Forgot your password?
typodupeerror
AMD Hardware Technology

First Round of AMD Athlon 64 Reviews In 248

Posted by CowboyNeal
from the comparing-notes dept.
wrinkledshirt writes "Here's a bunch of AMD Athlon 64 reviews, courtesy of 8Dimensional." AcesHardware and HardOCP match the Athlon 64 line against the Pentium 4 Extreme Edition. amdmb, FiringSquad, and SharkyExtreme take a closer look at the FX-51. AthlonXP and PCStats have glowing reviews of the chips. Digit-Life compares the new Athlon 64 with Opteron and a Pentium 4. LegitReviews and Overclockers.com.au also both have succinct reviews of the FX-51. Overall the reviews speak very highly of the Athlon 64 and the FX version of the chip, with the only downside being the cost, especially of the FX chip.
This discussion has been archived. No new comments can be posted.

First Round of AMD Athlon 64 Reviews In

Comments Filter:
  • Isn't this a major benefit of more bits? You can play with the concept of mapping disk like memory, since you have plenty of address space for any size disk. The Athlon 64 might be a kludge, but introducing more bits is a good thing.
    • Given that there is no way to fit more then 8GB on an ATX MB and that requires spending 700$ for each of the eight 2GB DIMMs I'd say that the extra "bitness" is overhyped. You will not see consumer systems with more then 4GB any time soon.
      • by Anonymous Coward
        If you have lot of RAM or so programs will tend to exhaust the 32-bit address space before actually consuming all the physical memory. One example is that creating a thread might represent the reservation of 64KB of address space for its stack, while only 8KB of physical memory is actually committed.

        A 64-bit address space is probably a good thing once a program is allocating 2GB or more of address space.

        • Yes, yes it is.

          Some apps require large blocks of contiguous memory - and with only 2GB of address space available, you can actually run into address space fragmentation problems long before you run out of physical memory. There simply isn't a large enough span of addresses available to map the memory into.

          Other things compete for address space too. System DLLs map themselves into various places, leaving too-small gaps between them. Threads reserve 1 MB each, for the stack grow. Some PCI boards (e.g. HiD

      • That's true, but new form factors are on the horizion and extra bits now means that when form factors do more memory, you won't have to patch the software and redesign CPU's again.
      • by ianezz (31449) on Saturday September 27, 2003 @03:02PM (#7073270) Homepage
        You will not see consumer systems with more then 4GB any time soon

        The point about mmap(2) is to let the system (the VM subsystem of the kernel) manage the caching for the userland processes using it, avoiding extra copies to/from buffers in userland and eliminating in several cases the need for custom caching code (processes don't have to worry about data being available in RAM: the kernel automatically takes care of that when needed).

        You don't need gobs of memory to do this, but in order to work on large amount of data you need a large address space, which is what 64 bit architectures provide. Of course, the more physical memory you have, the less the kernel has to swap pages in and out, but the main point is not that.

        A little example to clarify: in order to keep things simple (instead of needing two intermixed caching systems, one for the VM and one for disk accesses), the Hurd just mmaps the whole partition. This means that the maximum size of a partition has an upper limit given by the size of the addressing space, which is 4GB on 32bit architectures (actually less, since in that address space you have to keep also the code that uses the mmapped data, so it's more like 2GB/3GB). A 64bit architecture comes very handy here, given the current size of hard disks.

        • The point about mmap(2) is to let the system (the VM subsystem of the kernel) manage the caching for the userland processes using it, avoiding extra copies to/from buffers in userland and eliminating in several cases the need for custom caching code (processes don't have to worry about data being available in RAM: the kernel automatically takes care of that when needed).

          Most applications, however, do not take full advantage of mmap. Most applications don't even need to take advantage of mmap. Those tha

      • The practical memory limited for a consumer 32-bit machine is not 4GB but 2GB. The OSs needs some of the virtual address space (you can cut it down, but that decreases peformance because you have to jump through hoops to map the whole buffer cache) and hardware needs some as well. All told, up to a gig of your virtual address space is gone before you even talk about memory. Since 3GB is an odd amount of memory for a machine, 2GB is the limit you usually see in motherboards. 2GB is a pretty close limit. Soon
      • The Tyan Thunder K8W [gamepc.com] mobo is the dual Opteron board to have. Perhaps not for "consumers", but certainly as a workstation board it's unparalleled.

