Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
IBM Hardware

More on the PowerPC 970 386

functor writes "Ars Technica's Jon Stokes has a treatise up covering the microarchitecture of the high-performance 64-bit PowerPC 970 microprocessor, due to be released by the end of the year, that goes over in detail how this chip is put together, and how we can expect it to perform. This is the follow-up to Stokes' article detailing the PPC 970's design philosophy. 'It appears to hold quite a bit of promise in bolstering Apple's currently almost obsolescent product line, and it appears to have been designed explictly to fulfil Apple's requirements. To say the least, the second half of this year looks to be pretty interesting as Apple's product line promises to become competitive performance-wise with IA-32 and x86-64-based PCs again.''
This discussion has been archived. No new comments can be posted.

More on the PowerPC 970

Comments Filter:
  • Inaccuracy, Part 1 (Score:5, Interesting)

    by 11223 ( 201561 ) on Wednesday May 14, 2003 @08:10AM (#5953424)
    Unfortunately, the vector performance of the G4e has been consistently bottlenecked by Apple's lackluster motherboard and chipset designs--specifically the anemic frontside bus and memory subsystems that Apple has saddled the PowerMac line with.

    This implies that the decision of how much bus bandwidth to give the G4e was up to Apple - which it was not. Motorola designed the processor (for Cisco, depending on who you believe), and Apple made do with the anemic MaxBus at 133mhz that they got from Motorola.

    Apple'd be putting DDR400 on the G4 right now if they could. None of this (well, except the decision to go Moto) was their fault.

    • As far as I know, none of the other IA-32 motherboard manufacturers are making *dual* processor boards with 166 MHz (333 DDR) bus for use with 333 MHz DDR memory. Sure, there are plenty of IA-32 single cpu boards running bus and memory synchonously at DDR333, a few at DDR400, and even a few running DDR266 bus/333 memory asynchronously.

      So my point is that Apple's offering, though somewhat expensive, is unique.
      • by functor ( 31042 ) on Wednesday May 14, 2003 @09:06AM (#5953778) Homepage
        No, but IA-32 motherboard manufacturers go a good number of steps further. ;) I recommend that you investigate Intel's Placer (E7505) chipset [intel.com] and motherboards based on it (several of Supermicro's offerings, as well as offerings from Tyan and other manufacturers, e.g. the Iwill DPL533 [iwillusa.com] and DP533 [iwillusa.com]. These motherboards support 133 MHz QDR system buses (coming to 533 million transfers a second), matched (quite well) with two channels of PC2100 DDR SDRAM (resulting in 4.267 GB/s of memory bandwidth that is actually utilizable by the processors, since the memory bandwidth matches the system bus bandwidth, unlike Apple's offering, which is bottlenecked by the system bus at just 1.333 GB/s, whether you have one processor or two). (And I'm certain that 200 MHz QDR Xeon chipsets are not far off in the future, since Intel in general appears to be headed in that direction.)
    • In a way that is a sort of a cop out. Apple could have invested in better chipset technology. A quad-pumped 133mhz FSB has already been done by Intel. What was to stop Apple from doing the same? Lack of a good chipset, that's all.
      • by 11223 ( 201561 ) on Wednesday May 14, 2003 @08:56AM (#5953708)
        You're completely wrong. The maximum speed of the FSB and whether it supports DDR (or QDR) is determined by the processor, not by the chipset. For the G4e, the maximum known speed at which MaxBus can operate is 167MHz - precisely what Apple uses.

        They can't make the FSB DDR or QDR without appropriate support from the processor, and that's exactly what they haven't been getting from Moto.

    • by Hannibal_Ars ( 227413 ) on Wednesday May 14, 2003 @11:25AM (#5954993) Homepage
      I should've been more clear on this point. My real problem with the current G4e situation, aside from the 167 SDR FSB, is the fact that it's a shared bus topology, which is just ridiculous. To my knowledge, there's nothing stopping Apple from putting out a chipset that gives each G4e a dedicated FSB (even if it's still 167MHz SDR) to the chipset.

      As far as the low MHz and SDR situation, I've also never been totally convinced that Apple wasn't partially to blame for this either, unless they just have zero clout with Moto SPS.
  • by Space Coyote ( 413320 ) on Wednesday May 14, 2003 @08:11AM (#5953430) Homepage
    Why this had to be posted the morning before my presentation to my supervizor is a clear indication that the universe is against me.

    Time to hide my network cable until the end of the day.
  • by Anonymous Coward on Wednesday May 14, 2003 @08:13AM (#5953441)
    I sold my G4 tower some time ago becuase it was not fast enough to compete with my winders boxes. I'll jump back on the Apple platform when the 970 ships, assuming it's all that. Lets just hope the entry level unit is ( for Apple ) somewhat affordable.

