Stories
Slash Boxes
Comments
Search

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

4.7GHz IBM Power6 Spotted

Posted by kdawson on Sun May 20, 2007 09:19 PM
from the more-power-to-ya dept.
IBM Hardware
Ilgaz notes that The Register has posted benchmark results from Oracle 11i running on four 4.7GHz Power6 chips. Quoting: "The speedy chips confirm IBM's boasting that Power6 would arrive near 5GHz. They also show that IBM's customers have a lot to look forward to in terms of raw performance." Rumor has it that the Power6 chips will be announced on Tuesday.
+ -
story
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.

4.7GHz IBM Power6 Spotted 50 Comments More /

 Full
 Abbreviated
 Hidden
More
Loading... please wait.
  • Power6 sounds like it's going to be pretty damn cool - Perhaps Apple made a mistake jumping to intel so soon...

    *sighs* I for one yearn for the days of smugly ending any performance argument with some PC user with "Well, we've got Altivec & Altivec is magic."

    • Re:Did Apple make a mistake? (Score:4, Insightful)

      by jcr (53032) <jcr@ma[ ]om ['c.c' in gap]> on Sunday May 20 2007, @09:24PM (#19203651) Journal
      Nope.

      If Apple had been waiting until now for a 64-bit chip they could put into a portable, they'd be in Very Big Trouble.

      The PPC has a lot going for it, but Apple made the right choice.

      -jcr

      • Re:Did Apple make a mistake? (Score:5, Interesting)

        by mrchaotica (681592) * on Sunday May 20 2007, @09:30PM (#19203731)

        What I don't understand is, since Mac software has to be Universal nowadays anyway, why Apple doesn't just permanently keep its lineup as a mix of PPC and x86, picking whichever chip suits the particular machine they're designing at the time? Power6 Xserves along side Core 2 laptops... it sounds good to me!

        • Re: (Score:3, Informative)

          by kestasjk (933987)
          If your software has to take advantage of both chips it probably won't be optimized for either, and it's also a lot of extra unnecessary work.

          • Re:Did Apple make a mistake? (Score:5, Insightful)

            by Nasarius (593729) on Sunday May 20 2007, @10:11PM (#19204063)
            How much software is "optimized" for a specific architecture, beyond what the compiler does? How much "unnecessary work" is there, beyond what has already been done in the creation of universal binaries? It's extra work for Apple, but essentially none for the vast majority of application developers.

              • Re:Did Apple make a mistake? (Score:4, Informative)

                by fritsd (924429) on Monday May 21 2007, @01:25AM (#19205381) Journal
                It's tricky but not as hard as you make it sound. Look at Debian [debian.org] if you don't believe me: this picture shows what percentage of the programs is compiled for each architecture: stats.png [debian.org] it's usually over 95%. This includes little- and big-endian (mips, mipsel), 32-bit and 64-bit (x86_64), and weirder (s390). Also note the x-axis on the picture runs from the year 2001 :-) And yes, I know, compiled doesn't mean it actually also works :-)

                As to why Adobe can't be bothered to create a working flash player for (at least) 64-bit AMD64: I have no idea; we can't see the source so we can't see how difficult it would be to port it.

        • Re:Did Apple make a mistake? (Score:5, Insightful)

          by suv4x4 (956391) on Sunday May 20 2007, @10:09PM (#19204053)
          why Apple doesn't just permanently keep its lineup as a mix of PPC and x86, picking whichever chip suits the particular machine they're designing at the time? Power6 Xserves along side Core 2 laptops... it sounds good to me!

          As a software developer why you should work twice more (OSX intel / OSX ppc) to produce a piece of software that will work on roughly 2-3 % of the desktop computers out there?

          If Apple would keep randomly altering their hardware and require compatibility with a range of completely different architectures, in the end it'll completely alienate the developers. As Microsoft knows very well, developers, developers, developers are you best asset in this fight.

          Furthermore, no, being Universal binary is not a requirement, and I know few companies which release only Intel versions of their Mac software (example: Adobe's Soundbooth)

          • Re:Did Apple make a mistake? (Score:5, Insightful)

            by mrchaotica (681592) * on Sunday May 20 2007, @10:38PM (#19204261)

            As a software developer why you should work twice more...

            By "work twice more" you mean "check an option box in XCode," right?

            If Apple would keep randomly altering their hardware and require compatibility with a range of completely different architectures, in the end it'll completely alienate the developers.