        8 DIMM slots (4 attached to each CPU, unlike every other workstation board out there) means 8 GB using cheaper 1 GB DIMMs, or 16 GB max. It also means up to 10 or 12 GB/s of total bandwidth :-) No other dually motherboard I have seen offers both AGP and memory connected to each CPU.

        Add to that an AGP Pro slot, a few PCI-X slots (100 MHz and 133 MHz), 4-way SAT

  • Hmm.. (Score:4, Funny)

    by darkov (261309) on Saturday September 27, 2003 @02:03PM (#7072966)
    Is it possible to have a first round of reviews after an earlier story about Athlon 64 reviews?

    fp
    • Yes it is:

      1) The earlier story was round 0, or
      2) the earlier story was the pre-match bragging part, or
      3) some other reason (not profit, though)

  • Wooah! (Score:1, Funny)

    by Sir Haxalot (693401)
    Remember bigger numbers =! better.
    Think N64 and Playstation.
    • by Gldm (600518)
      Do the math! (Jaguar)
    • Re:Wooah! (Score:2, Funny)

      by Anonymous Coward
      Correction =! != !=
  • Enough! Leave me alone with my abacus...

    Um, anyone knows how many bits can an abacus counts to?

  • Hang on.. (Score:4, Insightful)

    by madprof (4723) on Saturday September 27, 2003 @02:06PM (#7072986)
    Give it a bit more time. The motherboards can still be tweaked to get more performance out of the system. I think we should wait a bit before really making decisions, and get in a whole load more real world testing too.
    Benchmarks are not always entirely, although often can be, illuminating.
    • The FX will also probably perform a bit better once it's freed from having to use registered memory. Wait for the 939pin version.
    • Re:Hang on.. (Score:3, Informative)

      by captainclever (568610)
      I had to buy a new box as a pg database server recently..
      almost bought a dual opteron, but chickened out and went for a Xeon instead.

      the suse distro that supports it is still a bit shaky and i wanted to wait for some good bench results.

      maybe my next server will be 64bit.. :)
      • Re:Hang on.. (Score:5, Informative)

        by FatherOfONe (515801) on Saturday September 27, 2003 @02:46PM (#7073191)
        We have a dual Opteron 1.8GH and a 2GH on order. We found that with RedHat and Oracle the Opteron in 32 bit mode beat the crap out of a quad Xeon for the stuff we do.

        Just an F.Y.I. Again this was with 32 bit code. I tried the RedHat BETA and it wouldn't even boot up without locking.

        So given that Oracle cost us over $20k a processor, we saved over 40 grand!

        • So given that Oracle cost us over $20k a processor, we saved over 40 grand!

          I didn't know anyone actually bought processor licenses. I'm surprised that a dual Opteron beats a quad Xeon. All things being equal (i.e., 3 GHz Xeon compared to 2 GHz Opteron) I would have thought that hyper threading would give the advantage to the Xeon.

          • Re:Hang on.. (Score:5, Informative)

            by WoTG (610710) on Saturday September 27, 2003 @04:30PM (#7073800) Homepage Journal
            I don't have any firsthand experience with this stuff, but I have read a lot of the reviews and whatnot. I gather that the difference is in scaling to 2 way (and 4 way) boxes. While a single 3GHz Xeon and a single Opteron are pretty competitive, the Opterons are built for dual and quad processor work - especially with memory intensive applications.