    The current pro line of G4 is a joke. They cant come out with 970 computers fast enough.
    • by TheRaven64 ( 641858 ) on Wednesday May 14, 2003 @10:20AM (#5954400) Journal
      I sold my G4 tower some time ago becuase it was not fast enough to compete with my winders boxes.

      I'm intrigued. What do you do that makes a 1.25GHz G4 feel slow? I'm still using a 1.33GHz Athlon and it feels quite fast. I keeps thinking about upgrading the CPU, but really can't see the point. I rarely use more than 20% of it as it is...

      • by Doc Hopper ( 59070 ) on Wednesday May 14, 2003 @10:39AM (#5954566) Homepage Journal
        I concur. The only times I feel the age of my 933MHz Pentium III is on certain levels in Unreal Tournament, and when I'm compiling a new kernel and realize it takes three times longer than the newest benchmarks.

        The 970 looks exciting, but I personally am anticipating the release only for the dramatic price reduction it should bring in the older Macs. At that point, I'll go get that Powerbook -- once I can pick up a decent one for less than $2,000.
      • I'm intrigued. What do you do that makes a 1.25GHz G4 feel slow? I'm still using a 1.33GHz Athlon and it feels quite fast. I keeps thinking about upgrading the CPU, but really can't see the point. I rarely use more than 20% of it as it is...

        One thing to remember about Apple's machines is that while the G4 processor, if the code is really tuned can hold it's own with processors at a much higher clock speed on OSX not very much is very tuned and the overhead associated with running OSX demands a lot mo
    • by King_TJ ( 85913 ) on Wednesday May 14, 2003 @12:53PM (#5955840) Journal
      I guess I don't quite follow.... As someone else already asked, what exactly are you trying to accomplish with your machine(s)?

      I've been using strictly PCs for 5 years or so now, after a brief stint with a "Performa" Mac mini-tower that didn't turn out too well.

      My "high end" PC system is a Pentium 4, 1.8Ghz tower with Promise EIDE RAID and a GeForce 4 video board.

      I'm pretty happy with it, but I really wanted a good system to run OS X and some of Apple's incredibly well-done video editing packages (Final Cut, iDVD, etc.). I just broke down and bought a dual-processor G4 1.42Ghz tower. I certainly don't feel it's "slow" at all! I'd say it performs at least on par with my P4 system, if not a little faster at certain tasks. It boots into OS X a lot more quickly than the P4 boots into Windows XP, for one thing.

      Sure, the 970 processor will be great -- but the people complaining that the current PowerMacs are "horribly underpowered" must be "benchmark junkies", worried about having the best stats for the sake of stats (bragging rights?).

      Like I say, I consider myself very much a "power user", and for a long time, I didn't think Apple really had the "price vs. performance" in the right place on the curve. But with their recent price drops, plus "speed bumps" to their G4 offerings - I think they still have a very competitive setup to tide them over until the 970 is done.

      At the end of the day, you don't plunk down $2000-3400 for a "pro" Mac G4 or PowerBook because you're worried about having the "most Ghz". You do so because it offers an OS and specialized applications you can't get in the PC world. (These days, you might also do so to avoid the Microsoft licensing nightmares. A "family pack" of OS X lets you load it on any 5 systems of your choice for a price not much more than 1 single copy of Windows XP Pro, for example.)
  • It is competitive ! (Score:5, Interesting)

    by Fefe ( 6964 ) on Wednesday May 14, 2003 @08:15AM (#5953448) Homepage
    Who knows whether it will still be competitive in several months when they actually want to offer it.

    On the other hand Apple users won't have much of a choice, and neither has Apple.
    • by mirko ( 198274 )
      Actually, since Jobs got back to Apple, he's not especially disappointed Apple users : there was the iMac, the iBook, the iWhatever, the Airport, the G4...
      Each time there was a leap forward so I guess this will give the concurrence some nightmares.
      But you are right, until then, Apple took huge risks.
      It's just a good think someone did, otherwise the market would still offer prehistoric beeping-XT.
    • by dasmegabyte ( 267018 ) <das@OHNOWHATSTHISdasmegabyte.org> on Wednesday May 14, 2003 @10:39AM (#5954561) Homepage Journal
      Sorry, that's bullshit. Apple has whatever choice apple wants to make, and they've had talks with other chipmakers including AMD. If the 970 fell through, they'd go with something else. If they had to, they could switch to an x86 architecture without batting an eye. It worked pretty well for SGI.

