Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
AMD

More Drooling Over The Opteron 201

Posted by Hemos from the looking-pretty dept.
bradv writes "I havent heard much about the new 64bit chips from AMD lately and was excited to find this article to satisfy my appetite for a little while longer. Probably more info than most people will ever care about. "
This discussion has been archived. No new comments can be posted.

More Drooling Over The Opteron More

More Drooling Over The Opteron

Comments Filter:

  • The OPTERON (Score:5, Funny)

    by Ogrez ( 546269 ) on Monday December 30, 2002 @10:25AM (#4981432)
    How can I not love a 64 bit processor that for some reason makes me think of the Transformers...
    • Yes, Opteron is a 64-bit Gestalt composed of two 32-bit chips.

      Athlons unite and form Opteron.

    • Hehe, funny-funny...

      But seriously, I don't love it because it's essentially a year late from when it was originally hinted to be launched, and several months late from when it was officially supposed to be launched according to the company's old roadmaps. And it still won't be here relatively soon.

      I seriously wanted my next machine to be a Hammer machine, and it would be if only they were out by now--which they should have been. All my machines, except for one 486 laptop, have been AMD based. However, my 800MHz Athlon died, and I need to upgrade *now* without waiting for the Hammer, which disappoints me greatly. So, the best PC I can build from existing parts is going to be a dual Xeon box. I'd rather have a single Hammer because its 64-bitness and strong, efficient performance at lower clockspeeds appeal to me--but I'm buying a Placer motherboard and dual Xeons because they're here *now*. They also offer a guaranteed upgrade path to probably 3.5GHz or better since Intel is good about keeping Xeon platforms viable for some time. Whereas, if I were to buy a non-Hammer Athlon right now, I'd be almost guaranteed to have little upgrade room by comparison.

      AMD just lost a long-time customer because of their lateness to market with Hammer, and I'm sure I'm not the only one.
    • What is the BFD?

      Some of us have been using 64-bit processor systems for several years now: can you say "Alpha", "MIPS" and "SPARC"? And cheap too, can you say "eBay" Little GNU/Linux Boys and Girls?

  • OMG look at all those pins (Score:4, Funny)

    by Nevermore-Spoon ( 610798 ) on Monday December 30, 2002 @10:26AM (#4981436)
    makes me feel prickily all over

  • If this chip... (Score:3, Insightful)

    by craenor ( 623901 ) on Monday December 30, 2002 @10:27AM (#4981438) Homepage
    Doesn't perform..and I mean really perform...I'm not sure if AMD will be with us much longer, which would be a shame.

    • Re:If this chip... (Score:2, Interesting)

      by Insane One ( 63516 )
      That would really hurt AMD but I don't think they would be gone. Other devices that use processors maybe what keeps them a float like pda's, pvr's.
      The whole cpu market is going to slow down more so then now. We are at the point where you can get a nice pc for $700 and a 6 month old one for $400 which is only 4% slower then the $700 one.
      What can you not do with 2.0ghz p4 or AMD 2200+?
      I can burn a cd while listening to mp3's chatting in msn/icq with 4 or 5 browser windows open. The processor (p4 2ghz) is only at 50% if that.
      Point is, once every one that wants a pc has it why would they need anything else? Is 4ghz going to much more then what you have now?
      The server industry will help keep both chip makers running but as far as desktop/laptops go it will slow down.

      • Re:If this chip... (Score:4, Interesting)

        by rotwhylr ( 618309 ) on Monday December 30, 2002 @11:20AM (#4981762)
        I'm sure that there are people here other than myself that were thrilled with all the "extra" power that the 486dx had over the 386 (no internal math coproc.) Sooner or later, software abilities and user demands will eat up the available hardware performance.

        More to the point, though, 32 bit vs. 64bit architecture is about more than clock speeds. In fact, typically I've seen 64bit cpus debut at lower clock speeds than contemporary 32 bit cpus.

        [Author hereby warns reader of his intent to use an analogy. The analogy is not designed to be airtight, or absolute proof of anything. It is intended to convey a point of view. Any attempt to stretch/abuse/extend said analogy beyond its intended limits will likely result in confusion.]

        Comparing a 32-bit cpu with a 64-bit cpu with half the mhz rating is roughly like comparing a 10k rpm, 4-cylinder motorcycle engine to a 5k rpm v-8.

        The bike will take one person (maybe two) and a small amount of cargo, and carry them at outrageous speeds. To carry more people or cargo, an SUV with the v-8 would do a better job.

        [Author briefly has a vision of a motorcycle tooling down the highway with an SUV v-8 crammed into it, penguin bumper stickers adhered all over it.]