            Yeah, just like how the wide range of different architectures most UNIX software runs on alienates developers...

            ...oh, wait.

          • Re:Did Apple make a mistake? (Score:5, Informative)

            by jcr (53032) <jcr@ma[ ]om ['c.c' in gap]> on Sunday May 20 2007, @10:40PM (#19204275) Journal
            why you should work twice more

            Building universal isn't twice the work. Most apps don't have any intrinsic byte-order dependencies, and very few people ever wrote CPU-specific code that depended on Alitvec (for example).

            -jcr

            • Re: (Score:3, Informative)

              by suv4x4 (956391)
              Building universal isn't twice the work. Most apps don't have any intrinsic byte-order dependencies, and very few people ever wrote CPU-specific code that depended on Alitvec (for example).

              You can tell that to the Flash developers who worked their ass off to deliver the Intel version of Mac Flash quickly.

              That little player has loads of ASM and SIMD instructions to be able to pull off what it does in this size and this speed.

              Also you're not accurate about Altivec, multimedia apps like Photoshop make very goo
              • Dude, all work's been done.

                Either, you're doing a mac specific app and use the Accelerate.framework which handles conversion to SSE3 or Altivec depending on the platform... ...or, you'll just pull in all the SSE work you did from the Windows Flash runtime since it's the same chip and these are all not OS dependent.

                Same thing for Photoshop. The plugin architecture makes it hella easy since they should have started with plugins for all the heavy stuff anyways. Recycling! It's not just for cans.

                • Re: (Score:3, Insightful)

                  by suv4x4 (956391)
                  you'll just pull in all the SSE work you did from the Windows Flash runtime since it's the same chip and these are all not OS dependent.

                  The argument was about keeping PPC or not. So how do you pull that from SSE on Windows Flash?

                  Furthermore, if you're a startup, who writes version 1 of a software, where do you "pull" this from?

                  The accelerate framework is a toy, for serious work, you need to code it manually.
                  • What do you need that Accelerate.framework doesn't provide?

                    Have you filed a feature request? http://developer.apple.com/bugreporter/ [apple.com]
                    You can use a free developer connection account to do so. If it's a feature that could be useful to multiple developers, there's a decent chance it will be added.
            • I think that you basically mentioned the only real place where there's a market for PPC: on servers. Although I've always been a big fan of the Power architecture (I have a dual-G5 spaceheater sitting under my desk that I'm writing this on, right now), I don't think that offering G5 PowerMacs along side Intel PowerMacs would really do anything besides confuse customers and potentially make the platform less appealing for developers who don't realize how easy Universal code is to produce. So I think that's a non-starter.

              However, keeping OS X Server (which under the hood really isn't that different from regular old OS X, but it's marketed as a totally different product) Universal, and producing PPC XServes in addition to Intel boxes, might not be a bad idea. PPC XServes have always had a fair bit of popularity in the HPC and scientific-computing segments over x86, and for servers, a lot of the software in use is OSS anyway and is architecture-agnostic by design. So they wouldn't really be confusing any developers there -- most of the software that runs on OS X Server is either supplied by Apple, or is OSS, or (in the case of custom HPC code) may have been written/optimized specifically for Power/Altivec in the past already, so they'd be saving their customers work by offering a PPC product.

              I think there could be a lot to gain by keeping a PPC model around. They might not even have to do too much hardware design; if they didn't burn too many bridges with IBM on the way out, they could probably use one of IBM's Power-based blade-server boards in a 1U case...particularly with the way Cell hasn't been selling, IBM would probably be happy for the microprocessor sales.

              • Re:OS X Server = PPC/Intel, OS X = Intel (Score:5, Interesting)

                by Amiga Trombone (592952) on Sunday May 20 2007, @11:56PM (#19204813)
                I think that you basically mentioned the only real place where there's a market for PPC: on servers. Although I've always been a big fan of the Power architecture (I have a dual-G5 spaceheater sitting under my desk that I'm writing this on, right now), I don't think that offering G5 PowerMacs along side Intel PowerMacs would really do anything besides confuse customers and potentially make the platform less appealing for developers who don't realize how easy Universal code is to produce. So I think that's a non-starter.

                I'd agree with that assessment. Also, consider that desktop/laptop CPU's have different requirements than server CPU's. One of the reasons Apple dumped PPC was that IBM wasn't earning enough on chips optimized for desktops to invest in the necessary R&D to keep them competitive with x86.