      Not that it would necessarily be in their best interest. Apple, as a company, represents an ALTERNATIVE, and therefore they try to maintain alternative hardware choices: SCSI over IDE, ATI over NVIDIA, USB over PCI, the one button & metakey paradigm, Flat Panel over CRT -- and the big one, RISC over CISC. Some of these choices have panned out great. Some have flopped miserably...yet despite the doomsayers, Apple is still afloat after 30 years of "forcing" people to buy "crazy" proprietary gear. With "only" 3% of the market, but it's a huge freakin' market, and their margins are gigantic. Part of the reason for this huge margin is that they are the only hardware player, and are therefore able to name their own prices.

      As for Apple users having no choices, that's also crap. We always have choices: buy Apple, or buy something else. Keep the current machine and upgrade it, or buy a new one. And Apple will still make the old chips for a long time after the 970 has started smokin' competitors...at least, that's what they did every other time they made a generation jump in processors.

      Apple users vote with their dollars and so it's in Apple's interest to do whatever people are most receptive to, which is what they as a company seem to be best at anyway. After all, they got me to spend $538.72 on an mp3 player. This whole "no choice" thing is BS -- you choose the platform, not the hardware. That's the Apple way.
      • "If they had to, they could switch to an x86 architecture without batting an eye. It worked pretty well for SGI."

        No, they couldn't. Every app would need to be rewritten, right on the heels of the 9 to X transition that isn't even finished yet. Switching to x86 is a complex nightmare that may be Apple's doomsday plan, but it is far from simple.
      • SCSI over IDE

        Apple went pretty much all IDE about five years ago.

        USB over PCI

        Mac's have had PCI slots since the first PowerPC based units became available. In fact, back in those days many PC's still had VLB, and only Pentiums had PCI slots. Further, since they were 100% PCI there was no bottleneck due to legacy support (ISA) Also, USB's importance was such that it replaced SCSI for external, high speed devices...
  • competitive, sure... (Score:5, Interesting)

    by Anonymous Coward on Wednesday May 14, 2003 @08:17AM (#5953461)
    The PPC 970 will not really make the Macintosh competitive with modern PC's. It will make it competitive with PC's from the beginning of this year, which are not the fastest available any more, and will be even slower when compared to the machines that are available when the PPC 970 ships, which is the very earliest that Apple machines based on it can ship. It will however go a long way to catching up, and take off a lot of the pressure caused by the abominable performance of today's dual processor G4 machines when compared to even inexpensive PC's.

    The other unkown in this is the price. PPC 970 based Apple computers may be significantly more expensive. Motorola loses hundreds of millions of dollars each year on their semiconductor business, and IBM does as well. Still, IBM may want to look at Apple and the PPC 970 as a PROFIT center, rather than a LOSS center, like Motorola does with Apple and the G4.

    The PPC 970 is great news for Apple, but it is still a bone thrown to them while the x86 PC is feasting on the meat of the Intel and AMD processors.
    • by Trurl's Machine ( 651488 ) on Wednesday May 14, 2003 @08:44AM (#5953633) Journal
      The PPC 970 will not really make the Macintosh competitive with modern PC's. It will make it competitive with PC's from the beginning of this year, which are not the fastest available any more, and will be even slower when compared to the machines that are available when the PPC 970 ships

      "Being competitive" does not equal to 'having more computing power". Look how small is this thingy's power consumption! I guess when 970 ships, we will have similar situation as we have right now. x86 machines will consume enormous amount of power and dissipate enormous amount of heat, what usually results in this nice "quadruple augmented turbofan" sound that accompanies most PC desktops or "not enough battery life even to watch a full DVD" laptops. Not to mention that if you actually put this laptop on the top of your lap, you might get your testicles hard boiled.

      And Apple will launch yet another series of slower but cool machines - both in terms of look and heat dissipation. Which actually is pretty much what we have already.
      • ...and as the processors become faster and faster, speed becomes less of an issue.

        When you can compile the kernel or compress a movie in 2 mins, who will really care that it can be done in 1?

        The only reason for getting the 4 gig over the 2 would be to have the ability to play your games at the cinematic detail level.
        • When you can compile the kernel or compress a movie in 2 mins, who will really care that it can be done in 1?

