31 Stages: What's this, Baskin Robbins?

Flip back a couple of years and remember the introduction of the Pentium 4 at 1.4 and 1.5GHz. Intel went from a 10-stage pipeline of the Pentium III to a 20-stage pipeline, an increase of 100%. Initially the Pentium 4 at 1.5GHz had a hard time even outperforming the Pentium III at 1GHz, and in some cases was significantly slower.

Fast forward to today and you wouldn't think twice about picking a Pentium 4 2.4C over a Pentium III 1GHz, but back then the decision was not so clear. Does this sound a lot like our CPU design example from before?

The 0.13-micron Northwood Pentium 4 core looked to have a frequency ceiling of around 3.6 - 3.8GHz without going beyond comfortable yield levels. A 90nm shrink, which is what we thought Prescott was originally going to be, would reduce power consumption and allow for even higher clock speeds - but apparently not high enough for Intel's desires.

Intel took the task of a 90nm shrink and complicated it tremendously by performing significant microarchitectural changes to Prescott - extending the basic integer pipeline to 31 stages. The full pipeline (for an integer instruction, fp instructions go through even more stages) will be even longer than 31 stages as that number does not include all of the initial decoding stages of the pipeline. Intel informed us that we should not assume that the initial decoding stages of Prescott (before the first of 31 stages) are identical to Northwood, the changes to the pipeline have been extensive.

The purpose of significantly lengthening the pipeline: to increase clock speed. A year ago at IDF Intel announced that Prescott would be scalable to the 4 - 5GHz range; apparently this massive lengthening of the pipeline was necessary to meet those targets.

Lengthening the pipeline does bring about significant challenges for Intel, because if all they did was lengthen the pipeline then Prescott would be significantly slower than Northwood on a clock for clock basis. Remember that it wasn't until Intel ramped the clock speed of the Pentium 4 up beyond 2.4GHz that it was finally a viable competitor to the shorter pipelined Athlon XP. This time around, Intel doesn't have the luxury of introducing a CPU that is outperformed by its predecessor - the Pentium 4 name would be tarnished once more if a 3.4GHz Prescott couldn't even outperform a 2.4GHz Northwood.

The next several pages will go through some of the architectural enhancements that Intel had to make in order to bring Prescott's performance up to par with Northwood at its introductory clock speed of 3.2GHz. Without these enhancements that we're about to talk about, Prescott would have spelled the end of the Pentium 4 for good.

One quick note about Intel's decision to extend the Pentium 4 pipeline - it isn't an easy thing to do. We're not saying it's the best decision, but obviously Intel's engineers felt so. Unlike GPUs that are generally designed using Hardware Description Languages (HDLs) using pre-designed logic gates and cells, CPUs like the Pentium 4 and Athlon 64 are largely designed by hand. This sort of hand-tuned design is why a Pentium 4, with far fewer pipeline stages, can run at multiple-GHz while a Radeon 9800 Pro is limited to a few hundred-MHz. It would be impossible to put the amount of design effort making a CPU takes into a GPU and still meet 6 month cycles.

What is the point of all of this? Despite the conspiracy theorist view on the topic, a 31-stage Prescott pipeline was a calculated move by Intel and not a last-minute resort. Whatever their underlying motives for the move, Prescott's design would have had to have been decided on at least 1 - 2 years ago in order to launch today (realistically around 3 years if you're talking about not rushing the design/testing/manufacturing process). The idea of "adding a few more stages" to the Pentium 4 pipeline at the last minute is not possible, simply because it isn't the number of stages that will allow you to reach a higher clock speed - but the fine hand tuning that must go into making sure that your slowest stage is as fast as possible. It's a long and drawn out process and both AMD and Intel are quite good at it, but it still takes a significant amount of time. Designing a CPU is much, much different than designing a GPU. This isn't to say that Intel made the right decision back then, it's just to say that Prescott wasn't a panicked move - it was a calculated one.

We'll let the benchmarks and future scalability decide whether it was a good move, but for now let's look at the mammoth task Intel brought upon themselves: making an already long pipeline even longer, and keeping it full.

Pipelining: 101 Prescott's New Crystal Ball: Branch Predictor Improvements
Comments Locked


View All Comments

  • INTC - Monday, February 2, 2004 - link

    Ummmm yea, kinda reminds me of cooking an egg on an Athlon XP http://www.biggaybear.co.uk/Menu/Aegg/Aeggs.html
  • cliffa3 - Monday, February 2, 2004 - link

    something good to include on the mb compatibility article would be what boards would house the 2.8/533...i'm wondering myself if the E7205 chipset would...i have a p4g8x, and it would be a welcome upgrade with HT and all the other goodies if it oc's well.
  • Stlr22 - Monday, February 2, 2004 - link

    They didn't burn down, but the proc were running hot. Not to mention, these are the FIRST releases in the Prescott line. What's it gonna be like later on?....

    Just think, a P4 based computer that turns your living room into your very own Sauna!!....WHOOO-HOOO!!.....now that's what I call a bargain!

  • INTC - Monday, February 2, 2004 - link

    The message is clear: Anandtech and all of the other review sites didn't burn down so I guess it's not a flame thrower.

    Prescott is not as fast as I had hoped but is definitely not the step backwards as some were rumoring it to be. I think a Prescott 2.8 @ 250 MHz FSB will be really nice to play with until I see what Intel announces at IDF in a few weeks.
  • Icewind - Monday, February 2, 2004 - link

    The message is clear: Im buying an Athlon 64.
  • Vanners - Sunday, February 1, 2004 - link

    Did anyone catch the error in Pipelining: 101?

    if you halve the time for a stage in the pipeline and double the number of stages. Yes this means you can run at 2GHz instead of 1GHz but the reality is you're still taking 5ns to complete the pipe.

    Look at it like a motorbike: You drop down a gear and rev harder; you make more noise but you are still doing the same speed.
    The only reasons to drop down a gear are to break through your gears (i.e. slow down) or to rev significantly higher than the change in gear ratio in order to move faster (with more torque).

    The trouble Intel has is that they drop down a gear then rev 6 months to a year later.
  • kamper - Sunday, February 1, 2004 - link

    Just curious, Anand or Derek: what board did you use to get the 3.72 GHz oc? Obviously it wasn't the intel board used in the benches. I guess we'll hear all about this in the compatibility review though :)

    keep up the good work, that last point about smaller margins at higher clockspeeds (vs. Northwood) was cool. Let's just hope the pattern continues.
  • Stlr22 - Sunday, February 1, 2004 - link

    Seems to me like people either got cought up in some of the hype and expected to much or some people expected to little and that history would repeat itself (Willamette vs Palomino)

    The fact that the Prescott fared much better in it's launch compared to the Willamette might be a hint to not underestimate it. Prescott isn't really looking bad now, and I think it will hit stride faster then the Willamette core did.

    The next couple of years are gonna be really interesting.

    Damn, ya just gotta love it!
  • ntrights - Sunday, February 1, 2004 - link

    Great review!
  • KF - Sunday, February 1, 2004 - link

    I've grown to appreciate CRAMITPAL. If you read around the opinionated diatribes, he has some good stuff that people avoid saying for fear of retaliation. I suppose if I were in love with Intel, he would tick me off.

    But, it does look like Intel has created a CPU that should ramp up to speeds high enough to beat the A64 in 32bit mode, and that is all they needed to do.

    Regardless of how much heat that is going to take, Intel must have some way in the works to handle it.

    Looks like they might not charge an arm and leg for it, which is the biggest shock.

Log in

Don't have an account? Sign up now