It's not even that, at the server level it's getting very difficult. For people that demand them, SuperMicro / Tyan / Others. Hell, right now, the NforcePro 2200 chipset is solid.
There is just no advantage to Xeon, none. Need 64bit/100Mhz PCI slots? Opteron can give them to you (as can Xeon). Need Dual Core? Both processors can do it. Need a rack mount that isn't so blazing hot it sounds like a freight train with the fans running? Yeah, Xeon doesn't get that one. Even Intel knows it. The Xeon is way behind in updates, and it's getting farther behind. Right now, Bensley, the first dual core Xeons, were supposed to be ready to go with boards by the end of this quarter. Where are they? Nowhere. And while they should be good performers, they are still power magnets. Anand nailed something that I've been waiting for someone to say: http://www.anandtech.com/IT/showdoc.aspx?i=2644&p=6 ---------------- At 40-60% load, it costs $42 a month to run a Bensley system while an Opteron system would cost $25. Factor that over a year, and a Bensley system would cost $504 to run and an Opteron would cost $300. Now, if that doesn't pique your interest, let's say that you have 40 servers at a datacenter, with the same power characteristics. Over a year, it would cost you $8,160 more to run the Bensley system. ---------------- Now, for a single workstation, maybe that extra $200 a year isn't a big deal. But he's dead on, if your running a set of servers, which is what normally happens if you're big enough to be buying up multiprocessor racks, then it adds up. I'll give an example: a business client of mine stores terabytes of graphic data. Now, each client that they host has their own server they keep which stores their specific data, so they have remote access to it. (Architecture). Now, in their datacenter, they keep 85 independent servers. Some Linux, some Windows, whatever. What justification is there at all for them switching to these high-cost upkeep Xeons? None. And these "new" Xeons are better at power consumption (way better) then the old Xeons, which often could run as high as 115W. Until Bensley shows up, here's what you've got: http://news.com.com/2061-10791_3-5940478.html ------------------------ The single-core "Irwindale" Xeon for dual-processor servers has a TDP of 110 watts and a maximum of 120 watts, according to Intel data sheets. But Paxville for dual-processor servers runs at 135 watts and 150 watts for the comparable figures. In Xeon MP models for servers with four or more processors, the gap is significantly larger. The single-core "Cranford" has a TDP of 110 watts and a maximum power of 120 watts, but the dual-core Paxville Xeon figures increase to 165 and 173 watts, respectively. ------------------------ It's good that Intel is getting "down" to 95W, but even then that's the weighted average, not high end, as test after test shows they can't hold to that. But 173W? Hello! You want to talk about making life hard on people who construct servers to make it work out without making freight trains? I think Intel has a chance to really break through next year.. when they start to migrate the Centrino type technology - their good stuff - to the desktop. We should start seeing it in real by Q2. The question is how well they pull that off, because right now, it's not a matter of "favoring" AMD or anyone else, it's just that Intel screwed up the one thing they were always good at: executing delivering a product. CW ---------------------------------------------------------------------------- ---------- FIGHT BACK AGAINST SPAM! Download Spam Inspector, the Award Winning Anti-Spam Filter http://mail.giantcompany.com > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:hardware- > [EMAIL PROTECTED] On Behalf Of Hayes Elkins > Sent: Wednesday, December 28, 2005 8:19 PM > To: [email protected] > Subject: RE: [H] 1400-1500 Workstation > > >From: Winterlight <[EMAIL PROTECTED]> > >Reply-To: The Hardware List <[email protected]> > >To: The Hardware List <[email protected]> > >Subject: RE: [H] 1400-1500 Workstation > >Date: Wed, 28 Dec 2005 14:24:34 -0800 > > > >At 02:09 PM 12/28/2005, you wrote: > >>There's no way around getting a Athlon64 X2, any other choice is pretty > >>much a stupid one (especially for something designated as a > >>"workstation"). > > > >Gee, I don't know, I am using a dual Xeon 3.06 HT with 4GB RAM, and > raptor > >drives, as my video "workstation" and it seems to do OK. > > Sorry, but there is not a single reason to purchase a Xeon today over an > Athlon64/Opteron other than a preference for a sticker labeled "intel". > Are > Xeon's faster? No. Are they more power efficient? No. Are supporting > boards > and memory less expensive? No. > > Do Xeons suck? Of course not. You can build very fast stable systems > around > them. But - is there a reason to buy them in light of the Athlon64? No.