          Damn, I wish I had my copy of "Moon Is A Harsh Mistress" right now with me. I remember that there was a quote almost exactly the same as yours (of course, Heinlein did not mention specifically "compressing movies" or "compiling kernels", just something about calculations and miliseconds). Who says science-fiction predictions for XXI century are not accurate? Robert A. Heinlein predicted the spirit
      • by be-fan ( 61476 )
        The PPC 970's power consumption at 1.8GHz is 42 watts. If it debuts at over 2GHz like many people predict, it will easily be over 50 watts. At that point, the extra 20 or so watts a P4 consumes hardly seems like a big loss. Also, it is entirely possible to make whisper quiet 3GHz P4 monsters. The newest Dells don't even put a CPU fan on the heatsink. They have a huge heatsink, and cover it with a closed heat tunnel. At the end of the tunnel, they attach a very large, but very slow, and thus very quiet, runn
    • by 11223 ( 201561 ) on Wednesday May 14, 2003 @08:59AM (#5953736)
      The PPC 970 is great news for Apple, but it is still a bone thrown to them while the x86 PC is feasting on the meat of the Intel and AMD processors.

      As Nethack would say, "Ugh! This meat is tainted!"

      The 970 is fundamentally a 64-bit processor, and its performance must be evaluated in that context. The fact that the 970 will pull off amazing speed in the 32-bit arena only shows how well-designed this processor is.

      Keep in mind, the Hammer is only shipping at 1.4, 1.6, and 1.8 GHz - the same speeds the 970 is targeted at. And the 970 has the advantage of an ISA that was designed from the beginning to do 32 and 64 bit addressing, versus one that's a 64-bit extension of a 32-bit extension of a 16-bit micro with full compatibility to an 8-bit redesign of a 4-bit processor.

      • by Anonymous Coward on Wednesday May 14, 2003 @09:32AM (#5953981)
        64-bit extension of a 32-bit extension of a 16-bit micro with full compatibility to an 8-bit redesign of a 4-bit processor

        ...used by a 2 bit company that can't stand 1 bit of competition

        (for completeness)
  • Dual FPUs! (Score:5, Interesting)

    by EvilTwinSkippy ( 112490 ) <yoda AT etoyoc DOT com> on Wednesday May 14, 2003 @08:31AM (#5953551) Homepage Journal
    Reading through the article, its nice to see some real design going into a processor. Looking through Intel's last few chips, they've been upping ther clock speed and packing in more cache.

    Yeah, yeah, they are hog-tied because you can't easily re-compile the entire windows platform to use new instruction sets. Linux users, of course, don't have this problem (muhahahah).

    Did anyone else catch the bit on the twin FPU's? I'm just imagining what this thing is going to do with vector operations and frequency transforms.

    For most of you non-engineers:

    Most 3d vector operations are affine tranformations. Using a 4x4 array of floating point numbers you can translate, rotate, and scale. Works beautifully, but it's a lot of calculations.

    The Fast Fourier Transform (FFT) is used a lot in signal processing. It's a floating point monster.

    • Most 3d vector operations are affine tranformations.

      Most 2-dimensional transformations can be done this way too. Apple's Quartz subsystem uses matrix transformations just about everywhere it can get away with.

    • Re:Dual FPUs! (Score:4, Interesting)

      by functor ( 31042 ) on Wednesday May 14, 2003 @08:53AM (#5953688) Homepage
      Did anyone else catch the bit on the twin FPU's? I'm just imagining what this thing is going to do with vector operations and frequency transforms. For most of you non-engineers:
      • Most 3d vector operations are affine tranformations. Using a 4x4 array of floating point numbers you can translate, rotate, and scale. Works beautifully, but it's a lot of calculations.
      • The Fast Fourier Transform (FFT) is used a lot in signal processing. It's a floating point monster.
      I would imagine that the use of AltiVec/VMX single-instruction multiple-data instructions would be somewhat more effective in doing vector and matrix floating-point computations than scalar floating-point operations as provided by the dual FPUs -- especially on these smaller (4x4, 8x8) matrices. While, in comparison to the MPC 7450, the PPC 970 may have an inferior VMX unit (but with much more bandwidth and a higher clockspeed available), the SIMD instructions allow for more data to be processed per clock cycle than the scalar FPUs can, so I imagine that things like FFTs and vector arithmetic and transformations are better-suited to the use of AltiVec/VMX instructions -- perhaps even hand-tuned to get the best instruction scheduling and highest instructions-per-clock count (and hence computational throughput).
    • Re:Dual FPUs! (Score:2, Informative)

      by tortap-0 ( 306464 )
      While dual FPUs are useful this is what SIMD operations like Intels SSE and Motorolas Altivec are used for. The dual FPUs might be good for doing other work but DSP filters and Photoshop filters will use SIMD operations wherever they can.