                That is not an issue with servers, however, Power6 is already optimized for that purpose. Apple could probably offer a very attractive XServe indeed based on that chip. It would give them an offering that would outperform anything based on x86, making OS X a more attractive and versatile platform in general. I'd like to see them go for it.

        • Re: (Score:3, Interesting)

          by solios (53048)
          UBs are the new Fat Binaries - Apple dropped the 68k as soon as they were able but Fat Binary soft that could run on both 68k and PPC was around for quite a long while thanks to the big install base of 68ks. Now history's repeating itself - the only reason we have UBs is because of the hugehugehuge (proportionately speaking) PPC install base.

          Apple has no reason to go back to the PPC. The profit margin on intel kit is much higher..... and if you don't think it's about profit, ask yourself why all of the lo

              • Re:Did Apple make a mistake? (Score:4, Insightful)

                by mrchaotica (681592) * on Monday May 21 2007, @12:21AM (#19205007)

                Intel graphics are fine if all you run is Office...

                I've got Intel graphics on my X60, and I'm in the middle of installing a bunch of 3D games in Linux (Tremulous, FlightGear, Scorched3D, Neverball...); I anticipate that it'll run them just fine. It also works really well with Compiz/Beryl. Personally, I think it's a lot better than having an Nvidia or ATI chip, and not having 3D support at all.

                • Re: (Score:3, Interesting)

                  by solios (53048)
                  Try running, say..... Maya [autodesk.com]. Or Motion [apple.com].

                  Ultimately, the point I'm aiming at is that paying premium prices for bargain basement video really chafes my ass - if I'm going to lay down for kit that's twice the price of an equivalently powered wintel box, I'd like some name brand video and user access to all of the system memory.

        • Re: (Score:3, Informative)

          by Jeff DeMaagd (2015)
          POWER isn't PowerPC. I don't think it's worth keeping two architectures in parallel like that for long term, not with Apple's current volume. Apple doesn't make that many Xserves last I heard, something like tens of thousands per year, when the other server companies exceed that by as much as 20x. IIRC, HP was selling 200k 1U servers when Apple sold 12k 1U servers. I don't know what IBM's numbers are, but POWER-based workstations and servers were a lot more expensive than Apple's stuff.

        • Re:Did Apple make a mistake? (Score:4, Insightful)

          by Lobster Cowboy (605052) on Monday May 21 2007, @12:13AM (#19204951)
          Because one of Apple's new selling points is the ability for all new Macs to run Windows as well as OS X. Can you imagine the nightmare Apple would have if they mixed processors?
      • But Apple didn't have to switch - they could've gone intel for laptops & waited for power6 for high end.

        • Re: (Score:3, Insightful)

          by jcr (53032)
          But Apple didn't have to switch

          Remember the Sun 386i? Nobody believed it had a future, and that was a self-fulfilling prophecy. If Apple hadn't committed to a switch, then a lot of developers wouldn't have bothered to build their apps for Intel, since they could just run under Rosetta.

          -jcr

    • Re:Did Apple make a mistake? (Score:4, Insightful)

      by stoneymonster (668767) on Sunday May 20 2007, @09:24PM (#19203657) Homepage
      Huge increase in mac sales since the intel switch? Massive profits? Stock well over $100? Yeah they made a mistake. Look, sometimes business decisions are just that: business, regardless of whether they're the most exciting decision from a technical or geek standpoint.

      • Re: (Score:3, Informative)

        by Anonymous Coward
        The stock price is more related to the iPhone and phenomenal iPod sales.
    • It's interesting to see PowerPC on the other side of the megahertz myth for a change.

      Also, I bet Power6 would work great in minis and MacBooks.

      • watts (Score:3, Informative)

        by DreadSpoon (653424)
        The Power6 uses "under 100 watts in performance sensitive applications."

        WAAAY too much for a notebook or a mini.
              • You provide the wrong link, this [apple.com] is the proper one. 110 W, for the complete machine, and that's AC (so even if the CPU was the only DC component in the machine, it would end up consuming more from the mains).

                For comparison, I think the Core Duo TDP in that machine is something like 30 W, maybe a bit more.

