The future of Prescott: when Moore gives you lemons...

muddocktor

Here's a damn good read on the future of Presshott at Ars Technica. It explains a lot of what's happening with it and why the power consumption and TDP on it and everyone else's newest designs keeps going up even with process shrinkage.

If you read my Moore's Law article, then you know that there are two main ways to take advantage of decreasing transistor sizes. One way is to shrink the overall size of the CPU core and raise the clock speed; the alternative is to pack more functionality onto the same sized CPU die without raising the clock speed. In reality most folks do a mix of these two, but with more of one and a less of the other. Intel's Pentium 4 architecture (a.k.a. Netburst) is predicated on the assumption that the former approach — shrink the core size and raise the clock speed — will translate into both better performance and better sales, because it's easier to sell MHz than it is to sell added functionality. The problem with this approach is twofold and can be summed up with two terms: transistor leakage and power density.

They also hypothesize why Intel doesn't have any plans for adding hyperthreading to Pentium M any time soon.

So to sum up, Intel's reasoning appears to be that Prescott stands to gain more from hyperthreading than the Pentium M, so the latter won't be getting the hyperthreading treatment anytime soon because it just doesn't need it as bad.

Link to the Article right here.

Thrax

Because we got obsessed with MHz as a marketing number and made Prescott's pipeline so ridiculously long, Prescott benefits much more from a latency-hiding technique like hyperthreading than a saner design like the Pentium M.

I like that quote more.

Omega65

Ars-Technica muses on why Intel is producing a Dual Core processor based on the high wattage Prescott core instead on power efficient Dothan Core.

My first reaction upon reading the dual-Prescott rumor was incredulity. With its simpler architecture and higher performance per watt, Dothan, the desktop version of the Pentium M, seemed like a much more likely candidate for dual-core treatment than the large, hot, complex, power-hungry Pentium 4.

In the Geek.com interview, the Intel representative offered that the Pentium M's shorter pipeline is supposedly "not suited" for hyperthreading. Since there was no elaboration on that point in the Geek.com article, I'll do that here.

Source: Ars-Technica

Omega65

Posted to Front Page and threads merged....

Gobbles

my laptop has a pentium m, I wish it had hyperthreading...

Its not bad.. if only the rest of the parts were as fast...

I managed to convert a dvd to svcd in under 3 hours... I assume thats pretty decent. It was a 1 hour 29 min dvd...

Up with dothan down with everything else


Gobbles

edcentric

Omega, I think that we all know why there won't be a dual core Dothan, P4 would be dead and useless at that point.

Thrax

That's what Intel is banking on. They're ditching the P4 quite soon, which is why LGA775 was introduced. It <i>would</i> have been the Tejas socket, but it can just as easily be the dothan-based successor to the p4.

mmonnin

You wouldnt gain as much with Hyperthreading on the P-M compared to what the P4 gained with Hyperthreading. The P-M is so much more efficient that the pipe is so much more fuller compared to the P4.