Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
AMD Hardware

AMD Athlon 64 6000+ Launched And Tested 156

Posted by Zonk from the kicking-the-tires-lighting-the-fires dept.
Spinnerbait writes "AMD officially launched their next speed bump in the Athlon 64 product line, in the form of a new 3GHz part branded the Athlon 64 6000+. This new dual-core Athlon 64 sports 1MB of on-chip cache per core and is designed for AMD's Socket AM2 platform. This chip is still built on AMD's 90nm fab node and is comprised of some 227 million transistors. It also carries a thermal power profile of about 125Watts. Unfortunately, in all the benchmarks seen here, it was still unable to catch Intel's Core 2 Duo E6700 chip at 2.66GHz."
This discussion has been archived. No new comments can be posted.

AMD Athlon 64 6000+ Launched And Tested More

AMD Athlon 64 6000+ Launched And Tested

Comments Filter:

  • But hey... (Score:3, Funny)

    by Goaway ( 82658 ) on Tuesday February 20, 2007 @09:48AM (#18081354) Homepage
    At least it uses more power!

    • Re:But hey... (Score:5, Funny)

      by networkBoy ( 774728 ) on Tuesday February 20, 2007 @10:00AM (#18081498) Journal
      So they're about 2 years behind Intel on this? Prescott topped out around 135W IIRC, so AMD has 10W more to go...

      Burn karma, burn.
      -nB

    • Re:But hey... (Score:5, Funny)

      by moranar ( 632206 ) on Tuesday February 20, 2007 @10:05AM (#18081598) Homepage Journal
      Nah, with cool n' quiet we'll be able to run it at half of its power and clock speed. Like an Athlon 3000, for example, only more expensive.

  • DOS (Score:5, Interesting)

    by Short Circuit ( 52384 ) * <mikemol@gmail.com> on Tuesday February 20, 2007 @09:51AM (#18081380) Homepage Journal
    I've always wanted to try running DOS on a processor with 1MB of L2 cache...there's just something retro wicked about running an OS where the entire base memory fits in on-die cache.

    I have to wonder if qemu and the kernel's kvm will allow me to dedicate an entire core to a DOS image.

    • Re:DOS (Score:5, Interesting)

      by ArcherB ( 796902 ) * on Tuesday February 20, 2007 @10:17AM (#18081722) Journal
      I have to wonder if qemu and the kernel's kvm will allow me to dedicate an entire core to a DOS image.

      Or you could just boot off of a DOS formatted USB key. I remember hearing that the Athlon64 would run all OS's down to DOS 2, I believe.

      • Re: (Score:3, Funny)

        by Tanktalus ( 794810 )

        I'd hate to be part of that QA department...

    • Re:DOS (Score:4, Funny)

      by misleb ( 129952 ) on Tuesday February 20, 2007 @12:34PM (#18083598)
      Dude! Totally! And then you can run Wordperfect 5 at blazing speeds!

      There's retro... there's wicked... and then there's DOS.

      -matthew

      • Re: (Score:2)

        by Curien ( 267780 )
        Actually, DOS is used often enough in embedded environments.

        • Re: (Score:2)

          by misleb ( 129952 )
          Exactly. It is neither wicked nor retro. It is just plain ol' boring DOS. Barely any different now than it was in 1981.

          -matthew

    • Re:DOS (Score:4, Informative)

      by Rick17JJ ( 744063 ) on Tuesday February 20, 2007 @12:50PM (#18083910)

      My AMD 64 3800+ has FreeDOS [freedos.org] on the 2nd partition of my 1st hard drive. It is formatted as a FAT-16 partition. It is one of the choices on the GRUB boot menu [wikipedia.org]. I only boot up DOS every once in a while, but it does run on my AMD 64 computer. About a year ago or so ago I had IBM PC DOS 2000 installed on the 1st partion which also ran well. I later reformatted that partition as NTFS and installed Windows 2000 on my first partition instead. I still have FreeDOS on the 2nd partition. I have Slackware Linux installed on my 3rd partition and in that case I have 32-bit version of Linux running on a 64-bit computer. On a logical partition I have the AMD-64 version of Kubuntu 6.10 (Edgy Eft) Linux [kubuntu.org] which is what I like best and use most of the time.