            • Re: (Score:3, Informative)

              by c_forq (924234)
              The only Thinkpad in my requirements and price-range is $250 more than a Macbook, with a slower processor, smaller hard drive, and doesn't include bluetooth. (The X60 is $1,251.75 at sale price, and the Macbook is $1,000 after student discount).

    • Re:Did Apple make a mistake? (Score:5, Insightful)

      by l0ungeb0y (442022) on Sunday May 20 2007, @09:32PM (#19203753) Homepage Journal
      "Did Apple make a mistake?"

      No. Not at all.

      The Power series was the high end server class, meant for big iron.
      The PowerPC series was the vastly scaled down little brother intended for the desktop class.

      IBM wasn't all the interested in making chips for Apple.
      And who can blame them? Lower profit margins and less units sold.

      Intel is a much better match for Apple, which is a consumer grade CPU manufacturer. And since the switch, Apple has not had the embarrasment of lower performing CPUs and long waits on CPU upgrades that IBM and Freescale saddled Apple with.

      If Apple had stayed with IBM, they would have been pushed to the Cell processor. And that would be a bad PR move, running on the same CPU as your game consol runs on. And there would of course be no gaurantee switching to that processor family would result in better product cycles from IBM.

      Apple made the right choice, The relationship with IBM was no longer viable.

      • Re: (Score:3, Insightful)

        by LWATCDR (28044)
        Unfortunately true. The new Power series could work in a desktop or workstation but it will be expensive. I am sure that with enough money IBM could make a Power6 that was better than the Core 2... Except they wouldn't make a profit on it. As much as I hate loosing yet another better then the X86 ISA from the desktop Apple did the right thing.

    • Re: (Score:3, Insightful)

      by Short Circuit (52384) *
      I'll wait until the specs come out Tuesday* before I decide if the Power6 is interesting. Sure, it's got GHz. But how many ALUs does that cover? FPUs? What about its SIMD instruction capabilities? How long is the pipeline? Is it insanely long a la NetBurst?

      Not to mention that IBM didn't seem to be putting any resources at all into a low-power verion of the POWER5; What makes you think they would for the POWER6? Without a low-power chip, Apple would have a hard time making laptops with a decent battery

    • Power isn't PPC (Score:5, Informative)

      by Sycraft-fu (314770) on Sunday May 20 2007, @09:36PM (#19203783)
      Despite the similar name, and somewhat related architecture, IBM's Power line are not PPC chips and aren't suited for desktop use. That's not to say that some technologies from them can't go in to other chips, but drooling over what is essentially a minicomputer/mainframe chip is silly.

      The reason Apple switched is because, despite all the hype, Intel continues to make really fast chips for a good price. When Apple was on PPC I saw never ending arguments as to how much faster the chips were. All those never seemed to pan out in actual operation. Why that's the case isn't important from Apple's standpoint, they just want fast chips for low cost.

      I suppose if you want to long for the days of Altivec and talking about tech stuff you don't fully understand, that's great, however Apple has to be a bit more pragmatic and realise that while Altivec might sound cooler than SSE3, SSE3 is an API for a damn fast vector unit and that's all that really matters. Most people don't care about contrived benchmarks, they care about the wall clock benchmark, meaning how fast does the system do what they want, and further how cheap can they get that system for.

      • Re: (Score:3, Interesting)

        by DreadSpoon (653424)
        "IBM's Power line are not PPC chips and aren't suited for desktop use"

        Yes, yes they are PPC chips. In terms of core instruction set, they're the same. The PPC970 that Apple used for a short while were derived directly from the Power design, as I recall.

        The PC in PowerPC doesn't mean "Personal Computer." It means "Performance Computing." PPC is an instruction set, and Power is an IBM brand/product name. Many companies make PPC chips besides IBM, and the majority of those chips are embedded chips not at

        • Re: (Score:3, Informative)

          by Usquebaugh (230216)
          No they are not!

          Power != PowerPC

          PowerPC has a subset of the Power op codes.

          PowerPC is seen as a embedded/desktop platform
          Power is used in AS/400 and RS/6000 boxes

          Power code does not run on the PowerPC, lack of certain op codes

          I'm sure a better explanation is available on wiki, but they are not the same.

        • Re: (Score:3, Interesting)

          by 1lus10n (586635)
          I'm still waiting on the 128 way Intel/AMD ... or anything greater than 16 way that can keep up with the RISC, Sparc or POWER based systems.