      That is very bandwidth intensive work, moving alot of floating point numbers from memory, and this is where the 970 will be superior to the G4e. But this is also the strong point of the Intel P4 running at super high frequencies. The AMD Athlon 64 will clock for clock be c
  • Hehe (Score:5, Funny)

    by A Proud American ( 657806 ) on Wednesday May 14, 2003 @08:31AM (#5953554)
    This reminds me of an old joke one of my professors told once. Hope you don't mind that I share:

    Q: Why might IBM change the name of their 970 chip?

    A: Because they added 620 to the IBM 350 and got 969.999983605.
    • Re:Hehe (Score:3, Informative)

      by ZorinLynx ( 31751 )
      Awww jeez, that's just a permutation of the old joke:

      "Why did Intel rename the 586 to Pentium?"

      "Because they added 100 to 486 and got 585.999999878787775555"

  • by Dark Paladin ( 116525 ) * <jhummel.johnhummel@net> on Wednesday May 14, 2003 @08:33AM (#5953569) Homepage
    According to some rumor sites [macrumors.com], Apple may already have ordered several thousand of these chips for new machines to debute in middle of June.

    I'm not buying into it 100% myself, but as I don't plan on buying a new Powermac until next year (and turning my current one into either a Yellow Dog or OS X Server), I'm in no big rush.

    My expectations is that the Powerbook/iBook line won't be updated until next year, when IBM can get the power requirements down for the 970 or its successor.
  • Obsolete my a$$ (Score:5, Interesting)

    by motorsabbath ( 243336 ) on Wednesday May 14, 2003 @08:53AM (#5953689) Homepage
    It appears to hold quite a bit of promise in bolstering Apple's currently almost obsolescent product line

    Ha! I have a 867 G4 at home and it still rocks. Apple's line is certainly nowhere near obsolete, they're very different boxes than PCs.

    You multi-gigahertz fruitcakes crack me up. 3 GHz is a waste of processor power and energy for at 80% of the people that use computers.

    • Many academics treasure their old Lisp machines. Their love for the boxes does not change the fact that the rest of the world no longer cares, and that Lisp machines are obsolete.

      Besides, I take that to mean that you'll never buy another computer unless your current one breaks? Because, an upgrade would be a waste, wouldn't it?

      Perhaps I might feel sympathetic if Jobs hadn't strutted around with a "ho hum, another day, another supercomputer" attitude.

      • Actually, I've got a pile of them. The G4 is sitting right next to an Athlon running Linux and FreeBSD. The Apple still rocks, and is a match for the 1.2 GHz Athlon and is not in need of an upgrade, although I may get a new video card for it.

        I never buy PCs, I put them together myself - the next full computer I buy will be a PowerBook - I'm sick of lugging my PC around to do multi-track recording.

        I just disagree with the line that Macs are obsolete. They're great boxes. Even an 867 MHz Mac is 10x the pro
        • ALL computers are obsolete as soon as they are bought. Its not meant to be a slight against Mac users, its simply a statement of fact.

          Lose the insecurity.
      • Re:Obsolete my a$$ (Score:3, Interesting)

        by Xugumad ( 39311 )

        Besides, I take that to mean that you'll never buy another computer unless your current one breaks? Because, an upgrade would be a waste, wouldn't it?

        Systems I use for games (consoles and PC) are likely to get upgraded every couple of years, but that's because games are still pushing them to their limits. Systems I use for work are likely to not be replaced for a lot longer, unless there's a very good reason (I am planning to move to Mac, so that's a replacement). Nothing I do with my work systems, that

    • Re:Obsolete my a$$ (Score:5, Insightful)

      by gunnk ( 463227 ) <{gunnk} {at} {mail.fpg.unc.edu}> on Wednesday May 14, 2003 @09:16AM (#5953844) Homepage
      I have a 2.4GHz Dell Optiplex GX260 with 1 GB of RAM here at work. Next to it I have a 450MHz G4 with 640MB RAM.

      I use the Dell for Linux development work and to run a couple of Windows-only programs (Netware Administrator), but for everything else I use my Mac (email, word processing, web browsing, spreadsheets, etc.).

      Far from obsolete, this old G4 with OS X still provides a much better work environment than my Optiplex. So what if the processor is slow by today's standards? I'm still MORE PRODUCTIVE on my Mac. Isn't productivity the best benchmark for how good a computer REALLY is?
    • Re:Obsolete my a$$ (Score:5, Insightful)

      by bobbozzo ( 622815 ) on Wednesday May 14, 2003 @09:39AM (#5954040)
      Apple's line is certainly nowhere near obsolete, they're very different boxes than PCs.

      Macs are often used in publishing...

      We use Macs (Dual G4), Sparcs, and PCs (running Linux) to do massive batch processing of EPS files, converting them to PDFs and GIFs or JPEGs. (using Acrobat Server and/or Ghostscript)

      The PCs are able to run these conversions several times faster.