            Each Opteron has it's own memory channels in multiprocessor boxes. All memory is still shared throughout the system, it's just that there is more total memory bandwidth to go around as you add Opterons. In comparison, Xeon systems have the same amount of system wide memory bandwidth from 1 CPU all the way to 4 CPU's. The net result is that in many cases a second Opteron processor nets a gain 80% or more performance - which is a LOT better scaling than Xeons. This will probably be even more evident in future comparative reviews of quad CPU boxes since the Xeons will be sputtering on memory bandwidth fumes (relative to the Opterons).
    • Re:Hang on.. (Score:2, Interesting)

      by SD-VI (688382)
      Actually, I doubt the motherboards can be tweaked all that much. You see, usually any performance increase from motherboard tweaking comes from memory controller optimization, and the ClawHammer (Athlon 64) and SledgeHammer (Athlon 64 FX / Opteron) have an on-die memory controller. I'd look more to code optimization than motherboard revisions if I were you. This also means that the performance difference between chipsets is tiny. nForce3 will probably be very popular just because of nForce2, even though it
      • Re:Hang on.. (Score:3, Interesting)

        by Jeff DeMaagd (2015)
        You know, I can't disagree. Moving the memory controller must have greatly simplified the chipset design. Even the multiprocessing is greatly simplified as well as the CPUs are the ones that handle that, one may be able to use the exact same chipset for both dual and single chip board designs, possibly even quad designs. The down side is that a lot of dual CPU and definitely all new quad CPU systems need greater I/O, that which the single CPU designs rarely need, such as 64 bit PCI slots, at 66, 100 and
    • -1 Redundant.

      What you say is assumed for the first set of reviews. How long do you think we should wait? The answer is that we shouldn't, but we should be open to revising our opinion as new information and better testing arrives.

  • Don't forget Anandtech [anandtech.com].
  • It was either the Inquirer or The Register that had an interesting article saying that these CPUs (which are MP Xeons) still have their multi-cpu support enabled, thus saving astute customers thousands of dollars over their full-priced, $3,900 Xeon counterparts.

    • by MarcoAtWork (28889) on Saturday September 27, 2003 @02:24PM (#7073074)
      while it's very likely that this is the case for the reviewing samples (probably rushed out as fast as possible, I can't believe that Intel had plans for the EE all along given that there were 0 leaks) given that this is a paper launch by the time the EE will actually be available to customers I'm sure the MP support will disappear.
    • by Unoriginal Nick (620805) on Saturday September 27, 2003 @02:25PM (#7073086)
      It was either the Inquirer or The Register that had an interesting article saying that these CPUs (which are MP Xeons) still have their multi-cpu support enabled, thus saving astute customers thousands of dollars over their full-priced, $3,900 Xeon counterparts.

      The multi-cpu support may or may not be still enabled, but the P4 EE has a different pin count than the XeonMP, so you wouldn't be able to use it in the XeonMP boards anyway.

      • Makes ya wonder whether a company such as Powerleap [powerleap.com] might come out with a CPU adapter to support it. For a long time the Athlon MP series offered the only affordable SMP solution, especially if like me you found a pair of Athlon XPs which worked happily in SMP mode.
    • Yes, now all you need is some multi-socket 478 motherboards.... seen any?
  • The OS is 64 but most of the games/applications are 32 bits ?????
  • by Tumbleweed (3706) on Saturday September 27, 2003 @02:13PM (#7073017)
    The socket format will be changing soon, and once the upcoming changes happen, things will be much better. You'll then be able to use non-ECC memory, and the motherboards will be less expensive.

    Until then, yeah, the FX is freaking fast, but waaaaay overpriced, so don't bother.
    • Actually once you start having lots of RAM, ECC may become almost compulsory.

      If you have 128MB the chances of one bit becoming incorrect is rare. If you have 4GB it starts becoming more likely, also manufacturers may be starting to get close to the safe limits of their process with the high capacity chips.

      If ECC becomes more common it'll become more affordable. It won't be as cheap in the near future.

      The big problem for AMD I see is, even if you want to buy the stuff you can't.

      Once I stop seeing "preboo
  • by Anonymous Coward on Saturday September 27, 2003 @02:18PM (#7073037)
    Are avalible from SuSE [suse.com], Gentoo [gentoo.org], and Debian [jukie.net]!