        MHZ = speed, but speed does not necessarily equal power, and powerful does not necessarily equal useful.

        Or something like that ...

        • Comparing a 32-bit cpu with a 64-bit cpu with half the mhz rating is roughly like comparing a 10k rpm, 4-cylinder motorcycle engine to a 5k rpm v-8.

          This is a silly analogy that only shows that you don't understand what you are talking about. You don't magically get more per instruction with a 64-bit CPU, unless you have an application that really needs to do lots of 64-bit math. But if you *don't* then a 64-bit CPU can actually be slower because 64-bit pointers take up more space, so you increase the chance of cache misses.

          Note that we're not talking about a 64-bit bus as opposed to a 32-bit bus. Pentium's have had 64-bit busses from the get-go.

          The bottom line is that there's no magical speed-up from going to 64-bits.
        • [Author briefly has a vision of a motorcycle tooling down the highway with an SUV v-8 crammed into it, penguin bumper stickers adhered all over it.]

          No kidding.

          http://www.bosshosscycle.com/

          Or more reasonably:

          http://home.mira.net/~iwd/

    • Re:If this chip... (Score:5, Insightful)

      by Kevin Stevens ( 227724 ) <kevstev@@@gmail...com> on Monday December 30, 2002 @10:47AM (#4981556)
      I dont know why this needs to be continually restated but... AMD (Advanced Micro Devices, not Advanced Micro Processors) makes more than just PC processors! PC processor's are the tip of the iceberg when you consider embedded processors. TV's, VCR's, phones, dishwashers, etc. are all using embedded processor's now, and the # of products that use them in the future will continue to grow. Absolute worst case scenario, AMD will bow out of the CPU market. AMD is a huge company. One, two, three, or even four flopped products are not going to phase AMD. Look at AMD's website. PC processors are the most mainstream product they make, and thus the highest profile, but processors are just a cog in the AMD wheel. (And the same thing goes for Intel).
      • A company can easily fold just because people lose confidence in it. As far as the public is concerned, all AMD does is make CPU's in a big war with Intel that they are losing.

        AMD hasn't done enough to disparage people of that belief. Meaning that, in my opinion, unless they make some news in the CPU markets coming soon, confidence in them will further drop and their days will likely be numbered.
        • the embedded market is not prone to marketing armies and such like the mass market CPU market is. The embedded market is ruled by geeks who make informed decisions based on specs and their products needs. In fact, I would say that the only market where consumer confidence would matter is the CPU market. I guess it is nice to brandish the fact to EE dorks that they are the masters of CPU technology, but power consumption and things of that nature are much bigger issues when youre dealing with other products.
          • The embedded market is ruled by geeks who make informed decisions based on specs and their products needs.

            One company chose a CPU because that's what they used on the last project, and they wanted to ensure that reused as much of the code as possible. Sounds technical, except the "reuse" aspect was overhyped to make the product seem feasible, even though the CPU was severely underpowered for one of its requirements.

            Another company was going to switch CPUs because they were entering into a strategic partnership with the CPU maker, and wanted to ensure a consistent supply of flash chips.

            Applications with national prestige (or military use) often are required to use domestic products, even if there are superior foreign alternatives.

            Not as prone to marketing armies, yes, but purely based on specs and needs, not always.

      • Re:If this chip... (Score:5, Insightful)

        by Zathrus ( 232140 ) on Monday December 30, 2002 @10:59AM (#4981636) Homepage
        Yes, but as far as the PC market is concerned the original poster is correct -- if the Athlon64 and Opteron do not significantly increase AMD's market share then they're gone -- as in out of the market. Which means we essentially go back to a single chip maker monopoly for the PC market.

        If AMD wasn't around to spur Intel on (and vica versa) do you think we'd have a 3 GHz CPU available to the general public right now? Yes, you can question the need for one, or you can whine about the price, but the reality is that competition has significantly improved both prices and features.