      We are currently about to exceed the capacity of the current systems.

      Question: do we get an expensive 4-CPU Sparc, make a cluster of Macs, or do we get one moderately-priced dual-CPU PC server?

      That is what people mean when they talk about systems becoming obsolete; they can no longer keep up with what else is available or needed.

      That is also why Apple must keep up with the competition or die... even if MOST people don't need the power, SOME do.
      If they aren't going to compete on performance, then they must compete on features and/or PRICE.
      Features: It used to be that most good publishing or graphics software was only available on Macs; that is no longer the case. Now the only difference between a Mac and a PC running Windows is the GUI looks different.

      • I agree. My first post was applicable only to the end-user/personal computer experience. In a clustered or headless environment (where the interface is irrelevant) there's no reason to use anything but SMP Athlon/Pentium boxes. Note that SPARC just doesn't even enter into the equation anymore....

        Also, in the spirit of the original article, the 970 will really help Apple out in clustered/workhorse situations like yours. We use some of the 970's ancestors for clustered sim on AIX (has to be AIX/PPC) here at
  • by pchown ( 90777 ) on Wednesday May 14, 2003 @08:59AM (#5953739)
    Interesting, if you look at the pipeline design of the PowerPC [arstechnica.com] it is much closer to Intel than AMD [aceshardware.com]. The PowerPC pipeline has sixteen stages, the Pentium 4 twenty, and the Athlon ten.

    Presumably the P4 can reach higher clock speeds than the Athlon because there is less work to do at each pipeline stage. On the other hand a longer pipeline increases the probability of a stall, so the work done per clock cycle goes down.

    I'd speculate that the PowerPC ought, therefore, to be able to achieve clock rates approaching but not equalling the P4, since they are both comparatively "over-pipelined". At the same time, the PowerPC ought to deliver slightly more throughput per clock cycle because the pipeline is slightly shorter.

    Meanwhile, the Athlon will be running at a significantly lower clock rate, but delivering comparable throughput.
    • You may be right, but Athlon is MUCH older than P4 or 970.

      Perhaps you should look at Hammer if you want to make a comparison.
    • by Lebannen ( 626462 ) <(moc.nawori) (ta) (hsals)> on Wednesday May 14, 2003 @10:14AM (#5954343) Homepage
      As well as the depth of the pipeline, I believe the article also says you need to look at the width of the pipeline; it points out the G4+ is wide and shallow, the Pentium 4 is narrow and deep, and the 970 is wide and deep. You will therefore get bubbles in the 970's pipeline, but their effect is minimised and you're far less likely to get stalls.

      Combine this with the more intelligent branch prediction, out-of-order execution etc in the 970, and you're probably looking at a chip which is slightly less efficient clock-for-clock than the G4+, but more efficient than the Pentium 4.

      Integer performance wise, it looks like the 970 will be about equal to a Pentium 4 of 25-50% higher clockspeed; FPU-wise, and of course Altivec-wise, it looks like a monster. So; it probably won't outperform the current Pentium 4s at a lot of tasks, but will kick it about on other more specialised tasks, which is a big step over the G4+. We're not looking at a Pentium-crusher, but we are looking at something that will be vaguely competitive.

      Just gotta see how well it scales, after that, and whether 64-bit will mean anything for average tasks... and when it actually happens, of course.
  • How much will this help out apps like PhotoShop and AfterEffects, once they are re-compiled for the architecture?

    I've heard conflicting answers, one is that 64-bit will really shine with 3D apps but do little to help the performance of 2D number-crunching.

    Does this mean we'll see only nominal gains with Adobe's apps? Someone enlighten me.

    • Re:64-bit Adobe apps (Score:4, Interesting)

      by Halo1 ( 136547 ) on Wednesday May 14, 2003 @09:24AM (#5953910)
      Being 64 bit will not help 3D apps any better than 2D apps. The only real advantage of being 64 bit will be the ability to address more memory. This will be advantageous for any application (2D, 3D, database, ...) that can use so much memory. The reason is that most applications simply do not yet require many 64 bit integer operations, so the fact that a 32 bit processor executes those more slowly than a 64 bit one doesn't really matter.

      The 970 will be faster for most applications not because it is 64 bit, but mainly because it runs at a higher clock speed and has a much wider/faster memory interface. Some other architectural decisions (deep pipeline with aggressive optimization logic) will help somewhat as well, probably.

