Before reading this article, be sure to read our first Desktop Pentium M investigation, as we will not be revisiting any of the discoveries/conclusions in that article.

At this year's Spring IDF, Derek Wilson, Johan De Gelas and I all sat across the table from Intel's Justin Rattner and asked him a question that had been on our minds ever since Prescott's introduction.  The question went something like this:

Recently, Intel has shifted their focus away from ramping clock speed to increasing performance through other means such as exploiting TLP (Thread Level Parallelism).  Given that Intel's direction today seems to be in sharp contrast to the direction of the NetBurst architecture and especially the direction of Prescott, were the past five years of microarchitecture improvements and innovations essentially a waste? 

Rattner's exact answer encompassed a few items, but the main points were as follows:

1) At the time of Prescott's inception, clock speed is what sold, so clock speed was obviously the top priority; and they did quite well at that.

2) The past few years weren't a waste; after all, research continued in the direction of the Pentium M while all of the work on Prescott and Tejas were being done.

The second statement was particularly interesting because it was the first time that we received something tangible from Intel in regards to this question.  Clearly, significant elements of future Intel desktop processors will be derived from the Pentium M architecture, but as we saw in our first desktop Pentium M investigation, the time for the Pentium M on the desktop isn't now. 

As a mobile architecture, the Pentium M can't be beat.  That statement alone is something many assumed that we were contradicting in our desktop Pentium M article, but the purpose of that article was to look at desktop Pentium M performance, where we determined that it fell behind.  In the mobile world, without significant engineering investment, it is doubtful that the Pentium M will meet its match anytime soon. 

On the desktop, we discovered that there were a handful of limitations to the Pentium M's success:

1) 855GME chipset with only a single-channel DDR333 memory controller
2) Expensive motherboards and high total cost of ownership
3) Low floating point/SSE performance
4) Severely motherboard-limited overclocking

The combination of the four items above meant a few things.  While the Pentium M was an excellent contender in general use applications, its total cost of ownership was significantly higher than an Athlon 64 that performed similarly.  In other applications, the Pentium M simply fell behind the competition for architectural reasons. In those cases, its high price didn't help it out at all either.  The saving grace in many cases required overclocking, but the desktop Pentium M motherboards were far from overclocking monsters for those who were interested. 

In the end, our stance on the Pentium M as a desktop solution was that it's more of a wait-and-see proposition.  If motherboard manufacturers could produce cheaper, better equipped motherboards and if those elements improved performance, then the Pentium M would be worth another look as a desktop alternative. 

We hadn't expected such a solution to come around this soon however, but it has, and not in the form that we originally thought it would.  Both AOpen and DFI have indicated that they were working on updated motherboards based on the mobile 915 chipset, but that they were still months away.  So when we received word that ASUS had a solution to the Pentium M desktop problem, we were caught off-guard.  And rightfully so, as their solution is far from just a new chipset...

ASUS CT-479: Socket-478 to Socket-479 Adapter
POST A COMMENT

48 Comments

View All Comments

  • Avalon - Thursday, March 24, 2005 - link

    In Soviet Russia, the Pentium-M owns you!
    In all seriousness, though...I was reading the benchmarks and wondering where the 855GME desktop P-M 770 benchmarks were. You only had the 770 on the Asus board. While this is all fine and dandy, it doesn't show us the full extent of the benefits of the Asus pin adapter over using an 855GME desktop board. While you could extra some data, I would have liked 770 desktop benches as a comparison. All in all, though, the Asus adapter is definitely nice due to the price alone.
    Reply
  • Wonga - Thursday, March 24, 2005 - link

    I don't think anyone can criticise Intel too harshly for releasing the Netburst architecture. Yes, Willamet stank, and Prescott isn't exactly an efficient core, but a Northwood processor was the best CPU money could buy for about 18 months before the K8 hit the scene. So, it doesn't really matter how Intel got there, but Netburst was the architecture to have for those 18 months.

    Anyway, Anand, I'd appreciate it if you could see if those adaptors would work on other S478 motherboards :)

    Oh, and thanks for a great article! Perhaps that Pentium M horse that was beaten so much can now finally rest in peace!
    Reply
  • sprockkets - Thursday, March 24, 2005 - link

    While the boards and tech mentioned here are nice, why not also test some of the PM mini-itx boards at www.logicsupply.com and see how badly it can kick an Mac Mini's ass? OR, how well it can perform in such small space with little heat and noise output?

    Reply
  • xtknight - Thursday, March 24, 2005 - link

    amazing it beats the fx-55 at q3 source compile. maybe i need to get my eyes checked but does that say INTEL? Reply
  • mkruer - Thursday, March 24, 2005 - link

    Here is what I get out of this,

    1. P-M is better clock then the P4 and apparently the AMD64, (at least in some applications)
    2. Future versions of the P-M while adding better FP and SSEx Instructions will still not have 64-bit technology.
    3. Adding better FP and SSEx, will also increase the minimal thermal rating, resulting in a lower maximum over clocking ability, and higher heat dissipation

    All and all, it looks to be about the same as AMD64, when all the upgrades are added in.
    Reply
  • ncage - Thursday, March 24, 2005 - link

    #5 It probably wouldn't be that good for intel to drop Pentium M prices. Yes there would get more sales from techies like us but that wouldn't increase there sales that much. Where they get sales would be from labtops and their profit margins would drop for that that is why overall it would be stupid for intel to drop the prices on them. Reply
  • mlittl3 - Thursday, March 24, 2005 - link

    I'm sorry but I have to say a few more things.

    I counted up all the benchmarks (not including overclocking ones) and the 2.13 GHz PM beats a 3.2 GHz PIV in 22 out of 33 benchmarks. For a 1 GHz+ underclocked processor, the PM wins in two-thirds of the benchmarks. Now correct me if I'm wrong but doesn't the 2.13 GHz PM consume 25W of power and the 3.2 GHz PIV consume around 100W. That's four times the power for 66% less performance.

    Multiply that 4x waste by all the PIV's that Intel sold and we must have wasted megawatts upon megawatts of power for nothing. You can't say its technology improving because the PM is based on old technology.

    Intel wasted so much electricity and polluted so much more of the environment because what it couldn't save face when it released the 1.13 GHz coppermine PIII and had to recall it because of architectural problems. Therefore, it had to release the PIV to make itself look better from a marketing standpoint.

    This just blows my mind.
    Reply
  • EODetroit - Thursday, March 24, 2005 - link

    Nice article. While now we know that the Pentium M probably isn't practical for the desktop in most situations when price is considered, this review is important because it gives us something to compare the Yonah processors to when they come out. And those of us that were curious about the P-M on the desktop finally have some answers, so thanks for that.

    How high could it overclock, anyways? It sounded like you got it up to the point where the memory would be at DDR400 with the 5:4 ratio, and called it good. But how much more can it be pushed? What about with better cooling?
    Reply
  • Googer - Thursday, March 24, 2005 - link

    Thank God there are no Russian Jokes on this page! Reply
  • Googer - Thursday, March 24, 2005 - link

    I did not see any benchmarks of the Pentium M 770 on the 855GME so we could compaire APPLES to APPLES the only system it was run on was 865, It's a bit of an unbalanced compairison. Anand, Just how much of a differance does 865 offer the P-M 770 over 855? Reply

Log in

Don't have an account? Sign up now