The History of Banias

The face of mobile computing actually changed a little over three years ago, when Intel announced a chip they were working on called Timna. Timna was designed to be a desktop CPU for a new market, the sub-$600 desktop PC segment. The CPU and platform were going to enable this by integrating the memory and graphics controllers onto the CPU's die itself. If you're curious as to why Intel is so adamant about not integrating a memory controller on die, the failure of Timna could be a strong case against the idea.

At the core, Timna was based directly on Intel's Pentium III but the entire processor, from the way the chip was laid out to the way it was packaged was optimized for cost. Wherever money could be saved, Intel's Israel design team managed to cut costs, and in doing so became intimately familiar with the Pentium III architecture. Much of the technology and cost saving techniques that were developed in Israel for Timna ended up being used by other groups at Intel in everything from desktop to server CPUs, so even the Pentium 4s we've got today have a little bit of Timna in them.

As we mentioned before, Timna was to feature an integrated memory controller, but at the time it was on the drawing board Intel still had delusions of grandeur about RDRAM being the future of all memory technology by 2001. The high price of RDRAM coupled with the lack of demand for such a low cost chip (OEMs weren't too happy with nonexistent profit margins on sub-$600 PCs), led to pulling the plug on Timna just after the chip was completed.

For Intel's Israel design team, the death of Timna was like the loss of a family member. A CPU that they had worked feverishly on for quite some time was now never going to make it to fruition; and the most painful part of it all, especially to an engineer, was that the Timna failure had absolutely nothing to do with the design of the chip, and everything to do with the outside factors that a design team cannot control.

Things were so bad for the Israel design team that when they were immediately summoned to begin work on a dedicated mobile processor, they were almost in a state of depression. How would you feel if your blood, sweat and tears were poured into something that just ended up a lost memory? Luckily, for the sake of the Israel design center and Intel's mobile computing plans, the Timna team didn't spend too much time mourning their loss and quickly began work on designing Intel's first truly mobile processor.

The design team looked at the Pentium 4's architecture and quickly concluded that it wouldn't be appropriate for a mobile microprocessor. An extremely long pipeline for a CPU designed to run at very high frequencies was not conducive to building a processor that could run within a 24W power envelope. There were some features that they could borrow from the Pentium 4, but for the most part, the NetBurst architecture was not one that could be considered as the basis for the CPU that would end up known internally as Banias.

One benefit the Israel design team had was their intimate experience with the P6 architecture; although the experience was mostly from a cost reduction and integration standpoint, both of those areas left the team with a solid understanding of the inefficiencies of the P6 core as well as what could be brought over into Banias unchanged.

What was born out of Timna's ashes and the extreme dedication/talent of Intel's Israel design team was a processor that can easily be classified as a Pentium III with a Pentium 4 bus, but wrongly so. What Banias ended up being, was the most interesting microprocessor architecture that we've ever seen…

Index Longer than a Pentium III, Shorter than a Pentium 4
POST A COMMENT

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. Reply
  • 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. Reply
  • 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.
    Reply
  • 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). Reply
  • sprockkets - Tuesday, September 09, 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. Reply
  • 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?
    Reply
  • Anonymous User - Tuesday, August 19, 2003 - link

    looks good Reply
  • Anonymous User - Friday, August 08, 2003 - link

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

Log in

Don't have an account? Sign up now