      • Re:64-bit Adobe apps (Score:3, Informative)

        by darc ( 532156 )
        Actually, the performance gain is because 3D applications tend to use floating point matrix transformations, which are helped along greatly by the vector units on the processors. It's not really 32 vs 64, so as much as the advantage of having hardware better suited to lots and lots of matrix transformations.
      • Did you ever think that the reason there are few consumer level apps that will take advantage of 64 bit architectures is that nobody has a 64 bit PC yet?
        • by Halo1 ( 136547 )

          Did you ever think that the reason there are few consumer level apps that will take advantage of 64 bit architectures is that nobody has a 64 bit PC yet?

          No. And I still don't think that's the reason. How can you make a word processor better using 64 bit code instead of 32 bit code? A spreadsheet? A web browser? An email client? A terminal emulator? A shell? A pdf viewer? I stand by my original point that most consumer apps don't need 64 bit operations.

          Some video/image editing applications may be able t

          • Re:64-bit Adobe apps (Score:5, Interesting)

            by ivan256 ( 17499 ) * on Wednesday May 14, 2003 @11:54AM (#5955280)
            No. And I still don't think that's the reason. How can you make a word processor better using 64 bit code instead of 32 bit code? A spreadsheet? A web browser? An email client? A terminal emulator? A shell? A pdf viewer? I stand by my original point that most consumer apps don't need 64 bit operations.

            I anticipated your short sighted response as soon as I hit the submit button. I should have realized that you would think I was talking about optimizing existing applications rather than designing new ones for new problems and I should have said all the rest of this stuff the first time... I even had a horrific vision of the word processor analogy. It was scary. Anyway:

            I'm not talking about optimizing existing applications, I'm talking about new applications; programs that do things that we don't use our PCs for now. When we had 8-bit PCs nobody did photo editing or full color page layout on a PC. When we had 16-bit machines nobody used a PC for CAD. Now we have 32-bit machines and are moving to 64-bit. There will be some major tasks that will become possible with 64-bit PCs, but the software isn't there yet because the customers don't have the processors.

            Also, I can think of two applications that every single computer user runs that can benefit dramatically from 64-bits, and Microsoft is waiting with the code already written for widespread 64-bit processor deployment to release them: Operating systems and filesystems. Having a 64-bit virtual address space can make your OS much more elegant and efficient since every possible I/O operation can be memory-mapped at once. Similarly, it has already been demonstrated that large relational databases benefit from 64-bit addressing even without taking advantage of the additional memory capacity, and many next generation filesystems will be relational databases.

            Sure current consumer apps don't really need 64-bits. If they did we wouldn't have them. It's the PC apps of tomorrow that will benefit. If you don't care to do more with your PC then essentially what you can do today, just more quickly then keep buying 32-bit CPUs. They'll continue to be available for decades...
      • The only real advantage of being 64 bit will be the ability to address more memory.

        Guess what? Today's multimedia applications can use 64-bit addressing because they have become VERY intensive in the use of computer system resources in general. Why do you think computers that are used to render computer animation usually have several gigabytes of system RAM and very fast hard drive access?

        Going to 64-bit computing will make system resource-intensive programs like Adobe Photoshop, Adobe Illustrator, Adobe
  • by tjwhaynes ( 114792 ) on Wednesday May 14, 2003 @09:12AM (#5953819)
    With the POWER 4 chips from IBM knocking passed 1.7GHz now, it's a reasonable assumption to make that the PPC970 will clock at similar levels to the POWER 4 chips. So at release, 2GHz plus isn't out of th reach for the fastest chips. Remember that the Athlons with Barton cores aren't clocked much faster than that with the 3200XP clicking around 2.2GHz.

    It's not just AMD clocking lower either. The Itanium 2 isn't clocked that fast. Given that 32 POWER 4 1.7GHz processors smoked the 64 Itanium2 1.3GHz processors configuration in the latest TPCC non-clustered benchmark, the POWER and PPC architecture is capable of putting a lot more work through in the same number of clock cycles. There are a lot of nay-sayers trotting out the GHz-is-god line and it is particularly misleading for 64 bit architectures.

    Cheers,

    Toby Haynes

    P.S. Disclaimer - I work in SOFTWARE for IBM, not hardware.

    • Quoted from Loop Rumors:

      We received word that two large shipments of Power PC 970 processors went to Foxconn in Taiwan, under a purchase order from Apple computer. Twenty thousand 1.4Ghz PPC 970's and forty thousand 1.6Ghz PPC 970's have already arrived in their hands. IBM's inventory contains fifty thousand 1.8 Ghz PPC 970's, of which forty thousand are destined for Foxconn tomorrow (Wednesday).