      An easier way to run an old DOS program under Linux or Windows would be to just use the free DOSBox [sourceforge.net] program. In the past, I also used VMWare and had PC DOS 2000 installed on one the the virtual machines. With VMWare I was able to run Linux, Windows and DOS all at once.

  • Low power chips too (Score:5, Informative)

    by Bill Dimm ( 463823 ) on Tuesday February 20, 2007 @09:53AM (#18081418) Homepage
    In the full announcement [yahoo.com] they also mention new 45W single-core desktop processors: Athlon 64 3500+ for $88, and 3800+ for $93.

    • Re: (Score:2, Interesting)

      by D3m0n0fTh3Fall ( 1022795 )
      The review mentions this on the last page anyway. I too would be more interested in these low power chips. The ultra low power X2 would have to be the most interesting proposition for a home server... 35 Watts or something?

    • Re: (Score:3, Informative)

      by Curien ( 267780 )
      Yeah, if you can find them.

      A few months ago, I decided to build a fanless desktop.* The socket AM2 had recently debuted, so I decided to buy one of them and pair it with one of the new low-power Semprons. It took days just to /find/ someone who actually claimed to be able to get their hands on one. When I finally ordered one, the order got delayed -- first by one week, then by two. I think it finally arrived three weeks after I ordered it. And I'm lucky, I think. My supplier was in Germany; I don't think Am

      • Re: (Score:2)

        by egghat ( 73643 )
        According to heise [heise.de] the new 45 watt versions are the followups to the old 65 watts EE version (not the 35 watts EE SFF versions). While the EE SFF were essentially unavailable, the "normal" EE versions were available. At least here in Germany.

        Bye egghat.

  • cool (Score:2, Interesting)

    by jrwr00 ( 1035020 )
    I would still buy this over intel's processor, my god, that thing has alot of pins
    • Here's [cpu-collection.de] a processor that only has 16 pins...
    • I don't get it. You would still buy a worse performing processor because it's from AMD? I would only do it if it was cheaper.

      • Re: (Score:2, Interesting)

        by Tekzel ( 593039 )
        I only buy AMD. Even if it is currently a "worse performer" (and that is by a much smaller margin than that the Athlon 64 had over Intel's NetBurst processors for a long time at that). I like AMD and think they are good for the market. If they went poof over night Intel would rape the CPU market like marauding viking horde. As I am sure AMD would if Intel went poof. They keep each other honest.
  • What is the point of releasing a new iteration of an existing platform to bump up speed and still not catch up with the competitions products?

    Wouldn't they have been better served re-routing this R&D effort/money into something which would put them back on top of either the price or performance curves?

    • Re:Question for the AMD fans/afficianados (Score:5, Insightful)

      by God'sDuck ( 837829 ) on Tuesday February 20, 2007 @10:07AM (#18081612)
      Better served? Yes, of course. Possible in the short term? No!

      Both manufacturers hurry out minor iterations of their existing processor set while readying the next generation; it's a stop-loss tactic, since they can pop something like this out in the engineering equivalent of an afternoon, and it masks the fact that they're falling behind. Rather like the Pentium IV QRSTTurboMach5's that were coming out almost weekly back when Athlon was pantsing Intel. Intel knew they sucked just as much as we did -- but not releasing them would have terminated their share price.

      Besides -- your average Dell buyer only sees "New Release", not benchmarks.

      • Re:Question for the AMD fans/afficianados (Score:4, Insightful)

        by LWATCDR ( 28044 ) on Tuesday February 20, 2007 @10:22AM (#18081792) Homepage Journal
        Not only that but there are a lot of people with AM2 motherboards that might like to do a simple upgrade without buying a new motherboard. Not to mention that Dell, Gateway, and HP probably have a nice supply of AM2 motherboards and system that they can now sell with a faster CPU.
        I am still ever hopeful to see what AMD does at 45nm.

        • Re: (Score:2, Informative)

          by archen ( 447353 )
          Not sure why you would be in any hurry to upgrade if you're just worried about your mainboard. AM3 processors will work in AM2 sockets.
      • Comment removed based on user account deletion
        • The average Dell buyer barely knows how to turn the computer on.

