Huh? Mactel, for real?

Huh? Mactel, for real?

Summary: Niagara rocks. You want low power use for a laptop? How about an eight way 1.4Ghz SMP core with TCP/IP and cryptographydone in hardware - at 65 watts flat out. There are some serious software issues, but get past them and you've goteight to ten Xeons in the box - at 65 watts.

SHARE:
TOPICS: Hardware
16
Come on Steve, pull the other one. I know you previously referred to Intel as Plan B but really. I'm an Apple loyalist - bought my wife one in 1988 and she's never had anything else since. But this? Know what happens when a garbage truck collides with a Porsche? Remember your hardware sales at NeXt? know what happened at SGI? Great products that went Intel -and promptly became irrelevant in the market.

You had me believing you and trying to figure out what on earth might be behind the decision -until you mentioned that Intel's chips run cooler and offer a better roadmap.

They don't: unless you're comparing a Pentium M to an overclocked G4, but that's where the roadmap comment comes from too, isn't it? It's not that Intel has a better roadmap - it's that IBM's doesn't include you except as a subsidiary.

Unfortunately Intel doesn't have one either. In fact if they don't have a rabbit hiding in the hat, they're not going to last five more years as an independent - and you may think that's a BS prediction, but I'm not just the only pundit to call Cell right, something I wrote for Linuxinsider last year should seem a little more compelling today:

 

This will get more interesting if, as reported on various sites including Tom's hardware, IBM has been burning the candle at both ends and will also produce a three way, 3.5Ghz version of the PowerPC for use on Microsoft's X-Box. Whether that's true or not, however, my belief is that IBM chose not to deliver on its commitment to Apple because doing so would have exacerbated the already embarrassing performance gap between its own server products and the higher end Macs. Right now, for example, Apple's 2Ghz X-serve is a full generation ahead of IBM's 1.2Ghz p615, but costs about half as much.

Unfortunately this particular consequence of Apple's decision to have IBM partner on the G5 is the least of the company's CPU problems. The bigger issue is that although the new cell processor is a PowerPC derivative and thus broadly compatible with previous Apple CPUs, the attached processors are not compatible with Altivec and neither is the microcode needed to run the thing. Most importantly, however, the graphics and multi-processor models are totally different. As a result it will be relatively easy to port Darwin to the new machine, but extremely difficult to port the MacOS X shell and almost impossible to achieve backward compatibility without significant compromise along the lines of a "fat binary" kind of solution.

In other words, what seemed like a good idea for Apple at the time, the IBM G5, is about to morph into a classic choice between the rock of yet another CPU transition or the hard place of being left behind by major market CPU performance improvements

Good thing you had a Plan B, Steve; too bad it was the wrong one.

You know what the right one is? Ya, I know, it's disgusting but I'm going to quote myself again, and from the same article too:

 

So what can Apple do? What they should have done two years ago: hop into bed with Sun. Despite its current misadventure with Linux, Sun isn't in the generic desktop computer business. The Java desktop is cool, but it's a solution driven by necessity, not excellence. In comparison, putting MacOS X on the Sun Ray desktop would be an insanely great solution for Sun while having Sun's sales people push SPARC based Macs onto corporate desktops would greatly strengthen Apple.

Most importantly, SPARC is an open specification with a number of fully qualified fabs. In the long term Apple wouldn't be trapped again and in the short term the extra volume would improve prospects for both companies. Strategically, it just doesn't get any better than that.

Niagara rocks. You want low power use for a laptop? How about an eight way 1.4Ghz SMP core with TCP/IP and cryptography done in hardware - at 65 watts flat out. There are some serious software issues, but get past them and you've got eight to ten Xeons in the box - at 65 watts.

Sun's president, Jonathan Schwartz, put a nice invitation for you in his blog last Sunday. Maybe you should think about it, go for a drink with him, talk about the threat IBM poses to both of you. Get him to tell you about interval math on a software embedded array processor -you might like what you hear. And remember one thing: what you just did? Nobody at IBM is going to call a board meeting over it - but a Sun alliance? That would be different.

Topic: Hardware

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

16 comments
Log in or register to join the discussion
  • Goodbye Apple!

    Did Apple leave a suicide note? Perhaps this switch means that they will not have to write software to be able to use the dreaded and feared OS that "Most" people use and like! Too bad, I kinda liked listning to the Mac Bunch on these forums curse both the open source folks and MS! Now all we will have for amusement is the Open Source folks, yuck, they are far from being a wise as the Apple folks were! Oh well, one less computer builder to follow!
    blacksheepxlch1
  • Possible "seppuku"

