Slashdot Mirror


New Power-Sipping Chips From Intel

bradlauster writes: "As Reported on InfoWorld.com: Intel will be announcing new mobile processors on Monday. One is claimed to run at 500 MHz and consume less than two watts! How's that compare to Transmeta's Crusoe ... and has anyone used one?" yomahz points to additional coverage at news.com. Isn't it nice what price/performance battles can do?

8 of 115 comments (clear)

  1. Re:overclockable? by cvincent · · Score: 3

    Yes, heat is one thing in overclocking but ever wonder why some guy can crank is 500E to 750 with the OEM heat sink/fan combo and the guy next to him has to use a peltier? You have to have good quality silicon to get to higher speeds and thats why dedicated overclockers will look for a certain week manufactured intel chip like...week 28 from ireland or something. Its because they have finer silicon. If intel makes these things operate on low power...they also have to have the quality silicon it takes to get higher speeds. If they use a .18 die process then sure...it might look good but to keep the price down to compete it will probably be a .25 or .22 micron. Less power doesnt also mean less heat...sure there is a direct link between more wattage and more heat but this thing may run so hot its not even funny even tho its only at 2 watts! We will see I guess....

  2. This will be very bad for Transmeta by bwalling · · Score: 3

    I think there was a very important statement in the news.com story. Something along the lines of 'Intel just has to get close. If I'm most MIS guys, I'll choose Intel over Transmeta.'

    That's very true. Even products coming from the likes of IBM and Gateway, most corporations (read: where the bulk of sales are) will choose a product with an Intel chip. Unlike Linux vs Windows, it is harder for an IT guy to see the merits of a Transmeta chip over an Intel chip. This will be a hard battle for Transmeta to fight.

  3. Re:??? by mrbinary · · Score: 3

    Actually (to respond to my own post - does that lower your karma or get you some bad slash mojo of some sort?) the article on CNET is better, they indicate that the AVERAGE consumption is less than X watts of power. My question is: why does the 600Mhz PIII consume less than one watt (on average) yet the Celeron 500Mhz consumes less than two watts? Is this part of Intel's plans to push consumers towards buying the more expensive PIII (or am I just being paranoid at the end of a long night shift?)?

    --

    ----
    Slán leat agus go n'eirí an bóthar leat
  4. Transmeta's low-power design is much better. by ZanshinWedge · · Score: 5
    ugh, it's distressing how often I am reminded of the usefullness of setting my threshold at 2.

    Anywho, the reported 2 watts of power is most likely the minimum amount of power the cpu can consume (following the long tradition of hardware and software manufacturers reporting numbers that are theoretically important but have little bearing on average performance/usage and are often misleading). The Intel (and as far as I know, just about every other CPU manufacturer other than Transmeta) power conservation design is very very clunky (bordering on bletcherous) which is vaguely analagous to running a car for a short period and then coasting in order to decrease gas usage. Transmeta's design, on the other hand, is much more elegant and thoughtfull. More important is the ability of the Transmeta chips to fine tune power usage for any particular application (running at the lowest speed appropriate). Relating to the example above, this would be like running at 25 mph instead of alternating between coasting and 80 mph.

    Transmeta's design is not only superior and on average more likely to use less electrical power (and consequently generate less heat), but is also easier on the processor (preserving lifetime, not that that matters anymore), and in general provides for a "nicer" interface between the CPU and other components (for example, a processor that is running slower, but is nevertheless always on, responds to external interrupts better and more reliably).


    On a side note, I'm really waiting for RSFQ technology (Rapid Single Flux Quantum logic / memory, which (BTW) requires superconductors so I'll probably not see it anytime soon) which uses so little power that you should be able to operate in the 1 teraflops range using less than a tenth of a watt.

  5. Less power consumption, less heat, less fans ? by Ma�djeurtam · · Score: 5

    How about using these low-power CPU's in desktop computers ? Without going in the matter, it could be seen as an useless idea : why put a less powerfull computer ("only" 500 Mhz) in a desktop PC ?

    Mobile computers have one feature in common with iMac's : they haven't any fan. The current trends in computer is the more fans in the box, the better computer (see this recent Slashdot Poll, too). It's usual to find one big fan for the power supply, one or two of them on the CPU and on the video card, and even on the HD, the CD/DVD player, and on the CDR.

    It makes very noisy computers ! Now, I've noticed that the evening, just after turning off my computer and calming down in a quiet room, my ears continue to burden, just as they continue to buzz after a hard rock concert (just a slight exaggeration, here). This can't be good. Really, that noise annoys me.

    That's why I'd like to see low-power CPU's in fanless desktop PC's, just like in iMacs. I, for one, just don't care about speed on my desktop PC : my old P-II 266 Mhz is enough for my daily needs.

    (And I'm not even talking about how inelegant is the principle of fans itself.)

