Slashdot Mirror


IBM Releases Compiler for Power4 and G5

davids-world.com writes "IBM offers its optimized XLC compiler not just for Intel CPUs, but also for its own G5 processor (article in German at Heise). Unlike gcc, it is optimized for the G5 and achieves a major boost in speed, as first results show. I guess we will have to compare the new benchmark data (once available) with the data we get with the optimized Intel compiler for Xeon. The compiler is available for download now."

7 of 471 comments (clear)

  1. can it compile the kernel? by shaitand · · Score: 0, Troll

    worthless if it's not gcc compatible to compile the kernal.

  2. just in time.... by johnpaul191 · · Score: 0, Troll

    .....for Steve Jobs to have new benchmarks for the Mac Expo in Paris in mid September? great! more fuel on the fire.

  3. Re:Here we go again: by stingerman101 · · Score: 1, Troll

    Prove it, the Opteron alone costs $999/ processor and AMD hasn't sold more than 10,000 yet. IBM has already sold more G5's than the Opteron and Itanium combined in less than 1 Month. The Dual G5 is selling for 2699 at educational prices with 512MB of RAM, high-end GPU, SATA, PCI-X, Hyper-transport, AGP Pro 8X...

  4. Re:Here we go again: by stingerman101 · · Score: 0, Troll

    Bochs does and OPenOSX.com claims that MacBochs for the G4 is just as fast as VPC. So who needs VPC anyway?

  5. Re:Here we go again: by LilMikey · · Score: 0, Troll

    "The G5 could also run Linux, and via some emulators, Windows software as well. I think the G5s run plenty of software."

    So your solution is to pay twice as much to run emulated software 1/2 as fast. Don't get me wrong, it is a step above Apple's normal stance(pay twice as much to run 1/2 as much software 1/2 as fast) but as for me, I'll go the budget consious road.

    --
    LilMikey.com... I'll stop doing it when you sto
  6. Got something to back that up? by twitter · · Score: 0, Troll
    Maybe they think GCC is a piece of crap (which for high performance computing, it is).

    pinkployd, you anoy me.

    Do you have any kind of benchmark to back that silly assertion up? Can you even define "high performance computing" for me? If you mean scientific computing, gcc is in wide use. Do a little searching and you will find goodies like this. I'd love to see the licensing fees for a 1,000 node cluster. Put up or shut up, please.

    It's all nice and good of IBM to make a compiler for their chips. I hope they can be persuaded to donate chunks to gcc and think that it would help drive their own sales. It would be silly to hold their own work on C and FORTRAN against them.

    You, taking advantage of the news to flame GCC and all the good people who work on it, just shows what a nasty little troll you are. Why not spend your time constructively? I'd love to know what compilers you use to get your work done. Hey, you might even contribute to GCC in your spare time. Until then, piss off.

    --

    Friends don't help friends install M$ junk.

  7. Re:Let's be objective by siskbc · · Score: 0, Troll
    Do your homework. Apple modified the tests to reflect the actual shipping models, since they were running on prototype G5's. These issues have long been put to rest and Apple just updated their results the other day with actual shipping G5's

    And their modifications to the Dell? Since they got it as it shipped, and they modified it, they didn't test it as it shipped.

    Get out of denial, x8 is not a religion, it's a processor for goodness sake.

    Interesting. You seem to be much more riled up than me - I would advise you to consider the same. Btw, I use both macs and intel machines at work, and I don't have any preference at all, so it would be hard for any actual denial to exist. I might question you on the same basis, however.

    For what it's worth, you'll never see a PC user really get upset because someone insulted his processor, but I could point to numerous examples of this phenomenon with some mac users, though obviously not all.

    --

    -Looking for a job as a materials chemist or multivariat