Does High-Speed DDR3 Help AMD's FX? Four 8 GB Kits, Reviewed

Making A Case For High-Speed RAM

The primary goal of today’s test was to find out if the reports of an AMD FX-based performance boost with high-speed RAM were true. The secondary goal was to find the best module set to get us that performance advantage. And while most of today’s benchmarks demonstrated negligible performance gains going from DDR3-1600 to DDR3-2400, one application really did jump by 6%. That’s a small gain in performance for such a big push in data rate, but it’s still noticeable. Dropping some of our RAM to CAS 6 at DDR3-1600 didn’t give us the same speed-up.

So, we know that the reports of improved Bulldozer performance with high-speed RAM are at least partly true, at least in certain types of applications. The next question is, which module set best gets us there?

This wasn’t supposed to be a value shootout, but one thing that a quick look at DRAM prices show is that only one of the top two module sets is currently available for purchase. The $25 price increase for G.Skill’s Ripjaws X DDR3-2133 would account for only 1% of a $2500 machine’s total price, and the RAM is capable of proving far more than 1% performance advantage in some applications.

Still, the best reason we can think of to buy high-end memory is to show off a big overclock. And so, a value chart that shows a maximum data rate divided by the baseline (DDR3-1600), compared to an actual price divided by the average price for all modules ($69), might not be all that important. Still, it’s nice to see that G.Skill holds its value when compared to top-brand DDR3-1600 CAS 9 modules.

A performance per price chart would be far more brutal to G.Skill, unless we add total system cost to the equation. The problem is, everyone's total system cost is likely to differ from ours. Parts like high-end cases and optical drives have no impact on today’s benchmarks, but would help make the case for G.Skill’s RAM by diminishing its relative price to the overall system.

Since high-speed memory only provides a marginal performance benefit to a small number of all the applications on your PC, we can only recommend the step up to a subset of enthusiasts willing to pay that extra money. At the end of the day, you will need to be the judge of whether or not the added expense is worth the marginal performance increase. Regardless, this was still a review of four high-performance memory kits, and G.Skill produced the best purchasable kit in this test. Therefore its Ripjaws X DDR3-2133 CAS 9 gets our stamp of approval.

This thread is closed for comments
65 comments
    Your comment
  • Rockdpm
    Interesting!. Still would wait for Pile driver. But since i do have a Kingston 8GB kit and a Sabertooth 990FX... may just have to try it one day
  • jdwii
    Great article this proves a lot of rumors!

    People can finally stop saying how unfair it is to test Amd with slower ram when it supports higher speed ram even though higher speed ram cost more money. It would actually be unfair to Intel its not their fault Amd needs higher speeds to compete(which it doesn't) with Intel's 1333 ram controller.

    Ram speeds do how ever make a big difference on APU's.
  • mayankleoboy1
    would love to see the same review for Ivy Bridge processors with and without iGPU HD4000.

    and do include more tests in the review.
  • sarinaide
    More of the cheap stuff is my mantra.
  • Crashman
    sarinaideMore of the cheap stuff is my mantra.
    Yeh yeh, but I really can't complain about paying $70 for 8GB of DDR3-2133, especially when it overclocks like that.
  • Reynod
    AMD has enjoyed the performance benefits of an integrated memory controller for more than twice as long as Intel. And yet, it seems that Intel sets today's standard for acceptability.

    Replace the last word with "performance" crash.

    Good article which gets answers a couple of questions people have been asking - well done !!

    :)
  • jaquith
    Thomas thanks for the review.

    First note:
    *KHX2400C11D3K4/8GX - is a Quad Channel kit specifically designed for the X79 ; link http://www.kingston.com/us/company/press?pagename=n1111c&year=2011&prLanguage=english_emea
    *WQ213UB4G vs WQ213UX16G (kit) - is ONE stick of RAM (not a matched kit) and its 'kit version' is Quad Channel kit specifically designed for the X79 ; link http://www.supertalent.com/datasheets/WQ213UX16G.pdf

    After that it makes sense that those (2) "kits" (in one case 4 individual sticks) failed @ Rated, and I wasn't interested in researching the others. I've seen other more in-depth testing on the AMD FX-8150 indicating that overall there's a slight advantage for the DDR3-1866 kits on a most of the AMD FX processors. Though @ 4AM 'to me' I'll edit this post after a few cups of coffee.