        Will AMD, the company, go under? Doubt it. But they can't stay in a losing market much longer, and right now the x86 market is a massive loss leader for them.
        • Well, under the unlikely scenario that AMD completely drops out, what will probably happen is that Intel will drive prices up, and start making killer profits again (which is probably what AMD was drooling over five or however many years ago when they jumped in the CPU market), and then AMD will jump back in. AMD has been at this game awhile time now, and for a really long time was in a FAR worse position than it is now. In the K6-2 days, AMD was pretty much a joke, and was at best fighting for the low end- low profit segment of the market. Now, geeks see AMD on an equal footing, AMD has (well maybe had) high end- high profit competitive products, mobo producer support (this used to be a really big problem), and much, much more sway in the industry. I do not know what their bottom line is in the CPU business, but I bet they are alot happier now than in the K6 days.
          • LMAO @ "5 or however many years ago" but you're pretty much on point. AMD's marketshare has skyrocketed from what it was when the Athlon was first introduced. Their current parts are very competitive with the differences in performance between the platforms being essentially academic to the people that would ultimately use them. The problem now is one of perception, a problem that can be overcome if Hammer has a smashing (couldn't resist... sorry) debut, performance and stability wise.
      • This is true. However, AMD is second to Intel in both flash and processors, and has been bleeding money for at least four quarters [yahoo.com]. Intel ships more flash, has been quicker to market with their new StrataFlash [ebnonline.com] (vs. AMDs MirrorBit), and has better manufacturing capabilities. This allows Intel to set the stage on pricing and AMD to *hope* Intel doesn't drop prices too much because AMD has to go lower and watch profits piss away. In addition, Celeron continues to pressure the Athlon processors because the Duron was dropped. In addition, the new P4 "Banias" processors will leave AMD out because they won't even come close with a mobile processor for a while. Now, if Athlon64 dominates, and I hope it does, things will all change... AMD did well when the Athlon first came out and had several consecutive good quarters. Lets cross our fingers.

        JOhn

        P.S. Yep... I shoulda used preview...

      • Re:If this chip... (Score:3, Informative)

        by cheezedawg ( 413482 )
        Well, in Q1 2002, desktop processors accounted for 76% of AMD's revenue (according to their S&P stock report). Thats a little more than just a "cog in the AMD wheel" if you ask me. Their flash memory products only accounted for about 17% of their revenue during that same time period.

        The fact is AMD really isn't that big of a company. I think they have around 13,000 employees (compare with ~80,000 for Intel), and their revenues have been dropping like crazy ($500 million in Q3 of this year, compared to ~$6 billion for Intel). I don't think either AMD or Intel could really survive if they lost thier PC processor revenue. If you want to see a company that could survive losing a major product, look at Motorola or IBM. Now those are truly HUGE companies.
    • "If this chip...Doesn't perform..and I mean really perform...I'm not sure if AMD will be with us much longer...

      Good catch. We should probably avoid buying AMD products until things turn around for them.
    • I personally will not be drooling over this processor until Lightwave is ported over to it to take advantage of its processing capabilities.

  • more info ... (Score:5, Informative)

    by blandthrax ( 575357 ) on Monday December 30, 2002 @10:28AM (#4981448)
    SPEC results linked from The Inquirer ... here [theinquirer.net]
    • It looks like Athlon64 performance is going to be quite good. But even if it weren't, I hope AMD wouldn't hold up the release of the Athlon64 over concerns with benchmarks. If the price is reasonable, we'd buy them right now even if they ran at half the speed of a top-of-the-line Pentium4. The ability to address greater than 4 Gbytes of memory directly just outweighs even fairly significant differences in raw CPU performance. In different words, even a slow pointer dereference is still a lot faster than read/seek/write.

  • News for Nerds (Score:4, Funny)

    by Anonymous Coward on Monday December 30, 2002 @10:32AM (#4981470)
    "Probably more info than most people will ever care about."

    Yeah... And yet that is surely why you posted it on Slashdot - "News for Nerds".

  • Yea, what he said. (Score:4, Funny)

    by FreeLinux ( 555387 ) on Monday December 30, 2002 @10:32AM (#4981474)
    more info than most people will ever care about

    That's a great reason to put it on the front page.

  • Opteron is a tipping point (Score:5, Insightful)

    by Anonymous Coward on Monday December 30, 2002 @10:34AM (#4981480)
    Maybe I'm reading too much into this chip introduction, but I have a strong feeling that Opteron will be a classic tipping point in the CPU business. In other words, it will either be a blockbuster success, help AMD a lot, and take some of the wind out of Intel's sails, or it will flop and AMD will wind up being bought out by someone like IBM. I really don't think it (and AMD) will just muddle along.

    I think this is accurate because of the architecutural choice AMD made--instead of going with an all-new architecture, ala Itanium, they instead blew out the x86 system to 64 bits. That level of division in the CPU market at this time feels like it will have a very significant effect on the balance of power.

    • Re:Opteron is a tipping point (Score:5, Informative)

      by costas ( 38724 ) on Monday December 30, 2002 @10:58AM (#4981631) Homepage
      Well, Mr Cringely thinks the Opteron will be a huge win for AMD and makes quite a nice argument about it (based on hearsay though): link [pbs.org]

      (top sci/tech link from memigo currently; yes it's the holidays but a few things are happening /.ers...)
    • In other words, it will either be a blockbuster success, help AMD a lot, and take some of the wind out of Intel's sails, or it will flop and AMD will wind up being bought out by someone like IBM. I really don't think it (and AMD) will just muddle along.


