Athlon 64 Memory: Rewriting the Rules

by Wesley Fink on 10/1/2004 12:45 AM EST


Back to Article

  • xeoph - Wednesday, March 09, 2005 - link

    I also wanted to mention it in this artical if it hasnt been already.

    The pricing above the ultra X is for Ultra Platinum. The link is a little misleading because it doesnt give the exact tech specs.

    simply a mistype. m-400-512x2glx1gb3200dc is what your looking for, little X between gl and 1gb.
  • mervine - Sunday, October 10, 2004 - link

    hi Wesley
    You mention the newer Samsung TCCD chips overclock better on AMD64 than the older ones, How long have these new chips been avaliable?
    The reason I ask is I live in europe and stocks over here might not be as new as in the states. I'm looking to buy some ocz rev2 but is there any way to tell if its using old or new revision TCCD?
  • PrinceXizor - Tuesday, October 05, 2004 - link


    I appreciate your point. I suppose I would have been vaguer in my statements than you were, that is my only point (probably a matter of subjective opinion of course). Something like...

    "In the course of testing, it was discovered that we were being adversely limited by the power supply used in our testing rig. Switching to a higher wattage power supply unlocked the potential of many of these DIMMS. While the actual problem with the original PSU requires more analysis time than we had for this review (lower wattage? lower rail amperage? lower rail tolernaces?), hard-core overclockers should carefully inspect the specifications on their PSU's if they plan on maxing out the OC on their memory modules. Switching power supplies, as we did, may help you to reach the overclocks attained in this review."

    Something like that is all I would have liked to see, instead of a blanket recommendation for a higher wattage PSU. That was my point.

    All this being said, I still enjoyed the review! I look forward to more memory module reviews in the future.

  • jynxycat - Tuesday, October 05, 2004 - link

    good review, i was on the edge between the PQI and Crucial sticks, not really knowing what to expect from the PQI on an AMD setup.

    finally a full review of what the respective sticks will achieve on the AMD platform, very much appreciated.
  • Blappo - Sunday, October 03, 2004 - link

    That is true. It might be interesting to see a comparison of PSUs to see which one would give the highest overclock. I would guess that the PSUs with high ratings would perform well, but there might be some lower power PSUs delivering good quality power that perform well, or some high output PSUs that perform poorly because of low quality power.
  • darkwaffle - Sunday, October 03, 2004 - link

    I'm not sure how much of a direct comparison you could make between the 3 PowerStream models, because according to the +5 rail differs on the 420 and 470, and the +12 rail is different on all of them.
  • Blappo - Saturday, October 02, 2004 - link

    I would be curious if using either the 420W or 470W OCZ PowerStream models would change the amount the memory could be overclocked? Since they are in the same model line then any differences when OC would be attributed to the amount of power available, and not the quality of power available. Reply
  • LocutusX - Saturday, October 02, 2004 - link

    I've got the one Wesley mentioned. It's the Enermax EG465P-VE - the *Q4 2001* model. That same model which is *currently* available in stores has slightly different specs. Anyways, I think I'll keep in mind a possible upgrade to the Powerstream since that would probably help my OC situation as well.
  • KingofFah - Saturday, October 02, 2004 - link

    (This is more directed to #20)
    As Wesley said:
    +3.3V - 38A
    +5V - 44A
    +12V - 20A
    -5V - 2A
    -12V - 1A
    +5Vsb - 2.2A
    are the specs of the 465 they used.

    Considering they switched to 520W OCZ Powerstream:
    +3.3V - 28A
    +5V - 40A
    +12V - 33A
    -5V - .5A
    -12V - .5A
    +5Vsb - 2.0A

    Even though there is a drop in amperage on the 3.3 and 5v rails, there is a great increase in the 12v. CPU, GPU, HDDs, Optical drives, fans all get there power off of the 12v line.

    Since w = v x a, if you want a good power supply, it almost certainly will have a lot of watts on it, but you only find good amperage on high quality models... So maybe you can see why they said +500 quality PSU??

    My question is: Which rail does the memory draw from? Is it still the 5v?
  • Marlin1975 - Saturday, October 02, 2004 - link

    Wesley Fink, since it has been shown many times that more and more current systems need better power. Why not do a review of power supplys. Everything form top dollar name brand ones to cheaper high "watt" ones to see if they put out what they say at each rail and if they help or hurt a system when doing normal work to over clocking?
    I know power supplys are one of the most over looked items and I am building a HIGH power system right now and can not find any real reviews other then other people saying use or don;t use brand X

  • Bugler - Friday, October 01, 2004 - link

    I ordered from Newegg today. I did not see a place on their site for just a 512mb stick. I ordered the 1gb kit.
  • Bugler - Friday, October 01, 2004 - link