          Normal people do not upgrade their CPUs. They just buy new computers.

    • Re:Question for the AMD fans/afficianados (Score:4, Insightful)

      by Buddy_DoQ ( 922706 ) on Tuesday February 20, 2007 @10:17AM (#18081724) Homepage
      If you're an enthusiast with an existing AMD rig, why not just plop in a new CPU rather than a full Intel combo upgrade? If I was AM2 rather than 939, I myself would be down on this in a heartbeat. From the looks of things, overall it's about on par with Intel's bang-per-buck chips (E6600/E6700), sounds like a good move to me!

      Realistically, there's so much transition going on right now, DX10 cards, new operating systems, multiple cores, I think it's best to let this storm even out for another 6-12 months before considering a full upgrade. So for now, plop in that new CPU or GPU, if need be, and have fun!

      • Re: (Score:3, Insightful)

        by Lord Crc ( 151920 )
        If I was AM2 rather than 939, I myself would be down on this in a heartbeat.

        I also got a 939 rig, and I haven't quite understood the whole AM2 move from AMD. From what I've seen so far, AM2 doesn't bring a whole lot of improvements to the table, but what it does is equalize the upgrade costs between an AMD system and an Intel system. And in these days, that's hurting AMD bad I suspect.

        If AMD needs some easy cash, why not release something for the 939 system? A reasonably priced, speedy dual core for instanc

        • Re: (Score:3, Informative)

          by LousyPhreak ( 550591 )
          939=DDR Ram
          AM2=DDR2 Ram
          • 939=DDR Ram
            AM2=DDR2 Ram

            That's part of my point, but I fail to see yours. I was talking about taking one of the beefier X2's they had for the 939 and ship it cheap.
          • IIRC, DDR2 can actually be slower in some configurations than DDR, because of increased latency.
          • Somehow I doubt they would have been prevented from making Socket 939 motherboards that supported DDR2. I had Socket A motherboards in both the DDR and regular SDR variety. A new chipset to support it and they'd be set to go. Then regardless of if you have a 939 mobo with DDR or with DDR2, you still have access to the same processors.
            • You have to remember that Socket A chips used a northbridge chip as the memory controller. In order to change memory type for the 64 line of processors they need to change transistors on the CPU itself because that's where the memory interface logic is found. IIRC DDR2 also uses more signal pins, or at least differently configured ones, than DDR. And also the two memory types have different power requirements which may affect the CPU interface as well.

            • Memory controller is on die. (Score:4, Informative)

              by DrYak ( 748999 ) on Tuesday February 20, 2007 @01:00PM (#18084116) Homepage

              Somehow I doubt they would have been prevented from making Socket 939 motherboards that supported DDR2


              It does.
              With Athlons, the memory controller are on the northbridge (just like Intel's). You can put whatever memory on the mother board, as long you put the correct north bridge.
              So that's why you have both SDR and DDR Socket A mother boards.

              A new chipset to support it and they'd be set to go.


              That's not how is works with Athlon 64s : They have on die memory controller. The type of memory you can connect to the mother board is directly determined by the type of processor. And until recently each type of processor has it's own connector :

              Single Channel DDR : Socket 754
              Dual Channel DDR : Socket 939
              Dual Channel DDR2 : Socket AM2

              Only from now on will you have mutually compatible AM2/AM2+/AM3 mother board, which will use mechanically compatible connectors and the limitation will be only be the memory controllers on the chips (AM3 processors have both DDR2 and DDR3 and can got in all 3 motherboard. AM2 chips only have DDR2 and only go in AM2/2+ MB).

              On Athlon 64 motherboard, the nortbridge is nothing more than a controller in charge of peripherals and their busses and doesn't touch the memory at all. It's completly agnostic of the memory and only speaks "Hypertransport" to the CPU. It is mutually interchangeable with all mother board. And in fact you can find the exact same VIA KT880 AGP chipset on mother board from 754 all the way up to AM2, regardless of the memory.

              You can make different king of motherboard with the same chipset.
              But 939 Processor can only connect to DDR memory, so you're stuck with it.

              (On the otherhand, we could imagine building PCI-e nForce6 motherboards for Socket 754 CPUs and AGP KT880 mother board for AM3 connectors. But no company curently bothers.)