      Well, on a similar subject, whatever happened to Intel's Merced/Itanium chips? I've never seen those in anything but supposedly they're being produced. It didn't kill Intel when that flopped horribly. Face it, nobody wants to leave the x86 chipset behind. There's just too much software available.

      • You don't see Itanium chips because they cost a couple of thousand dollars each and are targeted for the high end server market only (Itanium was never meant to compete with Opteron). According to Intel, Itanium 2 is selling according plan right now.
      • We are evaluating a dual Itanium2 machine right now. It's screaming fast, but with the price point I doubt we'll buy any.
  • Verify the maximum amount of drool the chip can sustain before you start to droll over it.

  • Pffft (Score:3, Insightful)

    by Anonymous Coward on Monday December 30, 2002 @10:50AM (#4981573)
    I didn't think that was a very good article. There seemed to be a lot of guesses in there, none of which appeared to be particularly informed - or at least, they were not explained - and some of it sounded downright childish. Like:

    I don't know what Reserved might mean. One of the reviewers says that maybe in this case the processor turns into DSP. It's a mad idea, but if AMD realized it, this processor would be second to none in some kinds of operations. :-)

    or

    AMD realizes it, and at present they develop several independent versions of the compiler together with famous software development companies. I won't unveil their names - AMD will do it if necessary. You just should know that at launch the processor will have the required support of the compiler allowing using its architectural advantages.

    sorry?

    No, i'd rather read C'T [heise.de], at least they already have one of them chips on the test bench

    • Re:Pffft (Score:2, Insightful)

      by ergo98 ( 9391 )
      Personally I found the article very difficult to read: Clearly (or rather hopefully) English is a second language for the author. I don't fault them for it as my second language is l33t Quake3 online speak ("Ghey! Noobz!"), however it did make reading this English article rather difficult.

      It did seem rather void of information as well, and where there was information it was oft of dubious value. For example: If we interpolate, we will see that 512 KB must result in 7-8% gain in the SPECint 2000 (they got this value by dividing the gain of going to 1MB of L2 cache by 2. Of course the source of that information is absent so who knows). Wow, so 1GB of cache must result in a 15000% gain! Of course in reality such a simplistic interpolation isn't accurate, and indeed going to 512KB of L2 might yield 14% of the 15% gain of 1MB (depending on the test set).
    • It just has the minor problem beeing written in German, which might be as small problem for the current audience.

      For those people not interested in learning a new (human) language, I suggest the english version [heise.de].

      I suggest a article benchmark ;).
  • They tested with Slackware 1.2? They might want to try a version released within the past 5 years though. Perhaps version 8.1.
    • They need to test with both new and old x86 OSs to see whether their cpu's are really compatible. Slackware is the oldest mainstream linux distro around so it makes sense to test older linux compatibility with an old slackware.
  • Probably more info than most people will ever care about.
    In normal circles, that's an innocuous statement. But, when said on slashdot, it's like slapping us in the face with your glove...or triple-dog-daring us to read the whole thing.

    Of course, the majority of us still won't read the article.
  • All data are ECC protected.

    "Data" is a plural word, finally someone noticed.
    • In line with with the modern English usage,my datums are ECC protected, but my data is merely protected by ECC

      All your datums are belong to us!

    • Data hasn't become a "mass-noun" now, e.g. cereal, water, snow, etc? It's to the point now that "data are precious" sounds just as bizzare as "snow are precious."

      To back up my point, see the following entry here [reference.com], specifically the "Usage Note" section.

      Language changes. We aren't speaking latin anymore. Deal with it.

  • x86 continues to live (Score:2, Insightful)

    by Anonymous Coward
    Do we honestly need to keep breathing new life into x86? Whenever I see an article about intel posted, all I see is "x86 sux" posts (and I agree). However, when its Itanium vs. Opteron, its always "Go Opteron Go". Itanium actually uses a new instruction set, while Opteron keeps hacking x86 to work for yet another generation.

    • Re:x86 continues to live (Score:5, Interesting)

      by jbischof ( 139557 ) on Monday December 30, 2002 @11:25AM (#4981798) Journal
      oh you hadn't noticed the tendency of everyone to blindly dislike Intel regardless of what they do?

      Itanium has lots of cool new features that compilers could be using and people could be taking advantage of, but it doesn't have good backwards compatibility, and therein lies the problem.