  • saechaka - Friday, October 01, 2004 - link

    hey bugler where are you ordering your ocz platinum rev. 2 from? is there any way to find a place where you can order 1 512mb stick only? Reply
  • darkwaffle - Friday, October 01, 2004 - link

    I'm curious, for a socket754 user, is there really any reason why we couldn't (generally) follow these results? I realize that some of the overclocks may not be able to be achieved, but is it safe to say that the modules that perform highly on s939 will also perform highly also on s754 (In comparison to the other modules)? Reply
  • Wesley Fink - Friday, October 01, 2004 - link

    #32 -
    Corsair and Kingston DDR400 2-2-2 were both included in our recent 2-2-2 roundup. Frankly we did test both Kingston and Corsair in the early going for these tests and both did quite poorly on the Athlon 64 test bed compared to other recent TCCD modules.

    However, the Kingston and Corsair were early TCCD dimms and we are confident more recent dimms from these two major manufacturers would perform more like the OCZ, Geil, and G. Skill. Unfortunately we didn't have those more recent dimms to test, and we felt reporting what we had found would have been very unfair to Corsair and Kingston, who both produce excellent memory products.
  • ImJacksAmygdala - Friday, October 01, 2004 - link

    I'm sure OCZ and Geil are great memory for A64, but all I got out of that article was buy OCZ memory, buy Geil Memory, o ya and buy a top of the line OCZ power supply....

    What about Corsair and Kingston? How do they compete?

    This site is smothered with advertisement. Why make it so obvious in the articles? Thanks for the article though...
  • einsig - Friday, October 01, 2004 - link

    And re: the robust +12v rails. They are most crucial for A64 overlocking. I have an Enermax with 31A on two +12v rails and it makes the world of difference. Reply
  • einsig - Friday, October 01, 2004 - link

    One thing that really needs to be stated is that Clawhammer cores automatically set a command rate of 2T at speeds from DDR400 and up. you need to use A64 Tweaker to make the change (can be loaded at startup). I have a Clawhammer 3400+ that has been run on an ASUS K8V Deluxe and MSI K8N Neo Platinum. I was running XMS 2x512 3200LL Corsair, but it didn't want to OC even on the K8N (nothing OCs on the K8V because of the chipset). I now run Crucial ballistix PC4000 and it is incredible, however (as the article states) the command rate of 1T is really ideal. They should just tell people how to set that if they have a Clawhammer. Reply
  • Shinei - Friday, October 01, 2004 - link

    What's really exciting is that the two DDR600 overclockers are board-limited before they top out in speed (K8N Neo2s go to 300MHz on the RAM). I'd like to see what the RAM could do on a more extreme overclocking board, since it seems like these new RAM chips are capable of pushing on to DDR667 or even higher...
    With that said, I agree that the prices for this stuff is getting ridonculous. Cheaper RAM means more sales and increased usage of that 8 exobyte storage capacity the Athlon 64 has. ;)
  • Bugler - Friday, October 01, 2004 - link

    Thank you AnandTech. I have been waiting for greater clarification since you last recommended OCZ 3700 enhanced bandwidth ram and none could be found. I was balancing between that and ballistix. After today's review, I ordered OCZ Plat, rev 2 for the system I am putting together.

    Now if we could get some reviews and testing of the newer 90 speed AMD processors, hint, hint...

    I appreciate this site very much.
  • Wesley Fink - Friday, October 01, 2004 - link

    #22, #25 and others -

    The 465W power supply that was not up to the job had the following specifications:

    +3.3V - 38A
    +5V - 44A
    +12V - 20A
    -5V - 2A
    -12V - 1A
    +5Vsb - 2.2A

    I would never have had any reason to suspect issues with this expensive major brand PS based on those specs, but in fact it turned out to be the limiting factor in overclocking the memory.

    This issue requires more investigation as there are many possible reasons for my experience, but I felt an obligation to let readers know what we found in our testing. However, it is not fair to name brands without much more evidence.
  • Blappo - Friday, October 01, 2004 - link

    Hopefully the value memory roundup will also contain some generic memory as well. Knowing which premium memory is the faster is good, but I want to know how much of a difference memory makes on system performance. I don't want to spend 50% more for only 5% more performance. That money could be better spent somewhere else.