              As a side note, that's one of the reason why Athlon64 have a smaller cache :
              - Unlike Intels they're not limited by the bus speed for memory transfers. They have access to memory at full speed.
              - Memory access is direct, without having first to be processed by north bridge and latency is much lower.
              Of course now that DDR2 (and even more DDR3) have higher latency, these advantages don't shine any more.

              To see it by yourself can look at the trace on the mother board. On regular mother board, the north bridge is in the middle and has trace both to the memory and to the CPU. The CPU is only linked to the northbridge.
              On athlon64 mother board, the traces go from the memory to the CPU. The north bridge is only connected to the CPU.

              In fact now that the AM2/2+/3 familiy has been declared upward compatible, you may start to see the same kind of compatibility that we had back with the Slot-1 connector which could be used with the first Pentium IIs all the way up to the latest Pentium II Tualatins (given one uses the correct slotket).
              And this what exactly this is all about : AMD *does want* a stable socket so they can attract potential chip makers that will be interested in making specialized coprocessors that will remain compatible thru all upgrades from AM2 to AM3.

    • Re: (Score:2, Insightful)

      by Lazarus_Bitmap ( 593726 )
      It might not be the performance champ, but they've also priced it cheaper. So it provides options, and options are always a good thing.

    • Re: (Score:2)

      by ivan256 ( 17499 )
      By that logic, Intel shouldn't have launched any chips at all between 2001 and 2005...

      Sometimes you release a product when the schedule dictates in order to keep your existing customers happy.

    • Manufacturing != R&D (Score:4, Insightful)

      by grimJester ( 890090 ) on Tuesday February 20, 2007 @10:27AM (#18081848)
      The percentage of chips able to run at a given frequency rises as they tweak the process to make manufacturing more efficient. This is not a new factory, process or design. They make them already. Why not sell them?

    • Re: (Score:2)

      by Svartalf ( 2997 )
      It's fab turn. In order to actually catch up to Intel and possibly pass them they need to go to 45nm
      process parts and at least up the on-die cache memory (The MAIN reason Core Duos outpace AMD's parts
      is due to process size differences and the lower on-die memory that results from the same...). A new
      architecture would seriously do it, but the other things are more likely to bring them something.

      If you're wondering, they taped these things out probably 6 or so months ago and they finally
      passed all the confo

    • Re: (Score:3, Insightful)

      by Chris Burke ( 6130 )
      What is the point of releasing a new iteration of an existing platform to bump up speed and still not catch up with the competitions products?

      Uh, if you're behind, then it is even more imperative that you continue releasing parts that keep you competitive. If you were in 2nd place in a stock car race, would you refrain from pulling a tight inside turn because it would only close the gap with 1st, not actually allow you to overtake?

      Wouldn't they have been better served re-routing this R&D effort/money i

    • Re: (Score:3, Interesting)

      by vandan ( 151516 )
      That's a pretty strange thing to say. It's like saying "What's the point of all the other car manufacturers competing with Porsche? They should quit releasing crappy cars and invest in R&D until they can produce a better car than Porsche".

      Of course, not everyone buys Porsches, just as not everyone buys Intel's top-of-the-line chip. AMD's chips are always better value. Always.

      I've never bought the most expensive CPU available. I always go for the best tradeoff between price and performance. It's called

  • Speed Bump? (Score:5, Funny)

    by blcamp ( 211756 ) on Tuesday February 20, 2007 @10:01AM (#18081530) Homepage

    AMD officially launched their next speed bump in the Athlon 64 product line
    "Speed bump"? You mean it's supposed to keep my computer slow(er)?

  • Unfortunately? (Score:4, Insightful)

    by goldspider ( 445116 ) on Tuesday February 20, 2007 @10:10AM (#18081658) Homepage
    "Unfortunately, in all the benchmarks seen here, it was still unable to catch Intel's Core 2 Duo E6700 chip at 2.66GHz."

    What's unfortunate about it? It's just a fact.

    • Re: (Score:2, Insightful)

      by Anonymous Coward
      Actually, wouldn't you want AMD to answer with something a bit more competitive, raising the bar for Intel again at these processor speeds? Competition is good for the consumer, big time...