      • Itanium has lots of cool new features that compilers could be using

        yeah, if it weren't for intel pricing their compilers out of range of the avarage developer. As long as MSVC doesn't properly support it, that argument doesn't take you very far.

        Then again, .NET could change all that if the CLI is optimized for itanium...

      • Re:x86 continues to live (Score:5, Insightful)

        by Zathrus ( 232140 ) on Monday December 30, 2002 @12:48PM (#4982332) Homepage
        Itanium has lots of cool new features that compilers could be using and people could be taking advantage of

        Yes, and Intel would really appreciate it if someone would develop a compiler that takes full advantage of the Itanium. Really. Please. Because their own compiler is still struggling with the problems inherent in VLIW... yes, it's much, much, much better than it was a couple years ago, but it's still nowhere near where it needs to be.

        And it costs a fortune. But, hey, if you can afford a $9000 chip, you should be able to afford the compiler too.

        I like and respect Intel... I've grown beyond the newbie EE stance of "it sucks because it does", and recognize that they have some of the best minds working there, and that their fab processes are second to none. But Itanium has been a massive disaster for them, and they're now caught between a rock and a hard place. They can continue developing future revisions of IA64 and hope that someday their engineers figure out how to make it work well, work cheaply, and work fast with legacy code, or they can commit corporate hari kari and adopt x86-64 from AMD. Or they could do something similar, but different, to x86-64 on their own and just piss off everyone. Bad choices all around.

        The only chance Itanium has is if AMD flubs the Athlon64/Opteron launch. AMD will probably pull out of the market shortly after and Intel can gradually increase profit margins to the point where throwing cash at a losing proposition (IA64) remains viable. And eventually force everyone to transition, like it or not (which, admittedly, would probably be a good thing in the long run, but the short term would suck).

    • Re:x86 continues to live (Score:4, Insightful)

      by Kevin Stevens ( 227724 ) <kevstev@@@gmail...com> on Monday December 30, 2002 @11:28AM (#4981809)
      Well, x86 continues to live because of the absolutely enormous amount of software written for it. Could you even begin to fathom starting at zero again? Itanium failed for this very reason. Sure, one of the holy grails as programmers we are supposed to chase after is full portability, but technical and time constraints often make that a dream. It is unfortunate, but the only way I can forsee a momentous move to a completely new architecture is if Microsoft really monopolizes the entire software industry, and then gets in bed w/ CPU makers and agrees to port all of the apps to this new architecture, and then everyone will make a ton of money selling new hardware and software that is legacy free. I will place my bet that in 10-15 years we will still be having this discussion.

  • The biggest advantage and the drawback (Score:2, Funny)

    by Anonymous Coward
    ..is that the processor is based on x86 architecture.

  • 64-bit architecture at last... (Score:3, Insightful)

    by LotusFlower ( 634967 ) on Monday December 30, 2002 @11:34AM (#4981851) Homepage

    If AMD can deliver this on a desktop level, then Moore's Law can once again be considered applicable...

    Think about it - the main problem in terms of pushing computing power these days is electron migration, caused by extremely high clock rates.

    By doubling the word length to 64-bits, you can reduce the clock rate of the chip, and will still be able to perform more instructions per second than your top-of-the-range Athlon/Pentiums.

    This was always the case with graphics cards; the GeForce 256 was a big step up from the Riva chipset, due to doubling the word length.

    Supercomputers, such as the SGI Origin series, have been using 64-bit processing for quite some time now (MIPS processors), and while the Itanium series has its flaws (like a lack of backward compatibility), surely it's time to move on from the same old x86 architecture?

    We don't all have to wait for Microsoft to make their WinXP 64-bit version mainstream; there's no point in them pushing this until the 64-bit architecture breaks into the home market.

    Because the Opteron has this backward compatibility, then the 64-bit architecture will reach the home users, and they can upgrade to the 64-bit version as soon as it is deemed economically viable by Microsoft to release it.

    I wonder what kind of performance increase you'd get from a program such as SETI@home or Distributed.net by upgrading to a 64-bit platform...

    • Re:64-bit architecture at last... (Score:5, Informative)

      by jbischof ( 139557 ) on Monday December 30, 2002 @11:56AM (#4981970) Journal
      > Think about it - the main problem in terms of pushing computing power these days is electron migration, caused by extremely high clock rates.

      Electron Migration? what are you talking about. Processors continue to get faster and faster due to improved processing technology and increased parallelism. Leakage and electromagnetic interference from the clock signal are major problems today but who knows what scientists are working with nowadays.