    I'm glad that AnandTech did this article since everywhere else they only benchmark memory on Intel systems.
  • Blappo - Friday, October 01, 2004 - link

    PrinceXizor - I agree completely. However, with higher wattage PSU most of the extra power comes on the 12V rail. I think that is why nVidia is recommending "monster" PSU for their high-end graphics cards because it is easier to tell people to buy a 480W PSU rather than a PSU that can deliver 22A (or whatever) on the 12V rail. Reply
  • Spearhawk - Friday, October 01, 2004 - link

    15: 20A? You're kiding right? My oven are at 20A. Reply
  • decptt - Friday, October 01, 2004 - link

    My Ballistix performance
    "Dec" copyright since Duron600@1000
    Athlon64 Mobile 3200+ Rev.SH7-CG
    AMA3200BEX5AR-ClawHammer L2:1MB
    10x250 vid:1.5v+113%(~1.70v)
    Idle@36C Load(Prime95)@43C
    [10x255 worked @~1.97v Loaded(Prime95)@55C]

    Crucial Ballistix PC3200 512MBx2 vcore:2.70v
    Ratio = 1:1 ; TCl:2.5-Trcd:3-Tras:5-Trp:3 1T
    ATI AIW 9600XT 128MB BUS:75 vcore:1.6v
    DFI Lanparty UT NF3-250 Rev.A00 Bios.9/14(Beta)
    LDT/FSB@ 4X
    Thermalright XP-90 +Panaflo-H1B-92(FBA09A12H)
    Seagate ST3120026A
    DVD model DD0401

    Scroll(Tested@2.50GHz on Sep 30, 2004)
    CPU::ALU 11513 FPU 3954 iSSE2 5156
    Media:: Int 23895 Float 2562
    Ram:: Int 3780 MB/s Float 3779 MB/s
    3DMark03V3.4.0--1751 :(
    Super PI 1M 35s
    Super PI 2M 1M23s
  • quanta - Friday, October 01, 2004 - link

    Speaking of robust 12V rails, Enermax seems to be the only one make power supplies that can provide more +12V juice than most enthusists ever need, even for models without splitting +12V lines. Reply
  • AlphaFox - Friday, October 01, 2004 - link

    where did the power supply talk come from?? Reply
  • PrinceXizor - Friday, October 01, 2004 - link

    Just a further comment on the PSU "recommendation". I really think that the major tech sites are doing a disservice to the community when they keep recommending higher and higher total wattage PSU's. That is NOT the issue! As has already been pointed out, the key component is to analyze what voltage rail is is not providing enough juice. Just as clock speed is a poor indicator of processor performance, total wattage is a poor indicator of PSU performance. Considering that a major computer rig will rarely if EVER draw more than 250W of actual power, the key metrics for a PSU are the actual amperages on the various rails, particularly the robustness of the 12V rails. (Some newer PSU's are providing dual 12V rails for just this reason). So, does an enthusiast overclocking their rig need a 500 Watt "monster" or do they need a robust 12V line and tight voltage variance on those lines (as you drive stuff out of spec, the transients in those lines become more critical).

    I guess the point of my long-winded post is this...tossing off a "recommendation" like was done in this article (well-intentioned I'm sure) without addressing the actual issues involved seems to me to be habit that should be avoided.

  • rjm55 - Friday, October 01, 2004 - link

    As you pointed out, you can't do the kind of tests you did here with an Intel rig with a locked multiplier. It was good to see all the different memory speeds at the same CPU speed, which proved once and for all that higher memory speeds DO improve performance - even with slower timings at higher speed. Makes my mouth water thinking about how good DDR550 at 2.5-2-2 would be.

    I realize the performance differences weren't huge with just the memory overclocked, but most people will overclock the CPU AND the RAM and that will make a huge combined difference in performance. Tweaking is about squeezing the most from your gear, and you CAN get more out of memory at higher speeds.
  • eetnoyer - Friday, October 01, 2004 - link

  • saechaka - Friday, October 01, 2004 - link

    i can't seem to find a legit place to buy that ocz 3200 rev. 2. any suggestions Reply
  • Avalon - Friday, October 01, 2004 - link

    Excellent article. It's good to know what different memories can do on the Athlon 64 platform. Reply
  • ramclocker - Friday, October 01, 2004 - link


    The psu is probably around 20A on the 12V...I know from my testing 20A doesn't cut it anymore on a high end gaming/benching also have to remember that at high speed the ram will be drawing high levels of current also and the board will draw higher current due to heat etc.

    I found the article an excellent read due to the fact it finally proved to me with reasonably tight timings running high fsb over 2-2-2 at 200 is the way to go...running 2.4gig for all tests Wes was the wise move here...great work.
  • Blappo - Friday, October 01, 2004 - link

    The computer would probably wouldn't use more than 250W. I understand that you don't want to mention the make and model. The nVidia 6800 Ultra draws most of its power through the 12V connection to the PSU, where the ATI 9800 Pro draws its power from the AGP slot. What is the max current rating on the 12V rail for the 465W PSU that you were using? I agree that a high quality PSU is needed (although not necessarily high max rating). Reply
  • Wesley Fink - Friday, October 01, 2004 - link

    #12 - The Asus A8V is reviewed at and compared to other 939 boards. In memory testing we use a standard test bed to minimize variables.