    SO if you are complaining "theres no 64 bit os yet", stop complaining, leave the evil empire behind and see the REAL power of opensource.
  • welcome a plethora of processor reviewers
  • by _|()|\| (159991) on Saturday September 27, 2003 @02:38PM (#7073145)
    I haven't made all the rounds, but it seems like everyone is using the same two motherboards: Asus SK8N and MSI 8KT. I really like the looks of Monarch's Hornet 64 [monarchcomputer.com], with a uATX Gigabyte GA-K8VT800M. I'd like to see some reviews first, especially regarding chipset support under Linux. I'd also like to hear more about video drivers. I've heard that NVIDIA's drivers need some work. (Does ATI even have any?)

    We've got a couple of Opterons at work, one for 32-bit compatibility testing, and another for the AMD64 port. It's pretty cool to see this in Python on SuSE Linux 8.2 beta:

    >>> type( 9223372036854775807 )
    <type 'int'>
    SuSE Linux 9.0 for AMD64 is supposed to ship next month. Hopefully, it will be a little cheaper than RHEL 3.0 for AMD64, which will be more than twice the price of RHEL 2.1 for x86!
  • Slot, Socket, Boardfrequency, Memory Type, CPU Bitcount and CPU Class...
    Bit by bit (no pun intended) vendors are establishing a true real life randomness of standards. A shure sign that computers are becoming a comodity. Soon we'll see the same with operating systems.
  • Proper benchmarks (Score:3, Insightful)

    by NitroPye (594566) <coleman@n[ ]oy.com ['itr' in gap]> on Saturday September 27, 2003 @02:39PM (#7073155)
    Untill I see proper benchmarks done with applications built for 64bit chips I am going to go pffffft.

    I am still blown away that the FX did better then then 3.2 P4.

    Proper benchmarks include not using a 64bit beta stupid o/s like windows, a properly optimized linux (suse 64 or gentoo) and applications built for the chip. Openoffice, kde and kde apps, mozilla, some miscellaneous 3d engines running some impresive demos, maybe tenebrae quake. Tenebrae quake is great being that its open source and takes a huge amount of gfx and proc power.

  • What's this "organic" stuff? Apparently, plastic IC packaging (as opposed to ceramic) is now referred to as "organic".
    • by cosmo7 (325616) on Saturday September 27, 2003 @03:01PM (#7073261) Homepage
      Organic integrated circuits are produced without using fertilizer or pesticide.
    • The word organic means "composed of carbon, hydrogen, oxygen, and nitrogen". Petroleum & plastic fits that definition.

      There may be some confusion with a recently-invented marketing term, "Organic", which applies to agricultural products grown with sophistication below some arbitrarily-defined level.

      In reality, any food a human could survive eating is organic, by the scientific definition.
  • Most Only 32 Bits... (Score:5, Informative)

    by MBCook (132727) <foobarsoft@foobarsoft.com> on Saturday September 27, 2003 @02:52PM (#7073220) Homepage
    The reviews are basically all 32 bit, which is a shame. Linux is out there. I've only seen one review that did anything really 64 bit (running win32 programs on win64 doesn't count). The only 64 bit test they did under Linux was MP3 encoding. The test was the Athlon 64 running a 32 bit version they compiled of the MP3 encoder vs running a 64 bit version of the same program. The "bitness" was the ONLY thing that was changed.

    The results? The 64 bit version took nearly HALF THE TIME of the 32 bit version. This is the kind of thing we have to look forward to in some things (MP3s, video encoding, etc).

    The Athlon 64 is fast in 32 bit mode, and can beat a P4 many times. But when the 64 bit code comes along, the P4 will be taking one hell of a beating.

    PS: Sorry I don't remember which review had this test. I don't have time to go hunting for it right now.

    • by MBCook (132727) <foobarsoft@foobarsoft.com> on Saturday September 27, 2003 @02:55PM (#7073231) Homepage
      Found it! It was Anandtech [anandtech.com]. Check out the bottom of the 32 bit vs 64 bit [anandtech.com] page of the review.
    • by Sangui5 (12317) on Saturday September 27, 2003 @03:56PM (#7073577)

      The test was the Athlon 64 running a 32 bit version they compiled of the MP3 encoder vs running a 64 bit version of the same program. The "bitness" was the ONLY thing that was changed.

