Intel's Centrino CPU (Pentium-M): Revolutionizing the Mobile World

Hitting those high clocks

Quite possibly the most interesting part of the way Banias was designed, relates to the processor's clock speeds and what it took to achieve them. As we've already mentioned, the CPU features a longer pipeline than the Pentium III which does help it hit higher clocks, but where does the limit exist?

In the course of designing a processor, you will eventually discover that there are certain speed paths in your CPU that will run either faster or slower than your target clock speed. If you run into paths that run slower than your target clock speed, you're in trouble, since it means that you won't be able to reach the clocks you were hoping to without some sort of a redesign. In most cases, if you find that a path is running faster than your target clock speed (e.g. finding a path capable of running at 2.4GHz on a chip with a 1.6GHz target clock speed) then you're in a very good situation, as it means that there are parts of your chip that have fairly high ceilings. For the Israel design team however, this wasn't the case.

The design team actually went in and slowed down paths that were running above Banias' target clock frequencies, because if a path is able to run faster than it should, it means that you're wasting power. The benefit of this is an even more power efficient microprocessor, but the downside is a microprocessor that has a clear clock frequency wall.

It is unclear what the frequency wall is for the current 0.13-micron Banias, but it is very clear that one exists. We'd hypothesize that the wall is somewhere around 2GHz, but what happens afterwards? The Israel design team couldn't deliver on all of the promises of the most efficient and high performing mobile CPU around without sacrificing frequency headroom, so instead of depending on the architecture to enable higher frequencies, the design team turned to Intel's manufacturing.

The idea is that by the time the core hits its frequency limit, a smaller manufacturing process with faster transistors will be ready for transitioning to. For this year, assuming that Intel's 90nm transition continues as planned, the situation isn't all that risky, but looking forward, there is a big risk with this sort of an approach. From Intel's perspective however, there's not much else that could have been done if you want to have the absolute best mobile CPU possible.

In the end, placing faith in Intel's manufacturing process isn't a bad bet, but it is a risk that Intel's competition would not take simply because they lack the resources to potentially fail.