    #11 - The 90nm Athlon 64 tests should appear next week. We have just received 90nm 3500+ and 3000+ processors. AMD did not do a media launch on these processors, so we had to find them on the open market

    #9 and #11 - A Value RAM roundup is in the works, but it has been moved out a while because of a large number of new launches this month.
  • Deuce - Friday, October 01, 2004 - link

    It sure would be nice with tests also conducted on the Asus A8V. I'm still deciding between the two. Reply
  • PrinceGaz - Friday, October 01, 2004 - link

    Just to follow up that comment, I suppose the DDR533/2.4GHz results are actually the most useful out of them all when it comes to comparing those particular modules. All of them were fastest (at 2.4GHz) at that speed, except for the OCZ PC3200 Plat Rev.2 which was marginally faster at 8x300 for DDR600.

    Anyway, I'm looking forward to the reviews of the desktop 90nm A64 processors, and especially finding out how well each of them overclocks.

    And also the Value Memory review you promised a few weeks ago :)
  • PrinceGaz - Friday, October 01, 2004 - link

    Although all the (admittedly premium quality)memory could reach very high speeds, that didn't have much impact on performance.

    Taking the highest clocking brand as an example, the OCZ PC3200 Platinum Rev.2, from the DDR400 2-2-2 speed to the DDR534 2.5-4-3 speed which was the best result still at a CPU speed of 2.4GHz, the results were

    Quake 3: 516.3 -> 525.8
    Super PI: 80 -> 79 (lower is better)
    Wolf: 110.8 -> 112.7

    So running the memory at DDR534 instead of DDR400 provided less than 2% increase in performance. This is to be expected when you compare the real-world performance of S754 and S939. The only thing that is important is that the memory can do 1T command-rate to the maximum overclock of your A64 at default multiplier.

    I think the results on the highest memory performance page are probably misleading to some readers. It shows the Crucial Ballistix coming in at 536.5fps on Q3. Looking at the results I see that was at 9x278 for a CPU speed of 2.5GHz. Your CPU was able to reach over 2.6GHz so the performance in real world tests would have been somewhat higher with a 10x multiplier. Sandra results are irrelevant to most people.

    It would be better if you included an additional test in addition to Highest Memory Speed, and Highest Memory Performance. This would be Highest CPU Speed where the CPU is maxxed out, and the memory run at whatever multiplier gives best performance on real-world tests (ignoring Sandra). I suspect the results would be a *lot* closer.
  • AtaStrumf - Friday, October 01, 2004 - link

    Now about some value RAM tests? These modules are just too expensive for most of us. Reply
  • Jalf - Friday, October 01, 2004 - link

    Or maybe the "average" user would rather blow $200 on 1GB memory ;) Reply
  • Zebo - Friday, October 01, 2004 - link Reply
  • mkruer - Friday, October 01, 2004 - link

    If you get the chance, can you please test with 2GB of PC3200? I’m sure most would love to see what type of performance hit there will be with the larger modules vs. the smaller ones. Looking at the benches so far, it looks like even buying the cheap 1GB PC3200 modules will have negligible impact on the performance as long as the times are kept relatively low (under 3cls.) And one more big IF you could test 4x512 PC3200 with lower clock timings (2-2-2-5) vs 2x1024 PC 3200 with timings of (3-3-3-8) I’m sure that for the average user they would rather blow $400 for 2GB of slow memory then $400 for 1GB of fast memory. Reply
  • Zebo - Friday, October 01, 2004 - link


    p/p is horrendous for this stuff. It's too bad you don't include micron/crucial 8t in there which can also clock to 260 for half the price.
  • Kishkumen - Friday, October 01, 2004 - link

    I've loved all of these recent memory articles. For a while now, the current state of memory in general has been the fuzziest for me. Now I'm starting to get a clearer picture of where things are at and which direction to go. I'm still nursing along my old P4 Northwood, but the A64 plunge is imminent. Nice to see that memory development is keeping up at a strong pace what with 600 MHz speeds now a strong reality. Reply
  • RaistlinZ - Friday, October 01, 2004 - link

    Thank you for the great article! From your tests it looks like the OCZ 3200 Rev.2 is the best of the best. It performed near the top in every test and edged out the Crucial Ballistix at the highest speeds.

    I guess my choice for a memory upgrade is clear now. :)
  • klah - Friday, October 01, 2004 - link

    Seems you cut something off at the end of page 9

    "We have asked AMD to provide some insight into why we are "...
  • skiboysteve - Friday, October 01, 2004 - link

    excellent article, ill keep this in mind when I upgrade... im still pluggin on a TbredB @ 2.2 w/ a modded 9500nonpro Reply

Log in

Don't have an account? Sign up now