Slashdot Mirror
Intel Begins Shipping 64-bit Prescotts
Rucas writes "With a minimum of fanfare, Intel has begun shipping a version of the Pentium 4 with 64-bit instruction set extensions. The news came to light not via an Intel press release, but rather through the spec sheet for a new server from IBM. In the midst of the new IBM eServer xSeries servers based on the recently released 64-bit Xeon is a blade server powered by the 64-bit Prescott. This marks the first product appearance of the new CPU."
Now it's Intel running behind AMD :)
Scientia est Potentia
64 bit = twice the heat
Intel has been, in reality, behind AMD for at least two years. Now it just gets confirmed.
Bang for the buck means AMD wins hands down.
Professional Politicians are not the solution, they ARE the problem.
Ok, so Win XP 64-bit is in beta. Great! But where's everything else?
At what point are people actually going to start making 64 bit applications? I'm not talking 64 bit linux or anything like that, I'm talking consumer level apps and games.
I see a lot of people upgrading to 64 bit chips, but what good does it do if there's nothing to utilize them? Is it just for bragging rights or what?
I'm a programmer and I have yet to see a need to get a 64 bit chip.
We have secretly replaced these Slashdot mods' sense of humor with a rusty nail. Let's see if they notice!!
You know, I'm starting to feel sorry for Intel; they've had a terribly rough year with no performance gains (while AMD has run past them) and failures with the Prescott (overheating, big time). As we've seen on slashdot, they've recently released their roadmap for the next year and we don't even see speed improvements coming then. Well anyway, about the 64-Prescott. This seems like a very desperate move from Intel in the midst of all their problems -- they've had no official release for this new technology move (this seems sketchy ... this would normally be a humungous deal!) and, moreso, because they're basing this huge move on the Prescott, a chip which has been pushed beyond it's limits. There's reports everywhere about people's fans on heatsinks melting. I guess we'll just have to wait and see what Intel can manage to push out in the coming quarters -- it doesn't seem like them (as has been said by Intel's CEO in an intra-company memo) and, with their past, it only makes sense that they'll emerge from this alive, and may be in the lead again in the near years to come.
I need to know what proper cflags I should use.
Anyway
I'm still unconvinced about 64-bit computing for the present. I think most businesses will wait a long while before making upgrades based on this. One obvious reason is that software is compiled for 32-bit processors, but how much faster is say Gentoo compiled for a 64-bit AMD processor?
A lot of people's arguments defending 64-bit computing is that no software is designed for it. I'm sure I'm completely ignorant on this, but how well does gcc take advantage of it if I were to compile programs to make use of it?
Because the Opteron has an on-die memory controller. That can boost things up to 20% in some cases. It also makes designing motherboards easier because you don't need both a north and southbridge. It makes it harder to upgrade to a new memory technology, but it can be disabled allowing you to do that (I think). If they switched to that buffered "FB-RAM" or whatever (there was an article on the idea a while back on a big hardware site) that would fix that.
But anyway, Intel is stuck in a hard place. Because of the memory controller, their chips perform slower because of the extra latency, so they must ratchet up clockspeeds. The solution? An on-die memory controller. So why don't they do it? They CAN'T.
Intel has been pushing BTX for a variety of reasons (although most people blame Prescott's heat for it). But the way BTX is designed Opteron boards can't be made into a BTX form-factor because the memory is too far away from the CPU (there is too much electrical noise, IIRC). This means that Intel can't switch to an on-die controller without either changing BTX (what I think will happen because of AMD), or finding a way around the noise problem (little faraday cages?).
If you add in things like that the Intel chip only supports 36-bit address (I believe) while the Opteron handles 64-bit addresses (the actual bus is smaller right now, but that could easily be changed) and other performance factors (the top P4EE is outperformed in Doom 3 by a chip that costs more than $800 less, see the Inquirer) and Intel is in hot water.
All of this should be interesting to see what happens. Intel seems to be in trouble (performance wise, at least in the short term).
Comment forecast: Bits of genius surrounded by a sea of mediocrity.
1) "Not invented here". Actually, Intel does have a 64-bit platform, it's called the Itanium. They don't want to detract from their own product line by hyping this. They're marketing it like a way to extend your RAM and a way to get compatibility with those newfangled versions of NT that were once the province of AMD beta testers.
2) The 64-bit instructions are reportedly emulated and are not as fast as the AMD equivalent. Therefore they will make x86_64-specific optimizations seem slow. They'd rather you use it for the 40-bit pointers, but to keep the word sizes 32-bit and not to use those extended registers.
It's a half-hearted effort to get the compatilibity where it matters (OS, database) while exploiting the fact that most of the code is still x86_32 with a sprinkle of performance-critical SSE* and that runs fine on Nocona.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
I think the bigger part behind this is that Intel really didn't want to implement the whole AMD64 command-set. They just wanted to be "compatible enough". This is a defeat for Intel and they're not the kind of company that wants to own up to that. If worse comes to worse, they could ground up a X86-64 chip to compete, but I don't think they want to have to rely on it.
On the otherhand, Intel seems to be marketing these chips as a way to get more memory on your motherboard (aka the way Apple did at first, "Now you can have EIGHT gigs of ram instead of FOUR!!!"). They're hoping this will be enough for most people, considering where these chips will be marketed (low-medium range server).
Besides, Intel's said it plenty of times: Netburst is dying. It was a foolish hack in the first place, but at least it gave them enough time to breed the Pentium M into what it is today.
"Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
Ok, I'm feeding the trolls, apparently, but it's a legitimate question. And the answer, of course, is:
As long as Microsoft leads the desktop operating system market, and as long as people need backwards compatibility to apps compiled for x86."
Remove those two requirements, and you'll see a different architecture become dominant. But, really, is it likely to happen any time soon?
Nobody's complaining that Ars was first with it. The complaint is that this /. article is a cut-and-paste of the first paragraph of the Ars article, with the links changed so they don't go to Ars.
I don't think anyone would be complaining if the submitter had written their own summary of the eetimes article. What's lame is not only taking the entire submitted paragraph from another site without credit, but also removing the link *in* that paragraph to a previous Ars article.
Cheers
-b
If I wanted a sig I would have filled in that stupid box.
None whatsoever. They expression of it, however, is protected by US copyright law, which the US Constitution gives Congress the power to create. So they don't own the information (you could write a paragraph in your own words with the same information) but you can't just copy'n'paste theirs (except when that falls under fair use, which this probably does, so it's a moot point). Whether this is state of affairs is desirable is another matter entirely, though I'm cool with it.
You're joking right? That is barely true. I suspect for integer performance, the Opteron is the strongest out of all of them. Look at the SpecINT scores, it is dominated by x86 at the top end (Xeons & Opterons) and they are way above everything else at all.
For FP, you're half right. The Itanium2 & POWER4/5 are more than a match for the Opteron and will beat it (especially the POWER5). SPARC & MIPS are waaaay slower. No-one uses them for raw MHz performance, more for a large number of CPU systems.
My REAL WORLD tests show the Opteron is 33% quicker MHz for MHz over Sun's UltraSPARC3/3i/3+ processors. That is a problem for them when the Opteron clocks so much faster too!