    • Re: (Score:3, Interesting)

      by Machtyn ( 759119 )
      It's unfortunate to AMD and those who would support AMD. The customer gets a hotter, more power hungry processor, that is probably just as, if not more, expensive than a cooler, slower GHz rated Intel processor that outperforms the Athlon.

      • Re: (Score:3, Informative)

        by gmack ( 197796 )

        It's unfortunate to AMD and those who would support AMD. The customer gets a hotter, more power hungry processor, that is probably just as, if not more, expensive than a cooler, slower GHz rated Intel processor that outperforms the Athlon.

        The AMD processors are cheaper than the Intel chips and the difference becomes even more noticeable when you throw in the difference in motherboard costs. I was pricing this out the other week when I wanted to upgrade. I considered a core2 but then looked at the total

    • Re: (Score:2)

      by Junta ( 36770 )
      It is unfortunate obviously for AMD.

      But in the aggregate, it is unfortunate because the announcement of a new processor release suggests a hope of pushing the highest achievable performance up, and it is unfortunate that a new product does not fulfill such a hope. This is an industry-wide, vendor neutral way of expressing how it is unfortunate.

      But, ultimately, it's probably because the writer is an AMD fanboy and really would like to claim some victory over Intel above and beyond what existing products cou
      • What is stopping him. Just do what Apple Users have been doing for Years. Make their own benchmarks and show how it beats their competitors.
        • AMD still does maintain a stronger memory architecture, stream numbers out of AMD platform systems are much higher than comparable intel.

          Of course, this processor does nothing to widen that gap or anything..

    • Re: (Score:2)

      by Goaway ( 82658 )
      Facts cannot be unfortunate in your world?

    • Re: (Score:2)

      by Kjella ( 173770 )
      What's unfortunate about it? It's just a fact.

      I'd like them to one-up each other every day of the week. And the day they don't, I want it to be the one with deep pockets to move from dividends to R&D that is falling behind. Then again, the Core 2 Duo/Quad is pretty much exactly a result of this... Whenever Intel are seriously threatened, they push out damn good products. Last time around was when AMD was trying to reach the laptop market and they came up with Pentium M, also a damn good processor for it

  • Not a very helpful benchmark (Score:5, Insightful)

    by RailGunner ( 554645 ) * on Tuesday February 20, 2007 @10:21AM (#18081774) Journal
    From TFA: The OS used was Windows XP Pro SP2.
    A 32 bit OS. The real strength of the AMD 64 architecture is running in 64 bit mode - benchmarking this chip compared to other 64 bit architectures would be far more helpful than running a 32bit Sandra tests and Photoshop tests on it.

    Not a very helpful benchmark. I'd like to see these chips compared running 64 bit OS's - and compare the speed and throughput of applications like Apache, Oracle, PostgreSQL, MySQL, PHP / Perl scripting, and raw image processing - not Photoshop, where most of the time is spent waiting on the user to do something.

    • Re:Not a very helpful benchmark (Score:5, Insightful)

      by Pizza ( 87623 ) on Tuesday February 20, 2007 @10:33AM (#18081900) Homepage Journal
      Perhaps more to the point -- I'm curious about the raw integer performance of the AMD64 vs Core2 parts. A great deal of the extra performance that the Core2 parts demonstrate is due to their single-cycle SSE engines (which the upcoming AMD parts will match), but if your code doesn't use SSE (ie your typical server app) then all of these desktop-type benchmarks are worthless.

      I'd also love to see a native 64-bit (integer) benchmark as well, both with and without SSE-enabled tests.

      • Re: (Score:2)

        by spotter ( 5662 )
        do you think OfficeXP uses SSE? as it beat it in that benchmark too
      • Core2 smokes the AMD64 in integer performance.
        See SPECint scores at www.spec.org

        This is intuitive because the Core2 is wider than the Opteron - which translates into more IPC.

