 2001, the 1.2Ghz MP K7s from which i still have 2 processors here, they blew away *every* PC on the planet. The fastest machine before that was a quad Xeon 700, but that K7 core is *so much better*, at least 20% than the P3-Xeon, that the compare was not very accurate even. The later introduced 933Mhz Xeons were called back to the factory because of bugs in them. So *nothing* could beat it at all. The later improved P4s (i guess that's the later northwoods) are improved at more points than just a bigger L2 cache, as seemingly initially was reported. A major difference obviously is that they have DDR ram nowadays. RDRAM at least for computerchess is a big failure IMHO. Yet the extra speed the northwood delivers i cannot explain just by DDR ram. I bet intel doesn't want to speak about it much it has to be something small that's internally changed which has a big impact (and probably was a small change but very difficult to manage). >On November 15, 2002 at 10:46:18, Gian-Carlo Pascutto wrote: > >>On November 15, 2002 at 10:38:10, Jorge Pichard wrote: >> >>>http://www.aceshardware.com/Spades/read.php?article_id=45000204 >> >>Not quite - the MP 1200 is old! Vincent has a pair of MP1600 >>now, and if I'm not mistaken, MP 2100 or MP 2200 are already >>available. >> >>-- >>GCP > >I forgot to read the day that vincent made the report. Three weeks ago I posted >this, to compare the MP 2100 vs Intel Xeon. I believe that intel took the lead >in the dual race, by introducing the Xeon 2.8 Ghz. > >http://www.anandtech.com/IT/showdoc.html?i=1641&p=6
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.