      IBM has listed as pending 2Ghz parts as well, which means that it will be in inventory within a month if their

  • by Mononoke ( 88668 ) on Wednesday May 14, 2003 @09:21AM (#5953886) Homepage Journal
    It appears to hold quite a bit of promise in bolstering Apple's currently almost obsolescent product line
    Other than a slight lack of processing horsepower, what exactly is "obsolescent" about Apple's product line? Everything they sell (well, except the iPod) can run the latest version of OSX, widely praised as the most advanced OS in the world. Even Apple's 5 year old machines can run OSX. They only have one machine left that even bothers with a CRT, and that's only for economy's sake.

    I'm sorry, but I don't see anything even approaching obsolete in Apple's product line.

  • PPC970 and Linux (Score:4, Interesting)

    by Conspire ( 102879 ) on Wednesday May 14, 2003 @09:54AM (#5954180) Homepage
    The ARSTechnica article specifically mentions the possiblity that IBM will use PPC970 in poweruser targeted Linux desktops, and no comments here about that yet, which surprises me.

    I ran "my first linux" on a DEC Alpha 512mhz 64 bit box that I got fed up running NT 4.0 on. I instantly became addicted, and eventually forced my company to switch to Linux on every computer (causing mass protest in the beginning, then mass praise over the years as we have grown and have no MS Tax on the books).

    I now have a Powerbook G4 and love it, except it is a little lagging in punch speed sometimes. And, although I love OS X, now that my company is used to zero license and upgrade costs thanks to GNU/GPL/BSD software, there is no hope of mass migration to OS X and Apple hardware in the company. It just does not make sense after seeing the dollar savings of running Linux on all the desktops.

    There is, however, always a need for powerful workstations that run Linux, and IBM might be pulling a rabbit out of its hat with this one. Will be very interesting.

    At minimum, I would buy one for that "64 bit memorabilia" value, to bring me sweet memories of my first Linux love, the Alpha that rid me of winbloze forever.........
  • Kitchen Sink (Score:3, Insightful)

    by buddha42 ( 539539 ) on Wednesday May 14, 2003 @10:06AM (#5954275)
    Its interesting to see how Intel and Sparc are moving toward explicit paralellism and extremely "wide" superscalar designs. And alpha, pa-risc are goners in favor of Intels designs.

    And yet here we have the last man standing in the "RISC turned hopelessly complex" generation, the Power970. When you look at this things design they threw everything and the kitchen sink in there! Most interesting is that batch parallelism where an instruction for every type of execution unit is queued up and when they're all ready to go they're executed in parallel. It will be interesting to see if that can scale given the latency it introduces, and the likelyhood that you won't always be able to fill every unit.

    • Re:Kitchen Sink (Score:5, Informative)

      by functor ( 31042 ) on Wednesday May 14, 2003 @10:13AM (#5954333) Homepage
      The Pentium 4 is, in fact, designed to scale to high clock speeds exactly so that it can tolerate lots of pipeline bubbles in flight without ending up stalling for too long.

      A lot of these tricks (high decode bandwidth, multiple instruction queues [really buffers meant for reordering the instruction stream], branch prediction, etc.) are meant to reduce hazards such as pipeline bubbles as far as possible, and the PPC 970 does these hazard-reducing operations rather well, too.

      And, yes, we're now in the post-RISC world where instruction complexity (particularly in the realm of SIMD and streaming/explicit cache manipulation instructions) is growing because simple instructions clearly aren't enough to allow for great throughput increments.

      (Read some of Stokes' older articles in the Ars Technopaedia; I'm sure you'll find them interesting.)
  • by Anonymous Coward on Wednesday May 14, 2003 @12:01PM (#5955326)
    I manage labs of macs, totaling over 300 of them. The users of my labs have no complaints about performance. (They usually only start to whine when the find out the processor speed, but then they are conditioned to think that MHz is MHz; and when asked if their work is slowed, compared to what they do on a PC, they answer "no" it is not slower.)

    What is this every non-mac user keeps saying that their performace is out of line with PC's? I have on my desk a hepped up dual P4 and a hepped up dual G4. XP on the dual P4 does not "feel," in day-to-day operations with standard apps like Office or Photoshop, much different from the dual G4. Comparing the MHz does not tell me anything. Using both side-by-side tells me a more "real" story about things. Now, perhaps XP is significantly slowing things down? :-)

    I only notice a difference with some high-end 3D apps like Maya or Lightwave, *maybe* also with some high-ed vid apps like Avid's.

    I am looking forward to the 970 though. Actually, I am very curious about any 64bit CPU. Hopefully the "growing pains" for anyone to move to 64bit is negligible.

1 + 1 = 3, for large values of 1.

Working...