If money were no object in buying a new computer system, enthusiasts would almost always choose the fastest memory with the best timings and the largest overclocking potential. The reality for most, however, is that a new system purchase or upgrade is most often a question of where to put your money for the greatest performance return. To make those kinds of decisions on a system purchase, you really need to know the real difference in performance between buying the best and buying something half the price, and possibly putting the money saved elsewhere for a larger performance gain.

These questions of the real impact of memory speed and timings on performance are things that we have talked about in many memory reviews. AnandTech has always been an advocate of real world performance measurements, and we've shunned using just synthetic benchmarks in our testing of every type of component. This is not because synthetic benchmarks are not useful - they are often very revealing of component differences. It is because running just synthetic benchmarks can severely distort the picture of performance with real applications and real games. That is why we always use games and the pure number-crunching Super Pi in our memory tests. It is also the reason why we test using both Buffered (Standard) and Unbuffered synthetic benchmarks. We have found in much of our testing that the less commonly used Unbuffered benchmarks more closely mirror how games really respond to memory differences.

This quest for real performance differences is also the reason why we moved to testing different memory speeds at the same CPU clock speed in our Athlon 64 memory tests. The AMD CPU, with unlocked multipliers, allowed us to finally remove the CPU speed differences from our memory tests so that you could finally see the true impact of memory speed increases and memory timings on performance. As you have seen in past reviews, those performance differences are very real, although they are much smaller than many memory manufacturers might want you to believe. On the other hand, faster memory speeds and faster memory timings do improve performance, no matter what some nay-sayers are determined to prove.

Let's state, right up front, that if you're on a tight budget, memory is often a good place to save a little money by buying less than the best and moving that cash to a better CPU or a better video card. You are likely to get a bigger performance boost with the extra $150 that you save on memory by buying a faster CPU or an upgraded video card. But don't be misled. The upgrade to the next higher CPU would normally increase performance maybe 5 to 10% - about the same as the difference in performance between cheap, slow memory and fast memory with tight timings. Video cards usually yield more than a 10% boost for that same $150.

If you do need to balance and buy less than the best in memory, the need for help is even greater than if you can afford whatever you want. Some Value RAM is pretty basic and slow, and some is just about as good as the best and most expensive that you can buy. For that reason, it was well beyond time for our first Value RAM roundup. There are some disappointments and a few very pleasant surprises. We think that you will be surprised by how much performance there is in some cases, and how little in others, when you're shopping for Value RAM.

The Memories
Comments Locked


View All Comments

  • LoneWolf15 - Wednesday, April 13, 2005 - link

    Thanks for a good article. I usually post response for constructive criticism, but I ought to balance that out more. The Value VX OCZ RAM was particularly interesting and worth knowing about.
  • Wesley Fink - Wednesday, April 13, 2005 - link

    #79 - I double-checked the part number on the OCZ "Value BH5" and corrected the Part Number on page 2. Thanks for pointing this out.
  • ericeames - Wednesday, April 13, 2005 - link

    I belive that the idea of this roundup was good but it has some flaws:

    I dont think that getting memories directly from the manufacturer is a good idea. I know that this is how it "works" it makes the result less credible!

    The overclocking possibilities is not THAT important altough it should not be neglected.

    Compairing them with better brands was a good idea, it makes the results relative.


  • srstudios - Wednesday, April 13, 2005 - link

    Wesley, it seems that the part number is incorrect for the BH-5 OCZ shown on page two.


    ELDCGE-K for BH-5 2-2-2

    Nice article though, thanks for all the great work!
  • srstudios - Wednesday, April 13, 2005 - link

  • xsilver - Tuesday, April 12, 2005 - link

    another thing I have to clear up..... I don't mind extreme voltages used on a "performance" review but this was supposed to be a "VALUE" roundup.... so in this situation extreme voltages may not be warranted

  • xsilver - Tuesday, April 12, 2005 - link

    #62 wesley... you continiued analogy is flawed....
    the thing you forget to mention is that the ferrari only smokes the chevy IF and ONLY IF the racing alcohol is used (3.4v) and the racing alchohol is only available to people who buy say brand X tyres (dfi board).... so in essence you cannot separate the dfi board and the vx + bh5 ram.... they must be used together.....
    I stand by my statement that this "review" smells a bit like an advertisment for dfi and ocz

    I thought it was generally accepted that you can now get the same performance with 1:1 overclocking on loose cas3 timings as a lower mhz with tighter cas 2 timings.... so the need to push the ram to the highest mhz is unnessessary to get the best performance
  • JustAnAverageGuy - Tuesday, April 12, 2005 - link

    Poor Wesley.

    The crowds can never be pleased.
  • AtaStrumf - Tuesday, April 12, 2005 - link

    ChineseDemocracyGNR's post reminded me of one more thing that is sooooo... wrong with this article. 240MHz is nowhere near enough for 1:1 OC-ing of A64s, because unlike you, we don't have 2,4 GHz (4000+, 939, 12x multiplier) chips which cost a fortune, but rather 1,8/2,0 GHz (3000+/3200+, 939, 9x/10x multiplier) chips which cost much less. So if you do the math that is 2160/2400 MHz, which is not exacly the limit, at least not anymore, now that the 90 nm Venice chips are just around the corner.
  • AtaStrumf - Tuesday, April 12, 2005 - link

    I noticed that typo too Olaf van der Spek, but more importantly this very same Transcend RAM gave me quite a bit of trouble and I would not recommend it to anyone with an A64. I was getting strange errors on my ABIT KV8 K8T800Pro, A64 3200+ S754, 2x512 transend RAM combo, like consistent NERO Identity check failures after a DVD burn. Really annoying!!! There were other stuff too, but the point is after I switched to TwinMOS Twister 3700, PQI OEM 3200, Geil Value 3200, Crucial Ballistix 3200 CL2 (best RAM I tested so far), Corsair XMS 3700 (crap RAM BTW), APACER 4000 (TCCD chips; on A64 won't run with any other stick, eg. 1x512 Ballistix + 1x512 APACER, system only sees 512MB of APACER RAM; checked with many different memories) or any other memory for that matter, everything worked just fine, so there is something really strange wrong with Transcend RAM so I strongly recommend that all A64 users avoid it like a plague.

    I agree that the only good choice in this roundup was a OCZ VX Value. You shouldn't let manufacturers pick the RAM you test. Ask us what we want to see tested, we'll have plenty of ideas. The ones I'd like to see are Corsair Value (which I know is crap, I just want you to show in Anand's recent blog entry AT's much advertised backbone and prove to Corsair that you will not take their BS excuse and will still test their RAM, even if they don't want you to, because we, the readers, come first), then GEIL Value 3200 isn't all that bad, then TwinMOS Speed Premium, PQI 3200 OEM, Geil pc4000 Ultra Platinum isn't that expensive either.

    And please stop at 2,9V, because that is as high as most are willing to go, and don't test more that CAS 2.5 and 1T.

Log in

Don't have an account? Sign up now