Slashdot Mirror
Dual-Core Pentium 4 Slated For 2Q 2005
Quantrell writes "Today is the first full day of the International Solid-State Circuits Conference in San Francisco, and Intel has announced that dual-core Pentium 4s are
coming in the second quarter, one in the Extreme Edition line (no surprise there), and also the Smithfield
Pentium 4 800 series, which is the next so-called consumer desktop line. No word on pricing, yet."
Looks like the "Who is Winning the CPU War" line just shifted again.
DAMN YOU OCTODOG! DAMN YOU TO HELL!
Motorola/IBM have had multi-core PPCs demonstrated ever since the very, very first G4 lab units all those years ago.
Yet no one has ever productized a multi-core PPC. (Unless you count the Cell, which you probably shouldn't.)
Why is this?
Is there something about multicore technology which caused IBM/Motorola to decide it was not worth the bother of putting in a box and selling?
Inversely, is there something about multicore technology that makes Intel think we'd actually start caring about the P4 again once it's included?
I'm a little leery of getting excited about having more juice squeezed out of the P4 line, and maybe it's because I'm not entirely clued into the extent of the benefits gained from dual-core P4s. Are they doing this just to gain time before they introduce a new architecture?
I'm looking to build a new AMD-based system this summer, even if they are a little later-to-the-dual-core-table. As far as I can tell, this news doesn't present any substantive reason for me to change that plan.
Can someone more knowledgeable help me get some perspective on this?
.. pa-ra-bo-la, pa-ra-bo-la, 2 pi R, 2 pi R, where's your latus rectum, where's your latus rectum, 2 pi R
The jokes about the heat these puppies will pump out couldn't be more appropriate. An article at Tom's states that the Smithfield core has a thermal design power of 130W making it by far the hottest x86 CPU ever seen.
In contrast, AMD's dual core offering will offer no increase in TDP over their present single core designs.
For all intensive porpoises your a bunch of rediculous loosers
I mean, I assume that unless Windows is rewritten to take advantage of dual cores that you won't see much performance increase. And I assume that just getting OS support won't be enough for applications to really see much improvement either.
SO unless you're a reasonable l33t linux dude/dudette, or I've missed the boat (also possible I'm sure), where do I see the advantage of this system?
---
More craziness here too :-)
As underscored by the following strategy:
and also the Smithfield Pentium 4 800 series, which is the next so-called consumer desktop line.
Doesn't seem that long ago that people at Intel were saying absurd things like, 'consumers will never need 64 bits' or 'consumers will never need dual core'.
Hell, look out the window at El Camino Real and tell me how many of those consumers crusing up and down that road need those 4WD vehicles. Yet consumers buy them in droves. Consumers want, you don't offer, you surrender a market. Seems they've learned not to underestimate what consumers want (which often has little to do with what they need.)
A feeling of having made the same mistake before: Deja Foobar
Their ability to scale frequency is diminishing, lately, the ability to lay down more transistors meant bigger caches, which often has less impact than a second core would. Don't forget that nearly every CPU manufacturer is going dual core, not just Intel and AMD, it is just that they are relatively late into the game now.
I don't think it is too out of line to expect that programmers are going to start considering better multitreaded design. There are limits to what can be done, but for most software, that limit hasn't been approached, in my opinion.
HyperThreading is disabled in the Smithfield dual-core product too, so expect a mere 50% overall performance increase at the same clock speed (2.8GHz, 3GHz, 3.2GHz soon afterwards) for Intel.
I'm not sure what you mean, so I might be wrong. My understanding of what you said is that with HT enabled, you can expect a 50% performance increase. That is unfortunately not true. I'm part of a research group that's using P4s with HT, and the most realistic speedup you get is under 10%.
The problem with the chip is that eventhough two threads can have instructions executed simultaneously, they share everything. For example, if you run two floating point intensive applicaitons at the same time, they're both using the same floating point unit (units? I can't remember), which makes them basically sequential. Squeezing performance out of these is not easy.
I have some experimental numbers in front of me where on a 4 way SMP where each processor is a P4 with HT. The difference with a particular application is that going from one thread to four makes a difference from 16.0s to about 5.3s. But then going from four to eight (using all HT contexts in the machine) makes a difference of 5.3s to 5.1s.
AFAIK, normal Windows 2000 won't let you run more than 2 processors on any one box. But if you activate Hyperthreading on a dual core chip, you effectively have 4 processors--2 of which Windows will refuse to use. (I found this out when I got a dual Xeon Workstation at my job, but the doofi in IT wouldn't give me a Windows Server license because I had "no need".) Could this be the reason why Hyperthreading is turned off on the low-end dual cores? Because the average Joe won't want to spring for a Windows Server license to get the full potential out of his new box?
Great men are almost always bad men--Lord Acton's Corollary