Final Words

If you've read every word we've written about Centrino and the technology behind it and are impressed, then you haven't seen anything yet. The architecture and technology behind the platform goes far beyond what we're able to print, but the main point to convey is that the Pentium-M is not a Pentium III and not a Pentium 4, it's something very new and very different.

The approach Intel undertook with the Pentium-M is one that would be difficult for their competition (mainly AMD) to duplicate. It requires dedicating an entire design team to a project that will be taking significant risks and will be relying on perfect execution on the manufacturing side of things to meet product cycles. The sort of commitment Intel made with the Banias project required an incredible amount of resources, and put those resources at very high risk of being lost should anything have gone wrong. Maybe it was the passion that was burning within the hearts of the Israel design team or the desire to show Intel corporate what the team was capable of, but Banias was the first and only chip in Intel's history that was no more than 8 days late.

The future of the chip is also quite bright; there are a few tweaks that didn't make it into Banias that will be present in its 90nm successor known as Dothan by the end of this year. Dothan should be able to ramp frequencies up even higher, while providing an even larger 2MB L2 cache as has been leaked previously.

When looking at the present, the 1.6GHz Pentium-M is comparable to a Pentium 4-M 2.2/2.4GHz, with the 1.5GHz Pentium-M falling somewhere in the 2.0 - 2.2GHz range. For business/office applications, the Pentium-M is unbeatable, clearly outperforming the fastest mobile Pentium 4 at 2.40GHz. When dealing with games and media encoding applications, the Pentium 4 will be faster, but by no means will it be able to offer the battery life of a Pentium-M based system.

For the foreseeable future, the Pentium 4-M and Pentium-M will coexist, but the Pentium 4-M will be mostly restricted to the highest end desktop replacement notebooks, where battery life isn't as big of a concern. For all other segments, the Pentium-M and the Centrino platform will do their job better than the Pentium 4 ever could. The ultra low-volt Pentium-M processors will finally make high-performance Tablet PCs a reality and will be a good successor to current ULV Pentium III notebook solutions.

In the end, the Pentium-M and the Centrino platform in general will do more for mobile computing than any single technology we've seen in the past. If you want an idea of just how big of an impact they're having already, take a look at our roundup of the first four Centrino/Pentium-M based systems we were able to get our hands on in time for launch...

General Usage Performance
Comments Locked

8 Comments

View All Comments

  • zigCorsair - Wednesday, July 14, 2004 - link

    I thought it was a very informative article. Of course, I'll be upset if it's biased, but being a master's student in CS, many of the exact details I was looking for were in here, and for that I say thank you.
  • Zebo - Monday, May 10, 2004 - link

    I don't see whats so impressive. An athlon mobile 2600/2800 xp 35W version, which runs ~2000Mhz will kill these. To little to late.
  • Anonymous User - Wednesday, September 10, 2003 - link

    how the hell could this be a balanced and informative article when in their own analysis they ignored their own data?

    There is no mention of the anamolous nature of the BAPCO test..absolutely NOTHING...

    Its enough for me to question the competency of this site...and even to the point where I suspect that certain unethical compromises have been made.
  • Anonymous User - Wednesday, September 10, 2003 - link

    Yeah, I agree with Sprockkets... same reason Athlon XP loses to the P4 in this benchmark... someone was trying to make the P4 look better, and everything else look worse. Now all the sudden, this new great CPU is getting it's but kicked because of all the P4 optimizations (and probably non-P4 deoptomizations).
  • sprockkets - Tuesday, September 9, 2003 - link

    I wonder why the P4 trashes the PM on Content Creation Performance and nothing else? Maybe it's the stupid skewing toward the P4. Why else would it lose here and kick butt everywhere else? www.theinquirer.net has an article which brought this to readers attention.
  • Anonymous User - Thursday, August 21, 2003 - link

    "Without a trace cache, the design team was forced to develop a more accurate branch predictor unit for the Banias core. Although beyond the scope of this article, Banias was outfitted with a branch predictor significantly superior to what was in the Pentium III. The end result was a reduction of mispredicted branches by around 20%."

    Wouldn't he mean that the branch predictor was superior to the P4?
  • Anonymous User - Tuesday, August 19, 2003 - link

    looks good
  • Anonymous User - Friday, August 8, 2003 - link

    An outstanding well balanced article, after this read I feel I really know about Centrino. Thanks

Log in

Don't have an account? Sign up now