      Not really. While the "bitness" changes, what also changes is the number of registers visible to the compiler. The x86 ISA has been dealing with internal register rename as a nasty hack to deal with a sever shortage of programmer visible registers for a long time. This goes to show that the compiler is much smarter about register allocation than a hardware renamer can ever be. I'm interested in seeing the performance of common multimedia applications once hand-written core loops are available.

      And a note to those who are pointing to improved SSE2 support as the reason for the performance gain: they are comparing an AMD64 in 32 bit mode vs one in 64 bit mode. Unless GCC is being bass-ackwards, the SSE2 support should be benefiting the 32 bit mode as well. It appears that the only variables in this benchmark are the 64 bit math and the additional registers.

    • a ~30% increase in speed isn't "half the time", but it is a nice performance increase, and being only one test, it isn't enough to work from, and a more valid test would compare it to P4 running Linux too. Given that media encoding under ia32 is one of the weaker points of the Hammer architecture, then that means it can catch up and possibly beat the P4 here.

      In a way it is a shame that the alleged hardware review sites really don't check out Linux in part of their testing. Linux stands to score a major p
  • 64 bit resources (Score:2, Insightful)

    by JusTyler (707210)
    I've been struggling to find any good 64 bit resources for Linux.

    Basically, I want to know about all the 64 bit versions of major apps and systems, like MySQL, Perl, and so on. I know Perl is in 64 bit, because you can compile it to be, but what about stuff like MySQL, Apache, TomCat...

    Post your best 'going up to 64 bits on Linux/FreeBSD/elcheapo UNIX' resources here, and attract some karma :-)
    • You just recompile any app to "be 64-bit". If you want a pure 64-bit system you can do it, as long as you can also recompile all your drivers.
      • You just recompile any app to "be 64-bit".

        That may not always be true. It's possible, especially with the C/C++ language, that a programmer will have inadvertently built in assumptions about data types.

        For example, a program my try to read 4 bytes from a disk file into an "int". On a common modern CPU, that works fine. With an Athlon64 it'd leave half of the integer unfilled.
  • Until now, when only the Opteron was available for AMD64 support, there was very little effort for free OS support. There were efforts by RH and Suse for expensive enterprise server OS's..

    Now that the low end 64 bit chip is out, what is the best Linux Distro that is freely available, or at least cheap??
  • I know this is a CPU benchmark, but look at the GFX cards they use.

    Every benchmark shows the ATI 9800 Pro to be faster than the FX 5900 Ultra, in every benchmarked, (3dmark2001 included) except 3DMark2003.

    Using 3DMark2003 while informative, shows a negative performance compared to all other benchmarks. The raw FPS scores prove that 3DMark2003 is not giving true proformance of games out today.

    Be nice when HL2/Doom3 is out, we can compare and see if 3DMark2003 is providing true numbers for features not ou
  • by sundling (92926)
    http://www.aceshardware.com/read.jsp?id=60000253
    has a cooling technology that allowed them to overclock to a 2.8 Ghz Athlon FX. It was pretty impressive stuff, especially how well Age of Mythology did, even against the non-shipping P4 Emergency Edition.

    I can't wait until theres 64 bit games for this sort of thing. Of course, the first to be released will be Unreal Tournament. Oh yahh!! I know I'll own at least one of the AMD64 computers within the next year!

    I'd love to have a new AMD64 for my appli
  • Ugh... (Score:2, Interesting)

    by wicka_wicka (679279)
    Doesn't anyone realize that the Athlon 64 isn't going to be anything special until they start testing on a 64-bit platform???
  • Increased performance NOT because of 64 bitness

    From the articles, it can be clearly understood that the increased performance of of these new processors comes not from the 64 bit data bus, but from other technologies:

    1) increased number of general purpose registers from 8 to 16.

    2) dedicated bus to memory for each CPU; allows for much better CPU scaling.

    3) deeper pipelines

    4) improved design

    These things could be easily achieved with 32-bit CPUs. The fact that more than 4 GB of memory will be addressable

No hardware designer should be allowed to produce any piece of hardware until three software guys have signed off for it. -- Andy Tanenbaum

Working...