    Stéphane

    Have you checked out Badtech The daily online cartoon?
    Have you checked out Badtech The daily online cartoon?

    --
    Instant Karma's gonna get you, Gonna knock you right on the head (John Lennon, 1970)
  6. Re:Power Budget by stripes · · Score: 3
    Does anyone know what the typical power budget is for a laptop computer? How much power is used by the CPU, hard drive, DRAM and display?

    Depends strongly on the portable, and even how you use it. If you use the disk (or CD ROM, or anything it has to spin, esp. spin fast) that sucks a whole lot of your power. If it has to use active cooling, then the fan will suck power too. The backlight (if in use) sucks a lot of power also. The CPU can be pretty low on that list. Or fairly high (the 650/500Mhz SpeedStep things can suck a fair amount of power, maybe more then the backlight!). The CPU also makes heat (as does the drive) which can make the fan spin....

    If you look at a really simple portable like a Palm Pilot the backlight sucks most of the power, the CPU a distant second, and the DRAM a more distant third. AA rechargables last me over a month with no backlight use, about three weeks with a fair amount of backlight, or about 10 hours of dinking with "Pocket Life" with the backlight on (est from looking at the batt meter before and after 2 hours of said dinking). I've heard that it'll keep it's DRAM running for 2 or 3 months on "dead" battries.

    I would love to have a laptop computer that could run for days on a battery charge, even if the CPU was slow by desktop standards.

    The only thing that comes close is the Palm, and in part because you tend to turn it on, use it for three minutes, and turn it off for a while (what's that phone number? Gotta remember to get the car serviced this week. What time is it anyway? When is sunset today? [dee-DEE-dee-DEE-dee-DEE] Oh! Gotta run if I want to catch buffy tonight...). Still, it is a computer :-)

  7. Power Consumption info for notebook components by Guppy · · Score: 3

    Here's a chart I put together from Intel's Mobile Power Guidelines Rev. 1.00, which lists target power consumption values for notebooks. I've included Intel's numbers for the Maximum, Minimum, and Average (when running Winbench 3D) power consumptions, in watts.

    Please excuse the formatting, Slashdot seems to strip out extra spaces, so I put in underscores. Plus, I'm also having trouble with the lameness filter.

    _____________________ Mini Notebook_______ Full Size
    _____________________ Max__Min___Ave___ Max___Min___Ave

    CPU & L2 Cache ______ 6____0.36__5_____ 12____0.36__9.5
    Memory Controller ___ 2.1__0.05__1.4____ 2.4__0.3___1.6
    System Memory _______ 1.3__0.35__0.7____ 2.5__0.35__1.3
    Graphics Subsystem __ 1.4__0.2___1.0____ 3.0__0.8___2.4
    IO Subsystem ________ 2.0__0.04__0.5____ 2.68_0.4___0.6
    Audio _______________ --___0.7___--_____ --___1.6___--
    Modem _______________ --___0.3___--_____ --___0.4___--
    Hard Drive __________ 6____0.1___1.4____ 6____0.1___1.3
    DVD Drive/CD ________ --___--____--_____ 6____0.2___1.4
    1394 Controller _____ --___--____--_____ 1.2__0.003_--
    CardBus _____________ 2.5__0.67__--_____ 5.0__0.67__--
    USB _________________ 2.5__0.0125--_____ 2.5__0.0125--
    LAN _________________ 0.6__0.4___--_____ 0.6__0.4___--
    Power Supply ________ --___1.5___--_____ --___2.6___--
    Cooling _____________ --___--____--_____ --___0.5___--
    Other _______________ --___0.8___--_____ --___1.0___--
    LCD _________________ 3.05_--____2.8____ 4.75_--____4.3

    SYSTEM Average____________16.7_______________29.1

    The high maximum value for the hard drive comes from the power required for the initial spin-up. I didn't see a listing for separate L2 and CPU power consumptions, but as you can see, together they take up about 1/3 of the average power consumption in a full-sized notebook, with the graphics system (chip plus screen) being one of the components that takes up another big chunk.

  8. This will be DISASTROUS for Transmeta by Ars-Fartsica · · Score: 3
    The poster, and the quote cited from the original article are bang-on - if Intel can come close to Transmeta's power consumption goals, Messrs Ditzel and Torvalds will be shown the door shortly after their requisite fifteen minutes of fame.

    Strangely enough, Intel proved that branding can be applied even to something as disconnected form the user as a CPU, the ultimate "black box" that by rights should be brand-neutral from a user/consumer perspective. Nonetheless, the "Intel Inside" branding strategy has been a success, with most PC makers touting the Intel branding on their boxes.

    Don't underestimate the stickiness of a brand and a product that people know to work. Transmeta has none of that cachet - in fact, the Trnasmeta cachet is of ".com media darling" - which is probably exactly opposite to what CPU branding should represent.

    If Intel can get close to the Transmeta power consumption goals, its game over for Transmeta.