          When your running apache it doesnt make a difference. If you can get onto Oracle RAC then it will matter less, but for right now there is still a ton of business to be done on the high end of things. Sun's T1 chip is also a metric fuckton better and running web apps. Especially java. 32 threads, low power and so on.

          x86 has always been designed for mass use, and

    • Re: (Score:2, Informative)

      by Anonymous Coward
      POWERn != PowerPC

      They (mostly) share a common ISA but the chips themselves have always been quite different.

    • Re: (Score:3, Interesting)

      by Cadallin (863437)
      I'm going to give the very uncommon answer: Yes. Definitely.

      Why? If they were going to switch to x86, they waited way too long to do it. By the time the first Intel Macbooks shipped, IBM had had low power G5's available for months. These could have absolutely been user for a Powerbook G5. The desktops, of course would have been shipping POWER5 parts, what would have been the G6 (By the time these POWER6 machines made into Macs, they would have been the G7). The correct solution to many of the oth

    • Re: (Score:3, Informative)

      by DurendalMac (736637)
      Nope. This is the POWER6, not a PowerPC. IBM made the G5 (aka 970) as a derivative of the POWER4. IBM told Apple that they didn't want to make a derivative of the POWER5, so they were SOL on an upgrade path. This is not the kind of processor you would EVER see outside of a top-end workstation, server, or mainframe. It's not something Apple would have used.
      • As the transition sees developers pushing out binaries for both chips, I don't see a downside of Apple straddling the fence and using both types of chips. (The people who use paralells or virtualization will know what their software runs on....)

        They could also continue using that as leverage against intel or ibm when it comes time to price certain chips.

        (Assuming the intel contract only forbade them getting other x86 CPU suppliers, not CPU suppliers in general.)

          • Re: (Score:3, Informative)

            by TheRaven64 (641858)
            OS X (specifically XNU) uses a 4GB/4GB memory split, meaning that the kernel and applications both have separate 4GB address spaces. This means that any 32-bit application can access up to 4GB of memory. Most operating systems use a 1GB/3GB split, which limits the amount of space available for the application to 3GBs, but makes system calls cheaper since they don't need a full context switch (the kernel's address space is mapped into every process's address space, but marked as no-access for unprivilieged

  • Yeah, but... (Score:3, Funny)

    by wcspxyx (120207) on Sunday May 20 2007, @09:25PM (#19203679)
    ...does it run Vista?

  • an IBM p570 server showed an average response time of .625 seconds when handling requests from 2,100 users. That compares to a p570 with 2.2GHz Power5+ chips that handled .983 requests per second for 2,000 users.
    not bad, they were about due for a speed increase from those standard 3Ghz CPUs.

  • That's old school (Score:3, Funny)

    by Burdell (228580) on Sunday May 20 2007, @09:45PM (#19203871)
    I had a 4.77MHz IBM years ago. Oh wait, you said G, not M.

  • Why Apple/Intel was the right move (Score:3, Interesting)

    by zerofoo (262795) on Monday May 21 2007, @08:47AM (#19208003)
    The shop I work in right now is a mix of Dell and Apple hardware. We now buy all our Desktop machines from Apple - why?

    Intel CPUs.

    We can now run Windows and Mac OS on the exact same hardware. Dell has lost all our desktop business as the result of Apple's move to Intel. One hardware platform is very nice from a purchasing and management perspective.

    I'm sure we aren't the only shop with that strategy - and that's why Apple's conversion to x86 was a good decision.

    -ted

    • Re:Ho hum (Score:5, Funny)

      by SeaFox (739806) on Sunday May 20 2007, @10:14PM (#19204089)
      Because we can do no Moore at the moment.

      • Re: (Score:3, Insightful)

        by ricree (969643)
        I know that was just a joke, but I would like to point out that Moore's law is still continuing just fine at the moment. Most consumer processor designers have decided that instead of using the extra transistor density to increase speed, to use it for all these multi core chips that have been produced the past couple years.

    • Re: (Score:3, Insightful)

      by Icarus1919 (802533)
      Because most of us who have desktop pcs and stuff aren't running weather simulations or fragging at the highest possible FPS. My desktop runs at a little under 3ghz and it's just fine for me thank you. Most other people I know don't need that much power either.

All trademarks and copyrights on this page are owned by their respective owners. Comments are owned by the Poster. The Rest © 1997-2009 Geeknet, Inc.