        It has a 4 wide issue, wider fetch bandwidth, instruction fusion which makes it effectively even wider than 4, deeper reorder buffers, out-of-order load/stores, hardware prefetching into the L1 cache. I could go on and on. The performance speaks for itself. Oh and the process lead (65nm Intel vs 90nm AMD) means Intel can give you do

    • Re: (Score:2, Interesting)

      by Ecuador ( 740021 )
      Exactly.
      Very few hardware journalists can set up tests that are useful for people who don't just load Win XP to play the latest
      FPS.
      My office, just like my university lab before, is fully 64-bit linux running custom programs. Anyway, some people (including myself) posted some sample benchmarks here recently (http://hardware.slashdot.org/comments.pl?sid=221 6 76&threshold=1&commentsort=0&mode=thread&cid=17960 946), but I would like to see some serious benchmarking from hardware sites. Anandtech
    • A 32 bit OS. The real strength of the AMD 64 architecture is running in 64 bit mode

      Same with Core 2 Duo, I might add.
    • Then try Anandtech's [anandtech.com] analysis and benchmarks with Windows Vista Ultimate 64-bit.

      Summary, the Core 2 Duos still win with the X2 6000+s almost making parity with the E6700 cpus with 64-bit apps.

  • It's all about the cache... (Score:5, Insightful)

    by Zebra_X ( 13249 ) * on Tuesday February 20, 2007 @10:27AM (#18081840)
    AMD has been skimping lately on its cache. I have a sneaking suspicion that the majority of AMD's current performance issues are related to cache and lack thereof.

    The Intel chips carry 4 to 8 Mb of cache. The thing about the Intel architecture is that the cache is shared across both or all 4 cores. In contrast the AMD chips have a dedicated *tiny* 1 MB cache for the consumer chips and 2mb per core on the high-end parts.

    With that said, the reality of dual core computing is that one core is used much more heavily than the other. In Intel's case this means that one core is basically given the entire cache for its use - a significant performance boost when running a few tasks. In AMD's case the idle cache is inaccessible to the heavily loaded core.

    The reason that makes me think that the cache is the current bottleneck is that the memory controller on the AMD chip is significantly faster than Intel's. Given that fact one would conclude that in non disk-bound applications that require large amounts of memory (games) the AMD chips would pull ahead. This is not the case. Of course there is more than just cache at play here but the fact that the Intel chips has 4 to 8 times more cache available to it has to make a fairly significant difference.

    Check out my AMD FX-70 at http://amd4x4.blogspot.com/ [blogspot.com]

    • Re: (Score:3, Informative)

      by Joe The Dragon ( 967727 )
      AMD's upcoming kl8 chip will have L3 that can be used by all cores also amd chips don't need a lot cache as they have a build in memory controller and a better chipset to cpu, cpu to cpu link. In a 2-4 cpu server the direct cpu to cpu links with out havening to use the chip set also reduce the need. I think that Intel may have to add cache to the ram controller / main chip set soon in there 4 cpu severs. Also AMD cpu let you have more then one cpu to chipset link in a system so you can have 2 chip set like
    • The reason why AMD is losing these comparisons is that the core2 cores are faster. They're a newer design, and they're very good. Cache helps, and it helps more on certain applications, but the core2 is simply faster.

      However, just because AMD is losing at the top end doesn't mean every AMD system loses to intel. At work, I'm still getting Athlon64 systems because you can get A64 3500+ 512MB systems with DVI and 3 year warranties for under $600. They're plenty fast for the users and DVI connection will d

    • Re: (Score:2, Insightful)

      by archen ( 447353 )
      Have you ever really looked at benchmarks with the larger cache sizes? When I was looking into upgrading a processor I found that there were some Athlon64's that were pretty much identical except for the cache sizes. And what do the benchmarks show? Pretty much a negligible speed increase, which I found to be pretty surprising actually. When looking at the price difference you're much better off putting that money towards RAM. Maybe some server applications can better take advantage of the cache, but i

    • Re:It's all about the cache... (Score:5, Informative)

      by GauteL ( 29207 ) on Tuesday February 20, 2007 @11:03AM (#18082354)
      It may be important for some things, true, but a significant reason for the performance of the Core 2 duo is that most of the benchmark applications are heavily optimised for SSE, and the core 2 duo executes 128-bit SSE instructions in one cycle, as opposed to two cycles with the Pentium IV and the AMD Athlon 64.

      This is massively important as the core 2 duo can then operate on four 32 bit floating point numbers in one clock cycle instead of two.

