Forgot your password?
typodupeerror
Supercomputing Cellphones Java Hardware Technology

Mobile Phones vs. Supercomputers of the Past 247

Posted by timothy
from the crays-are-pure-sculpture dept.
An anonymous reader writes "The recently published Top 500 list of the world's fastest supercomputers is based on the Linpack benchmark developed decades ago by Jack Dongarra. This same test has been ported to Android mobile phones, which means that we can compare the performance of our phones against that of the supercomputers of the past. For example, a tweaked Motorola Droid can hit 52 Mflop/s, which is more than 15 times faster than the CPUs used in the 1979 Cray-1." But even today's most powerful cellphones don't come with an integrated bench.
This discussion has been archived. No new comments can be posted.

Mobile Phones vs. Supercomputers of the Past

Comments Filter:
  • 1979 tech still wins (Score:5, Informative)

    by Skarecrow77 (1714214) on Thursday June 03, 2010 @02:09PM (#32449104)

    For example, a tweaked Motorola Droid can hit 52 Mflop/s, which is more than 15 times faster than the CPUs used in the 1979 Cray-1.

    "The Cray-1 had 12 pipelined functional units" and had "floating point performance generally about 136 MFLOPS. However, by using vector instructiosn carefully and building useful chains, the system could peak at 250 MFLOPS."

    http://en.wikipedia.org/wiki/Cray-1 [wikipedia.org]

  • by flaming-opus (8186) on Thursday June 03, 2010 @02:24PM (#32449304)

    I thought it was strange that the article author was reporting that a cray 1 only produced 3.4 mflops on linpack, which had a peak performance around 130 mflops. Looks like the author doesn't understand the benchmark very well.

    If you look at the data quoted in the article, the n=100 result gives the Cray1 a score of either 3 or 12 mflops, depending on which entry you look at. There is no n=1000 result listed for the Cray 1, but one can expect, looking at the Cray XMP results, that it would be around 100, given the peak performance. The ETA10 would likely get a couple thousand mflops on linpack with n=1000.

    The Cray 1 is more than a little dated. That said, if you look at supers from the early 90's, they still can do things that modern commodity hardware can't. As fast as a xeon or opteron is, it doesn't have 300Gbytes/second of memory bandwidth. Even late-80's supercomputers exceed desktops in some metrics, though probably not in raw ALU performance if the data all fits into L1 cache. The cost to run a desktop, however, is pretty compelling, and they don't leak freon when they run.

  • by Anonymous Coward on Thursday June 03, 2010 @02:37PM (#32449490)

    I suggest you start here Mobiles through the ages [talktalk.co.uk] and come back here once you are familiar with the subject you are commenting on. Got to love people who are experts on subjects they don't understand.

  • by Anonymous Coward on Thursday June 03, 2010 @03:13PM (#32450020)

    Fail quoting Moore's law

    Moore's law allows for 1) more transistors in the same space (hence more power), or 2) the same transistors in a smaller space (such as the Droid).

  • by Anonymous Coward on Thursday June 03, 2010 @11:38PM (#32454550)

    > real business transactions, information handling / using / whatever is not much faster or efficient than it was in 70's.

    Huh? Those things got faster every decade. From simple levels (swipe the credit card yourself, instead of the cashier slooowly punching in numbers), to more sophisticated ones (doing stuff online, and it completes faster than it takes to load the order confirmation page). For other things, the speed of any single transaction didn't get much faster, but the total volume of transactions happening at any one time exploded. In some cases, many improvements stack - like air travel, where you can pretty much do all of your side of it yourself, complete with printing the boarding pass, and the computers handle figuring out the routes and schedules and seating availability for vastly more planes at once than were operated in the 70s.

    > You will see how the computer world stabilizes to same as any manufacturing - a couple of designers, a bunch of engineers, a lot of floor workers.

    Actually, that's one of the old myths. Programming doesn't really get any easier; it's not easier to do relative to the amount of work that needs doing, and it's not any easier to teach relative to the number of people who need to learn it. (IMO, we haven't even really gotten much better at teaching the *users* in the last decade). It's enough of a thinking job that the computer can't do it for us, and if you look into that in more detail you're likely to run into a formal proof explaining why, and a bunch of "technological singularity" AI articles.

  • by xded (1046894) on Friday June 04, 2010 @01:34AM (#32455116)

    But first try watching Primer [imdb.com], just to know what you shouldn't do...

    It's one of the best movies I've ever seen, but watch it 2/3 times before judging -- as suggested [villagevoice.com] by the director himself -- or use some reference timeline [freeweb.hu] when in doubt (spoilers ahead).

A language that doesn't have everything is actually easier to program in than some that do. -- Dennis M. Ritchie

Working...