      > By doubling the word length to 64-bits, you can reduce the clock rate of the chip, and will still be able to perform more instructions per second than your top-of-the-range Athlon/Pentiums.

      That is absolutely not true. Having 64 bits allows you to access a larger amount of total memory, and it lets you put more information in each instruction. The amount of data you can work on in any given clock cycle is proportional to the cache access and bandwidth and the register size (Neither of which inherently need 64 bit long instructions).

      To perform more instructions per second (or instructions per clock cycle) you need instruction level parallelism (ILP). This has been a major goal of processor manufacturers for many years now. Intel had two main ways of trying to increase ILP.

      1. Use an instruction set with inherently more parallelism - allowing you to issue multiple instructions at once - Itanium
      2. Try executing from more than one thread at a time - allowing you to use more of the processor per clock cycle - Hyperthreading (now on Pentium 4 processors).

    • Re:64-bit architecture at last... (Score:4, Informative)

      by pclminion ( 145572 ) on Monday December 30, 2002 @12:45PM (#4982321)
      Think about it - the main problem in terms of pushing computing power these days is electron migration, caused by extremely high clock rates.

      Say it with me: There is no such thing as electron migration. There is, however, something called electromigration and it has nothing to do with clock rate. The problem is that as electrons flow in a conductor, they collide with lattice ions and push these ions around a little bit. This isn't a problem in the macroscopic world since wires are so big, but in a microscopic (or nanoscopic) scale this can lead to melting and diffusion of the conductor into the surrounding medium. The copper atoms slowly diffuse into the silicon around them, almost like a gas (a very slow moving gas).

      Since these motions are caused entirely by the force of electrons colliding with the atoms, they are completely determined by the kinetic energy of the electrons -- i.e., how fast they move. And that in turn depends on the mean-free-path length (a property of the conductor) and the electric field within the conductor. It has absolutely nothing to do with clock rates.

      Newer, high-speed chips may suffer more from electromigration than slower chips, but this is only because the new chips have much thinner wires and are therefore subjected to a greater current density at a given voltage. I.e., more electrons flow per unit area, so the number of electron-atom collisions goes up.

  • we need this badly (Score:5, Interesting)

    by g4dget ( 579145 ) on Monday December 30, 2002 @11:39AM (#4981881)
    Workstations based on the Itanium or SPARC capable of addressing more than 2-4Gbytes of RAM are very expensive (above $10k). IBM's Power4-based systems are even more expensive.

    Lots of data-intensive applications desperately need more than 2Gbytes of RAM. If Opteron can deliver that for only a modest premium over regular Athlon-bsaed PCs, it will be a huge success. And if it can run existing binaries in 32bit mode and work with existing drivers, that's icing on the cake. There is just nothing else like it out there.

    As soon as they come out, assuming Linux does run reasonably well on them and there are no unexpected show-stoppers, we are going to buy half a dozen of them. We want a Beowulf cluster of these.

  • Lots of info...newbie reviewer (Score:3, Insightful)

    by dan g ( 30777 ) on Monday December 30, 2002 @11:51AM (#4981936) Homepage
    The article has some fairly interesting material, but what really amused me was how the reviewer didn't really seem to understand most of what he was writing about. He seems to have alternated between copying stuff directly from some marketing glossy and what he could get from a comp arch textbook...ususally following up with something like "I'm sure this is good for something or someone somewhere, beats me though!". You can almost see him scratching his head. It all starts when he is confused by the 'Resevered' entry in a table of register settings.
  • Is this the first time the guy has looked at processor specs or seen a pinout diagram or bitfield description? He seemed to get awfully excited about the word "reserved" and imagined some sort of super-computer hidden inside a magic bit. It means "reserved for future use" i.e. "it doesn't do anything yet so don't twaddle it or you will break something years from now when we find a use for it."
  • Just before it explains the processor core, the article lists tested operating systems. My question is, where the hell did they find copies of slackware 1.2 and 2.0? And why version so old?

  • Impressive Numbers (Score:3, Interesting)

    by MBCook ( 132727 ) <foobarsoft@foobarsoft.com> on Monday December 30, 2002 @12:10PM (#4982066) Homepage
    Scroll down on the HardOCP [hardocp.com] homepage and you'll see a graph from that German magazine pitting a 1.2GHz x86-64 against a P4 2.2GHz. Even at about half the clockspeed, it gets very close or beats the P4 most of the time. When it's behind, it's not by far. And sometimes when it's ahead, it's WAY ahead. And this is at about half the clock speed too. So if AMD can get these things out at about 2GHz, I'll be in line for sure.

