The Future of Pentium 4

[Asus]

Prescott is the only one left to continue as Intel releases it's new socket 775 and platform this weekend. Northwood stops at 3.4GHz with Socket 478. How will the new platform fair as they release 3.6GHz P4 Prescotts and beyond with new technology like DDR2?

Find out in Anandtech's review.

 

[jonbob]

Personally, I'm interested in what will become of P4M-based desktop chips.

 

[Fletch]

Argh - too much new technology! Im gonna be bankrupt keeping up with it all Interesting link Asus.

 

[Asus]

Personally, I'm interested in what will become of P4M-based desktop chips.

So far that is only rumor.
That's because people assume since they ditched Tejas that the natural adoption would be away from MHz/GHz but going to Intructions per Operation with the Pentium M redesigned as a desktop CPU. They would have to redesign it a bit since it really doesn't perform as well as desktop CPUs mainly because of the memory bus. It has lower FPU than the Pentium 4 which is something they may worry about, as far as I know.

Then again didn't they just anounce dual core prescotts is in their roadmap?
AMD actually taped-out their dual core CPU, ready for the Fabs.

btw, P4M is a Pentium 4 as a moble CPU.
Pentium-M is a whole different beast.

 

[lePobz]

Intel Schmintel. I'm much more interested in AMD Dual-Cores. Imagine a dual-cpu dual-core system. That's like 2ghz per core, x4. 8ghz, weeeeeee.

 

[Asus]

Should check out what Tommy Boy has to say about all of this. Link

 

[Asus]

And HardOCP's review. Link

I wonder if AMD will drop some prices now that Intel has launched their product.
Maybe they will wait until they actually see Intel product on the shelves, that will be awhile.

 

[crabcakes66]

Intel Schmintel. I'm much more interested in AMD Dual-Cores. Imagine a dual-cpu dual-core system. That's like 2ghz per core, x4. 8ghz, weeeeeee.

to bad that doesnt work in the real world....

 

[Asus]

Personally, I'm interested in what will become of P4M-based desktop chips.

btw you might want to read this 3 page article.
Intel may try to salvage Prescott and turn it into a dual core chip rather than bring in the P-M, which would be a real shame.

So let's say that you're Intel, and you've got this microprocessor architecture that's the backbone of your central product line, with millions invested into developing it and improving it. And let's say that the fundamental premise on which that microarchitecture was built craps out a few years too soon. What do you do? In Prescott's case, it looks like the answer is to try your best to adapt the old design to a new set of circumstances, rather than throwing it out the window. This means cranking way back on the clock speed increases, and taking advantage of Moore's Curves* by adding functionality to the die instead. This added functionality comes in the form of a whole other core, rather than a few new execution units and some cache.

So the dual-core Prescott design is Intel's way of making lemonade out of lemons. Because Intel can no longer rely on clock speed increases to bring performance gains, they plan to take their flagship processor design and head off down the same multicore path as IBM, AMD, Sun, and the rest of the industry. And of course, the rest of the industry is going multicore because of the power-related problems outlined above. The whole industry is suffering from the same headache, and now everyone has settled on the same cure; Intel was so addicted to MHz that they were a bit late to come around, but come around they have.

The perceptive reader will have picked up on the larger point that I'm making: a dual-core Prescott is a hack, made necessary by the fact that the single-core Prescott is going nowhere fast. So Intel is hoping that a 64-bit dual-core Prescott with hyperthreading will make for an attractive workstation offering, especially since they'll still have a clock speed advantage over whatever AMD has out at the time. Whether they'll have a performance advantage is hard to tell, but if this chip does see the light of day and if MHz still matters at all, it might help Intel squeeze more life out of what is an otherwise moribund architecture.

I'm of the opinion that a dual-core Pentium M derivative would make much more sense, and I'll bet that such a design is percolating somewhere deep within Intel. But it'll only come out if the dual-core Prescott fails to catch on.