    The 'ideal' AMD kits have JEDEC 'SPD' @ Rated 'cloned' (XMP) encoding which is optimized for AMD and sometimes slightly different so you're not 'stuck' as you put it @ DDR3-1333. Those are the better KITS to test on the AMD FX lines.
  • SteelCity1981
    Ram sppeds on a Bulldozer CPU means little and this just proves that so it doesn't matter if you are using DDR3 1600 ram or buying high performance RAM that can overclock to DDR3 2800 your not really seeing hardly any improvement with Bulldozer by doing so.
  • bartholomew
    Good article, very informative!
    Thanks :)
  • RAM speed also has a HUGE impact in controlled cache environments:
    http://thessdreview.com/our-reviews/romex-fancycache-review-ssd-performance-at-13gbs-and-765000-iops-in-60-seconds-flat/

    Even the best SSDs can barely muster 100,000 IOPs, yet even slow RAM can easily exceed 700,000 IOPs.
    Bandwidth, throughput, and latency are equally insane. RAM puts SSDs to shame exponentially more than SSDs put hard drives to shame. Contrary to what many "enthusiasts" believe, RAM timings are virtually irrelevant. Frequency is by far, much more important.
  • Crashman
    jaquithThomas thanks for the review.First note:*KHX2400C11D3K4/8GX - is a Quad Channel kit specifically designed for the X79 ; link http://www.kingston.com/us/company [...] glish_emea*WQ213UB4G vs WQ213UX16G (kit) - is ONE stick of RAM (not a matched kit) and its 'kit version' is Quad Channel kit specifically designed for the X79 ; link http://www.supertalent.com/datasheets/WQ213UX16G.pdfAfter that it makes sense that those (2) "kits" (in one case 4 individual sticks) failed @ Rated, and I wasn't interested in researching the others. I've seen other more in-depth testing on the AMD FX-8150 indicating that overall there's a slight advantage for the DDR3-1866 kits on a most of the AMD FX processors. Though @ 4AM 'to me' I'll edit this post after a few cups of coffee.The 'ideal' AMD kits have JEDEC 'SPD' @ Rated 'cloned' (XMP) encoding which is optimized for AMD and sometimes slightly different so you're not 'stuck' as you put it @ DDR3-1333. Those are the better KITS to test on the AMD FX lines.
    Read page 2. They're all Intel-optimized kits because that's the only thing any of these companies have. If you read something about an AMD-optimized kit being for sale somewhere, it's probably either a niche manufacturer or simply a marketing lie.
  • ta152h
    If you're going to spend extra money on memory, why not use the extra money for a real CPU? Meaning, from Intel, AMD doesn't make anything remotely competitive, and if you're getting performance parts to try to make it perform, you're probably better off starting with a processor that performs better in the first (second, and third) place.

    It's like putting a diamond in an aluminum ring. It makes no sense.
  • de5_Roy
    fx needs ddr3 2133 kits to reach 20~ gb/s on sandra bench with while a sandy bridge pentium can reach the same bw @ddr3 1333.
    http://media.bestofmicro.com/G/Y/327202/original/sandra%20mem.png
    amd apus are different story. those perform better in games with faster memory.
    only posted in case some wonder how much intel gets out of rams.
  • noob2222
    great article, 2 games and 2 programs that don't respont to memory speed must represent all games ever made.

    This article proves one thing, not every program responts to memory speed.

    http://vr-zone.com/articles/amd-fx-8150-memory-scaling-investigation--feeding-the-bulldozer/13704.html
    http://www.madshrimps.be/articles/article/1000220/AMD-FX-8150-Bulldozer-CPU-Review/4#axzz1wX1gAnfF

    Ill match your 4 cherry picked benchmarks against 2 websites that state otherwise.

    Throw civ V into the mix and your way off.
  • Crashman
    noob2222great article, 2 games and 2 programs that don't respont to memory speed must represent all games ever made. This article proves one thing, not every program responts to memory speed.http://vr-zone.com/articles/amd-fx [...] 13704.htmlhttp://www.madshrimps.be/articles/ [...] z1wX1gAnfFIll match your 4 cherry picked benchmarks against 2 websites that state otherwise.Throw civ V into the mix and your way off.
    Cherry picked for what, to prove that faster memory helps? Well guess what, it didn't help much. Or did you mean cherry picked for this article? Are you making blind accusations without even looking to see which benchmarks were used in the last several memory articles?