    Hopefully not. My guess is that apple won't use normal P4s or P5s, but a modified Intel processor with the added vector processing that AltiVec provides (apple is co-owner of the technology, so it can license it to Intel for apple's exclusive use). That should improve multimedia performance and simplify code porting, besides keeping a differentiator for apple (e-machines/compaq/dell won't be able to run vanilla apple apps at the same performance).
    It sounds fine BUT don't subestimate AMD or Intel themselves (they can in the future provide a similar engine for everybody else in the PC community that with little or no translation could outperform Altivec). It has happened in the past, like 3DNow; MS outwitting OS2; and the old IBM mistake that created the PC monster that almost killed it (they falsely tought that their propietary BIOS was enough to keep competition behind in the x86 market...).
    Any way, I am an apple fun because they are an innovating company, but Intel recently is behind in performance, power comsumption and creativity in their architecture, so if x86 is going to rule, my options are Linux (why pay for OS X?) or even better a future OS running on top of PS3 or XBOX360. Those will be killer platforms!
    oortizsilva
    • Seppuku? It doesn't sound like it

      I hate cross-posting, but as I stated in another talkback forum on the same subject, it will be a cold day in hell before Steve lets anyone run anything other than Mac OS on a Mac, intel or not, and there's no way in hell Mac OS will run on anything other than Mac branded hardware. They love being in control, and they love being elitist.
      On the other hand, if you ever manage to make linux boot up on the Xbox 360 and use all that processing power, please give me a call, 'cause i'm buying it right away
      SantiagoCrespo
      • Careful

        "On the other hand, if you ever manage to make linux boot up on
        the Xbox 360 and use all that processing power, please give me a
        call, 'cause i'm buying it right away"

        You might want to keep that quiet. The last guy who succeeded at
        hacking an XBox with Linux got sued.
        Jkirk3279
        • True...

          Although I couln't care less about the games, I want to make full use of a cell processor for serious processing needs, since obviously we're years away from seeing such a processor architecture in the desktop
          SantiagoCrespo
          • You are aware that the Cell is going in the ...

            ... Sony Playstation 3 and not the Xbox360 right?
            ShadeTree
  • I'll buy one!

    I'm a Linux user but always wanted a G just to have one. I like Apple always have. In fact I used to service a few. PC's are easy to build and work on, parts are everywhere. I really think people will buy more mac's People tend to think in big numbers (speed) when it comes to buying computers. IBM couldn't meet the ghz mark apple wanted. End users don't understand techie talk. They ask simple questions like is that one faster than this one? Yep I like OSX another virus free OS unlike the always infected and hacked M$ OS. I'll never leave Linux but I'll buy a MacTel when they come out. I will never go back to using a virus again!
    xstep
  • Correcting an omission -and a clarification

    I got a few emails overnight about this one; most of them to do with one of two areas:

    1- the relationship between the Pentium M and the G4; and,

    2- the nature of the Niagara CPU as a server, not a workstation, product.

    First: some people seem to think I compared the G4 to the Pentium M. It obviously isn't clear from the text, but I intended to suggest that Job's point about Intel having the better "roadmap" applied to the G4/Pentium M comparison. There's no question IBM's a generation behind in getting a G4 replacement out to Apple -and that the Pentium M, for all its problems, has more potential than an overclocked G4.

    On the other hand, IBM's 3 core, 6 thread, 3.2Ghz PowerPC base CPU for Microsoft's X360 suggests that they could have delivered for Apple. Similarly the master processor in the Cell grid is a full PowerPC (it even has an Altivec) intended to run at 3.9Ghz. Again, the question is why they choose not to apply this technology in their contract with Apple?

    The second type of question is inherently more interesting. The Niagara is intended to power servers, not workstations or laptops. Since it's a hardware SMP machine turning cores (and buses) off in software isn't much of a stretch - but the central argument is that the relatively low speed of each core makes the machine inappropriate to workstation processing.

    Some very smart people hold that view -but I think it's a software issue that's more perceptional than real. Pretend for a moment that you have ten 3.0 Ghz Xeons in a laptop (with magic cooling of course since that would run about 1,400 Watts) and then ask how a smart software designer would use that.

    The answer, I think, is that most would be turned off most of the time, but called into action as needed.

    If you're using the laptop to run Word, a single 1.4Ghz Niagara core would offer overkill performance -and if another one is running the OS and handling networking, you'd never notice any delays. On the other hand suppose you're flipping colors on a big shading job? Today that's handled serially but there's no real reason it can't be made parallel -meaning that the Niagara wakes up a few cores and delivers the results long before a P4 or Xeon could.

    Don't misunderstand, Sun's throughput computing represents a big change -but probably not as a big a one as IBM's cell, and those are the only two directions now known that can extend the reign of Moore's law another decade or more.
    murph_z
    • You have demonstrated more then a ...

      ... few errors in your nalaysis. The biggest one that only the Sun and cell approaches can keep up with Moores Law. Multicore also Keeps up with Moores Law as it has to do with the doubling of transistors and not the MHz of the processor. The other major problem you have is predicting the end of Intel in 5 years. It is more likely that Sun will be gone then Intel.
      ShadeTree
  • MacSPARC?

    As politically attractive as it might be for Apple to dump IBMs PowerPC architecture in favor of Sun's SPARC architecture, and as straightfoward as a SPARC port might be for MacOSX, it would not be in Apple's best interest to get in bed with Sun while Microsoft is on the other side of that bed.

    Placing Sun in the middle -- between these two would-be rivals (Gates & Jobs) -- would not be in Sun's (McNeely's) best interests either.