    As to people saying that AMD is dead if x86-64 doesn't work, I agree. They are basically betting the farm on the x86-64 chips. If they don't payoff, they'll most likely leave the desktop/server/whatever CPU market. They'll still be alive in microcontrollers and millions of other things, but they won't be competing with Intel for the CPU of your PC. If this happens, I'll be worried, becase we all know that we need a second big name in CPUs to keep prices in the "ludicrous and below" area.

    BUT... if they don't take off on the PC side, the chip is still superior to the little 1.x GHz PPCs that Apple is using. If they could be the new chip for Apple, then they could stay in the CPU market, and Apple could get a major contender again (CPU wise). I'd love this to happen. OS X is already proted (according to rumors, and we know that the kernel already runs on x86s, so it would be fast ported to the -64s, especially by AMD). Software would be easy to port from PCs to Macs (no endianess mess). Even as just a failed market expirament, this could mean alot to Apple, AMD, and Intel.

    All speculations, my opinions, and such. If you doubt me, send $200 to me and I'll consider your point of view better. The address is below....

    (address cut due to excessive donations)
    (WOOT!)

    • To move to x86-64, Apple would have to plan to phase out PPC entirely. There's no way the caliber of users we see on the 'Switch' commercials could handle having to know whether their machine was x86 or PPC before buying software at CompUSA.

      Yeah, Apple's been through a similar migration before, but when they went from 680x0 to PPC, their new architecture was fast enough to do a passable job of emulating the old CPU for legacy software. I'm not convinced that would work this time. Let's say the Opteron is released at 2.0GHz, how fast a PPC will it be able to emulate? Could the legacy software run at least as fast as it would on a low-end iMac classic? I doubt it, but I could be wrong.
      • A recent interview with the President of AMD said that he'd not been contacted by Apple. Apple apparently cross compiles a lot of their OS codebase on Athalons, but I'm told it is more just to keep options open in the future and to help find bugs. I'm not sure if that includes Aqua or iApps. (I doubt it)

        In either case the 970 with be in the same general class as the Opteron and will offer very good SMP. Apple's problem is that between now and perhaps September they are way behind the speed curve for desktop machines. (The portable market they do much better in, IMO)

    • If the development of the 970 by IBM is coming along as rumored, there is absolutely no reason why Apple would want to go to the Opteron over the 970. The 970 is also 64 bit, will probably achieve similar clock speeds, and most importantly is still a PPC chip. So rather than create a mess 10x worse than the switch from 68k to PPC by going from PPC to x86-64, Apple would be able to switch to a 64 bit architecture and still maintain backwards compatibility.

  • Hypertransport (Score:3, Informative)

    by kylegordon ( 159137 ) on Monday December 30, 2002 @12:17PM (#4982105) Homepage
    You may be interested to read about the HyperTransport capabilities of the chip at http://www.hypertransport.org [hypertransport.org]
    One thing I found particularly interesting was the SMP abilities of the AMD, through the use of Hypertransport. It allows multiple chips to be used on the same board without all the glue logic normally associated with SMP setups, so you can have arrangements like the Power4 and suchlike, without enormous amounts of additional circuitry.
    Funky stuff

  • An echo from the past (Score:4, Interesting)

    by Veteran ( 203989 ) on Monday December 30, 2002 @12:39PM (#4982280)
    Back in the early days of the 8086 there was a processor from Zilog called the Z800 (not the z8000 - which was a different chip). It was a super chip; it ran far more software than the 8086 - it was faster and easier to program - being directly compatible with the existing core of CP/M software. There was every reason to believe that the Z800 would wipe the 8086 from the computer market.

    The problem was that Zilog never actually got around to building the Z800; it was a classic example of vaporware.

    The real question for AMD is: can they build the Opteron? Sadly, the longer the Opteron is delayed the more likely it is to turn from silicon to vapor phase.

    I suspect that the real reason that the Intel X86-64 processor got canceled is that Intel decided that the Opteron was likely going into vapor phase. The fact that AMD has little to say on the subject sadly confirms this. The z800 was never officially dropped, it just faded away quietly - which is how vapor phase works.

    And yes, I have a manual from Zilog featuring the Z800 - so the documentation AMD has recently produced really doesn't matter much.
    • Well, first of all, the chips have been shipping and benchmarked.

      Furthermore, 64bit really matters to a lot of people; it's not just a few more instructions. And the other main alternative, Itanium, is hugely expensive, a pain to deploy, doesn't run a lot of software, and doesn't perform all that well.

      Sure, AMD may still fail to pull this off for a variety of reasons. But it really looks to me like they are very serious about making it work.