    • Re: (Score:3, Insightful)

      by Chris Burke ( 6130 )
      AMD has been skimping lately on its cache.

      Well, that's one way to look at it, another is that Intel has finally decided to unleash the flood gates on their own manufacturing and produce huge caches. Before the most recent generation of chips, Intel's desktop parts weren't sporting very big caches either. It was the Xeon MP and Itanium that were being granted gigantic caches -- I still maintain that Itanium's specfp score was mostly due to the amount of cache, since specfp 2000 should really be called spec
    • As others have pointed out, the AMD64 on-die mem controller means that memory access latency is significantly lower, making fetches from memory much faster and obviating the need for huge caches to make up for a cache miss. Again as has been pointed out, benches comparing the 512kB AMD64's to the 1MB AMD64's of the same clock and family typically show no difference, so cache is clearly not a bottleneck (except in the very few maths-eavy benches where the app fits entirely into a 1MB cache but not the 512k o
    • The Intel chips carry 4 to 8 Mb of cache.

      No. They have 2 or 4MB cache, with only the quad-core chips having 8MB.

  • silly but ... (Score:3, Interesting)

    by tomstdenis ( 446163 ) <tomstdenis AT gmail DOT com> on Tuesday February 20, 2007 @10:57AM (#18082242) Homepage
    Seems silly to release another 90nm part before the move to 65nm but keep in mind their are DIFFERENT LINES/TEAMS working at AMD. It's not like the production people working on retail 90nm parts are the same as the people testing new 65nm techniques.

    AMD is just trying to get as much non-idle time out of the fab as possible before they move everything to 65nm.

    It's the same reason why they make "el-cheapo sempron" parts and sell them AT A LOSS. It's better to lose a few bucks than a lot. And idle time in a fab costs a lot of money.

    Tom
    • I disagree. Idle time in any manufacturing operation can be viewed as an opportunity. For one thing, retooling to go to 65nm or 45nm or 30nm takes time, which might as well be spent earlier rather than later. Another consideration is maintenance, which is not retooling but repair and perhaps prep work for a future retooling effort if necessary.

      Cranking out processors just to give people something to do is not really minimizing costs, unless you're paying people to be idle. Even then, the energy and raw

      • Re:silly but ... (Score:5, Informative)

        by tomstdenis ( 446163 ) <tomstdenis AT gmail DOT com> on Tuesday February 20, 2007 @11:23AM (#18082620) Homepage
        Several things:

        1. They have several production lines. They make more than one CPU type at a time. They are capable of simultaneously producing/testing 65nm while making 90nm parts.

        2. Idle time in a fab is a KNOWN COST HAZARD. I'm not making this up. It costs money to keep rooms clean, pay the interest on the debt, etc

        3. Word on the street [when I was an AMD employee...] was the average processor cost ~60-80$ USD in raw materials/time/effort to make (assuming 100% yield). Yes, your opteron cost about the same to make (excluding yield problems) as that $50 sempron. So why make semprons if they lose money? Yes, I know I'm discounting yield which does contribute real cost to the processors. On the opteron side though, my take [personal op] was that most of the cost was to recoup the R&D not the production costs.

        Point is, both AMD and Intel produce low end parts that cost money. Even in the Celeron line which they call "mistakes" (e.g. parts with broken caches) that's not entirely true and is misleading. Even if you made a defective cache, it costs more money to just throw the die out, then to package it as a celeron and sell it at a loss.

        4. Intel cores are fast, but they're not the be-all. They still lack NUMA support which is handy in HPC environments (re: not your desktop). They're also not quite a strong in the ALU front (though from my crypto benchmarks are VERY VERY close).

        I'm by no means an AMD fanboi. Hell, my desktop is a core2. But I still love my 2-way Opteron workstation and get it to do things that run circles around the core2 (like hosting 15 engineers running simulations/verifications/etc).

        Buy what you need, not what some lame commercial on TV tells you. For many, the core2 is the best buy. It's fast, wicked low power and the cost isn't bad. For others, AMD is the better buy (cheaper) or simply more powerful (opterons).