    This 'marriage of convenience' between Sun and Microsoft should be expected to last long enough to dispatch RedHat and cripple Linux but beyond that it will probably not last as long as the IBM - Microsoft partnership lasted ... and we all know how that turned out!

    SPARCs technological advantages over PowerPC may indeed be real but switching to SPARC presents Apple with no opportunity to cut its production costs -- switching to Intel does!

    Right now, Apple is not on Microsoft's Radar -- nor is it likely to be as long as it poses no real threat to the Windows desktop. Linux, on the other, poses a serious threat to Microsoft's enterprise server strategy. It poses the same threat to Sun -- and that's why the Sun-MS partnership REALLY exists.

    Other than Sun, Apple is the only major player in the UNIX/Linux space who does not need to worry about lawsuit-happy SCO because OSX is based upon FreeBSD -- which has already been declared by the courts not to infringe upon SCO's IP interests in UNIX SVR4. Further, the OSX interface is rock-solid and could pose a real threat to the Liunux desktop space -- IF it could compete in that space at a more attractive price-point.

    If Apple is to improve its market penetration, it MUST bring its prices into line with other Intel alternatives. (Competing with Dell -- running Windows or Linux -- would be the prize nut to crack!)

    Otherwse, Apple's move to Intel is an exercise in futility.
    M Wagner
  • Take the blinders off, man

    Funny sort you are. Your loyalistic attitude toward Apple has blinded you to the huge strides Intel has made in their processors lately. Not only will they SMOKE anything RISC has, and yes, they do run cooler, another problem with AMD. Check out the new dual and quad core Intels coming down the pike, Steve Jobs actually made a correct decision for a change. Read his own statements in the interview posted elsewhere on ZDnet. Wake up, smell the Roses!

    Dan
    Crestview
  • Oh, please. It's profit & mindshare uber alles.

    If Apple has to make the move to x86, then it's not going to waste it's marketing money on fruitlessly explaining to the masses that an AMD chip is the same as an Intel chip. It's certainly not going to repeat the failed experiment of touting a relative unknown chip to the public.

    They tried that with PowerPC chips & they definately didn't win the war on being faster despite the lower clock rate.

    Besides, they left IBM because they ould be customer #4 behind M$, Sony & Ninetendo. Why would they want to be #2 behind a competitor for the high-end and server market like Sun?

    John & Jane Public know "Intel inside" means they can be comfortable about buying themselves or the little 'uns that PC. "AMD? PowerPC? Jeez, I just want a good computer and hey, that Mac is supposed to be a good system with a fast chip I've heard of".
    YuridaMan
  • It's not really about the CPU

    It's just another chipset. No big deal. Apple is still going to
    control the design of the motherboard and they use
    OpenFirmware. What killing M$ is that I know a lot of people
    always have to upgrade to the newest computer in 2-3 years in
    x86. They think is the OS (Windows) and to some degree it is
    because the older the BIOS the harder it is for the OS to deal
    with it. Apple use a FlashROM for OpenFirmWare so it is always
    easy for users to update. Has anyone ever really updated the
    BIOS on their machine. No. They just buy a new one when the
    old one doesn't perform as well anymore.

    OS X as been designed to work on the PPC and x86 - Darwin
    anyone? We don't like Macs because of the chip inside most
    consumers don't care about PowerPC this or x86 that. It's usually
    is it faster than my current machine? IBM won't or can't design
    the PowerPC to be 3 Ghz - regardless of Cell being a PowerPC
    derivative they have yet to really start putting those 3.5 Ghz
    chips out yet. People buy because will it run my favorite apps.

    No envision this. Your running you Mac OS X/86 computer and
    you want to play a game. The game is designed to run on
    Windows boxes so can you play it? Why yes because how hard is
    it to have an emulation layer mimic windows. The App doesn't
    know better and it just works. Which is what most consumers
    want.

    Apple will be able to gain consumers when the Windows apps
    just work on Mac OS X thru the Rosetta - yeah it's really for G4/
    5 PowerPC apps with the current Mac OS X. The price per
    computer really won't even need to come down because Apple
    will be able to go head to head with Dell/HP/Gateway. Yeah
    they'll sell you that cheap Windows box but try upgrading the
    graphics card or the soundcard or that ethernet card. They can
    with Apple computers but not with those other guys because
    those things are now apart of the motherboard. Have you ever
    tried to upgrade the graphics system of an HP? It aint easy if you
    don't have the right settings on the Motherboard.

    So when you say Mac your talking about an OS that is designed
    from the Motherboard up to just...well...work. M$ has to design
    Windows to work for 50,000,000 different setups and they still
    have major security issues. So stop whining because it's not
    really about the CPU...
    macshaggy
  • Why go sparc on the desk tops?

    When Sun isn't even going sparc

    http://www.sun.com/desktop/
    ajeskey
    • SPARC on the desktop?

      Sun isn't the only company that makes SPARC processors--in fact, as far as I know, SPARC is actually an open standard adopted by such companies over the years as TI, Fujitsu, and of course Sun. They could probably get some processors on this architecture from one of many distributors.
      Third of Five
      • Message has been deleted.

        myfevertoy