    • The real question for AMD is: can they build the Opteron? Sadly, the longer the Opteron is delayed the more likely it is to turn from silicon to vapor phase.

      If this were an issue Itanium would already be gone. Remember the original schedule?

      I suspect that the real reason that the Intel X86-64 processor got canceled is that Intel decided that the Opteron was likely going into vapor phase. The fact that AMD has little to say on the subject sadly confirms this. The z800 was never officially dropped, it just faded away quietly - which is how vapor phase works.

      I don't suppose it matters that AMD has demoed working silicon?

      And yes, I have a manual from Zilog featuring the Z800 - so the documentation AMD has recently produced really doesn't matter much.

      You are of course right in the sense that until the exact moment AMD actually begins shipping some volume of these chips at full speed, it is unknown if they will actually be able to do so. However, I think you're being very naive in your assessment.

      The best evidence I can offer of this is the Cray supercomputer being built using over 10,000 Opterons [eweek.com]. Trust me, Cray wouldn't risk it's fragile reputation and profits on "vaporware".

      Gee, I wonder why Itanic didn't get the design win? ;-)

  • Bob Cringely's take on it... (Score:3, Informative)

    by hdurdle ( 199425 ) on Monday December 30, 2002 @12:59PM (#4982414) Homepage
    Cringley has something to say [pbs.org] about the AMD/Intel 64bit/32bit situation, specifically how Microsoft will probably drive how it all pans out.
    • I'm not sure this Cringely guy knows what he is talking about.

      > Yamhill was cancelled because it might have caused confusion with Intel's other 64-bit chips. The unofficial line says Yamhill was a dog.

      Uhhh. No. Intel never admitted there was a Yamhill, and if there was it was a reaction to AMD's hammer chips. This wasn't a cancelled project that customers were promised, it was an exploration into 64 bit extensions to the x86 instruction set and would have conflicted with other processors.

      > While Intel has 64-bit chips in the pipeline, specifically the Itanium, which is the successor to the disappointing 32-bit Xeon

      1) Itanium is not the successor to Xeon, they are both being sold right now. Itanium is much more expensive and high-end. Xeon is cheaper (relatively) and low-end. They both address different market segments in Intel's eyes.

      This Bob character predicts that Intel will have to make a x86-64 processor (like Yamhill) that supports AMD's instruction set. However I don't think this will happen. Intel is too vested in its current processors to just give in and go AMD's way.

      Intel is praying that average consumers don't want 64 bits too badly, and they will stick with Intel products for a while and higher up servers will really want 64 bit, enough to adopt Itanium as it gets better and better with each revision (Itanium III is coming out soon).

      If Opteron does well, who knows what will happen. But it will be much harder for AMD than they originally thought. The same people going "I don't know about that Itanium chip" are saying "I don't know about that Opteron chip" and currently some ridiculously large ~90% of mid-range corporate servers use Xeon chips. Corporations like lots of support and extra features and reliability that desktop processors don't neccesarily have to have.

      If Opteron doesn't pan out - then Intel will be able to do what it wants with the 64 bit market. I don't know what that is but I would bet a bundle that they have something planned. You don't survive for long in this market without looking as far into the future as you can.

      • The only problem I foresee for Intel is Microsoft. If Microsoft decides to not bother to support your chip platform _or_ decides to support it slower (releasing software a year later, for example), you'll end up being nailed vis-a-vis your competitor who can run WinXP Pro without modifications with 64 bit support for their relevant application-level software.
  • I've been using 64 bit sparcs for a while and my N64 has a 64 bit cpu. 99+% of what runs on both platforms is in 32 bit mode because its faster. There is nothing that I deal with were doing atomic 64 bit operations is faster than atomic 32 bit calcs and the 64 bits means more crud to hit the stack on function calls, context switches and intterupts. If you look at the Power PC, you will find it takes 3 instructions to load a 32 bit constant they way most compilers generate it, the 64 bit flavors need 5. That can't speed thigns up too much can it? The people doing the N64 emulators found that there is no 64 bit code at all in any comercial game for the platform. So if 64 bits is so great, why have coders (that need the speed) not been doing it?

    I'm happy with 32 bit cpus. If I need longer word sizes, I need much longer word sizes that are byte addressable. Something in the area of 2k bits would be a nice start but I expect a full raster line in RGB would be even better but thats in the relm of vector processing and the video cards seem to have that down quite well.

    I know people will say 64 bits speeds up access to big file systems but I can't expect a few 32 bit adds with carrys aren't much faster than dumping 2x stuff to the stack everytime the dma controller hits the end of a 512 byte buffer.

Slashdot Top Deals

There is very little future in being right when your boss is wrong.

Close