Author: Robert Hyatt
Date: 14:31:35 02/22/03
Go up one level in this thread
On February 22, 2003 at 01:49:45, Jeremiah Penery wrote: >On February 22, 2003 at 00:40:39, Robert Hyatt wrote: > >>On February 21, 2003 at 23:54:14, Jeremiah Penery wrote: >> >>>On February 21, 2003 at 18:20:44, Robert Hyatt wrote: >>> >>>>On February 21, 2003 at 10:23:19, Jeremiah Penery wrote: >>>> >>>>>On February 21, 2003 at 09:47:27, Robert Hyatt wrote: >>>>> >>>>>>That logic is circular. They can make faster xeons but they can't make faster >>>>>>Itaniums??? >>>>> >>>>>Besides the fact that P4 is orders of magnitude less complex than Itanium, there >>>>>are several reasons why P4 can ramp up in clockspeed much faster than Itanium. >>>>>If P4 suddenly makes a huge jump, Itanium may be able to jump also, but nowhere >>>>>near as much. It's not as if Itanium machines are selling like hotcakes as it >>>>>is, so I'm sure Intel wouldn't be too happy if the Xeon starts eating into that >>>>>segment of the market more than it already has. >>>> >>>> >>>>I would think they would be more than interested in clocking beyond the PPC4, >>>>the >>>>alphas, the HP PAs, the MIPS 64 bitters, and so forth... >>> >>>Intel could probably have released 1.5GHz Itanium a while back, had it wanted >>>to, by using the 0.13um Copper process instead of the older 0.18um Aluminum one. >>> Instead, they've waited until Itanium3 (Madison/Deerfield) to do that. Since >>>Madison is practically just a shrunken version of McKinley, I don't see why this >>>would have been a problem. >>> >>>>They don't sell as many, but the profit margin is _way_ larger up there on that >>>>end of the server spectrum. >>> >>>Which is why they don't want Xeons eating into that segment. >> >>Xeons don't _touch_ that segment. IE I'd bet they would love to see Cray >>selling T3 machines with Itaniums rather than alphas. Or SGI selling challenges >>with Itaniums rather than MIPS. Or IBM selling SP2's with Itaniums rather than >>PPC4's. Etc... >> >>Xeons are nowhere to be found in that market segment... > >Are you at all reading what I'm writing? The supposition is that Xeon, if >sufficiently clocked relative to Itanium, would eat into Itanium sales. Of >course Xeons aren't in that market segment today, but that is not the point. If you clocked them 2x faster they wouldn't eat into that segment. Big server world is about something different, not a simple PC with one lousy bus that has no bandwidth to speak of. > >If Xeon suddenly clocked up to 5GHz tomorrow, and all other chips stayed the >same, it would certainly begin eating into that market segment. Assuming >something near linear scaling, a 5GHz P4 would have SPECint and SPECfp near >1800. That's _twice_ as fast in SPECint as any currently released "server" >processor (POWER/Alpha/Itanium), and almost 25% faster in SPECfp than _any_ >processor. SPEC is not that important for servers. Ditto for mainframes. I/O is the name of the game, and multiple channels/busses is the issue. The PC is still a PC, whether it is 500mhz or 50ghz.
This page took 0.01 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.