        Tom
        • Idle time in a fab is no less an issue for AMD than it is for a major pharmaceutical that needs to produce as much anti-cholesterol drug as possible before patent protection runs out. However, it is far easier to keep an idle room clean than it is to maintain a working production room's cleanliness, and costs for a line still have to factor into the overall account the cost of supply chain. In this respect, CPU manufacture is no different than any other supply chain manufacturer.

          -BA

          • Re: (Score:3, Insightful)

            by tomstdenis ( 446163 )
            You're missing a huge point. Every day that they're not making anything is a day of paying 1000s of wages, taxes, utilities and interest on the loans they take out to pay for the equipment. It's cheaper to make something, anything, that you can sell [and earn some low hanging customers] than so sit around doing nothing.

            Think about it, you have a pile of costs that don't go away. You can't just lay off/rehire fab technicians on a whim. These costs don't just go away because demand for Opterons is lower o
            • I'm not missing the point, and I think you need to understand how a supply chain works before continuing the debate. There is a price point that must be met before making something is more worthwhile than sitting idle. Making "something" is definitely not better than sitting idle in all cases.

              -BA

              • Ok, let me explain this so you understand. They LOSE MONEY on the low end processors. That's not a lie, it's not a myth, it's the fucking truth. They make them because it fulfills a need and keeps the fab operating. But part of the equation that is missing here is how much money goes into making the die vs. the chip. If a processor costs $50 to make, $35 of which is the die and $15 is packaging, then throwing out a die that doesn't quite make the cut is stupid.

                You already spent the $35, so selling the p

  • You make incorrect assumptions (Score:4, Interesting)

    by lcnxw ( 743741 ) on Tuesday February 20, 2007 @11:35AM (#18082762) Homepage
    The design and development of a processor has improved vastly since the days of borked multipliers. There are standard benchmark tests that engineers use to rate their designs in-house. If AMD chooses to go with smaller caches, I would imagine they have very good reasons.

    Perhaps in order to keep good performance when communicating between caches they need to keep the number of memory addresses low so that the overhead stays low. They decided that separate caches was a better model, and they currently have to maximize performance with this design.

    AMD might have favored their server market when choosing this design and separate cache works better for server machines. They may need to refine their architecture for the desktop market. Don't be so quick to accuse AMD of making cache mistakes without doing the math for find the theoretical best solution.
  • To comprise is to be composed of. The chip comprises 227 million transistors, or is composed of 227 million transistors.

    This is one of the rare cases where a common misuse isn't just a gradual development of language; it actually reverses the sense of a word, replacing the relation of the whole to the parts with the relation of the parts to the whole.
  • Comment removed based on user account deletion
    • Good point.

      Are we comparing AMD 2 cores with Intel 4 cores here?

      If so then I humbly suggest that the test might just be a tad skewed.
      • Not if the Intel CPU's are still price competitive, because that's all that matters to your end user.

        But no, these Core 2 Duo's aren't quad processors -- in that case they'd be talking of QX6800's, not X6800's.

    • Re: (Score:3, Informative)

      by jalefkowit ( 101585 )
      There's four basic AMD desktop chips these days. Here they are in order of performance, fastest first:
      • Athlon 64 X2 [amd.com] is their dual-core offering, available for Socket AM2 [wikipedia.org] and Socket 939 [wikipedia.org] motherboards.
      • Athlon 64 FX [amd.com] is their high-end single-core offering, available for Socket AM2, Socket 939, Socket 940 [wikipedia.org], and Socket F [wikipedia.org] (server) motherboards.
      • Athlon 64 [amd.com] is their mid-range single-core offering, available for Socket AM2, Socket 939, and Socket 754 [wikipedia.org] motherboards.
      • Sempron [amd.com] is their low-end ("value") single
  • Critical question here is whose compiler was used to compile the benchmarking programs??? and whether the benchmarks themselves were specifically compiled to take advantage of the instruction sets...

    There's only one truly independent test available... and that's how long it takes to emerge a default Gentoo install with the compile options set to match the respective processors. Everything else should be as identical as possible.
  • The only conclusion you can see from these benchmarks is that quad is better than dual is better than single. Between different CPU's of identical number of cores, the difference is negligible. They're all limited by memory bandwidth.

Slashdot Top Deals

Think of it! With VLSI we can pack 100 ENIACs in 1 sq. cm.!

Close