    Please, at least try to be honest. You already knew the Madshrimps article was explicitly dishonest because it used CAS 8 timings at all speeds (even 2133). Real-world memory supports tighter timings at lower frequencies, and requires looser timings at higher speeds. Madshrimps gimped the low-speed tests and boosted the high speed configuration intentionally.
  • noob2222
    take a small enough sample to make sure the results don't prove a thing. why not use some programs that do respond to memory speed?


  • Crashman
    noob2222take a small enough sample to make sure the results don't prove a thing. why not use some programs that do respond to memory speed?
    The test here was set up long ago to use 50% GPU-limited and 50% platform-limited games. Replace GPU with CPU for the other tests, and the hope was to find 50% of the benchmarks with noticeable performance gains. Of course that's based on a real-world scenario, where both GPU and platform-limited games coexist.

    Notice in the chart you showed, the difference between 2133 and 1866 is rather small. The baseline in this test was DDR3-1600 CAS 9 because that's the slowest memory a performance builder would use. In other words, the 1333 and 1066 results are irrelevant.

    The benefit you're seeing in some articles is reduced real-time latency, which I'm fairly certain is discussed in this article. DDR3-1333 CAS 7 has the same real-time latency as DDR3-2666 CAS 14, since the cycle time is inverse of frequency.

    If we assume for only a moment that DDR3-1600 is barely fast enough to fill the bandwidth needs, most of the difference over 1600 actually measures how long the CPU is waiting for instructions. Now I wouldn't actually assume 1600 is enough, I only assume that 1600 is the minimum speed that goes into most performance builds. After all, who would buy 1333 to save money in a performance machine when 1600 is so cheap?
  • noob2222
    CrashmanThe test here was set up long ago to use 50% GPU-limited and 50% platform-limited games. Replace GPU with CPU for the other tests, and the hope was to find 50% of the benchmarks with noticeable performance gains.


    the problem is both of those games, dirt 3 and metro 2033 are gpu limited. http://www.techspot.com/review/403-dirt-3-performance/page7.html
  • Crashman
    noob2222the problem is both of those games, dirt 3 and metro 2033 are gpu limited. http://www.techspot.com/review/403 [...] page7.html
    GPU limited at max quality and 1920. This test was at medium resolutions and details (the "high" setting in these games is actually mid-high). DiRT 3 in particular is supposed to be GPU-bottlenecked only at its highest settings.
  • noob2222
    If you have a copy of civ V, try it, you will laugh at the results, 1333 to 2133 is nearly 25% faster on the full render. There are benefits to faster memory, limiting the benchmarks limits the results.
  • jaquith
    CrashmanRead page 2. They're all Intel-optimized kits because that's the only thing any of these companies have. If you read something about an AMD-optimized kit being for sale somewhere, it's probably either a niche manufacturer or simply a marketing lie.

    Thomas -- I have seen (with my own eyeballs) the AMD so called 'optimized' kits with cloned e.g. DDR3-1866 with BOTH XMP and 'SPD' JEDEC (Duplicate) encoding.

    No LIE -- Google for yourself or look at one of my (1000's) of RAM posts.

    I of all folks here have seen and know plenty especially when it comes to RAM -- name another 'here @ TH' that knows more.

    I can only 'wish' one size fits all when it comes to RAM -- it clearly does not! Adjust the OC, CAS Timings and Voltages i.e. take some time and I'm confident unless those (2) I looked a in some way 'damaged' (possibility) can & will work -- try raising the CAS if Voltage fails -- play around in the BIOS.

    Frankly, I NEVER advise anyone, especially with >DDR3-1333 (even then I still do), to purchase anything BUT a fully matched set. G.SKILL in particular doesn't want to RMA (1) stick from e.g. an 8xDIMM fully matched set especially as either the Frequency <or> density is >4GB/stick. Heck the (stable) CAS per Frequency often doesn't work when you put (2) two Xerox (same PN) sets together.

    I am not looking to debate you, I linked B&W the info on the kits.
  • Crashman
    216041 said:
    Thomas -- I have seen (with my own eyeballs) the AMD so called 'optimized' kits with cloned e.g. DDR3-1866 with BOTH XMP and 'SPD' JEDEC (Duplicate) encoding. No LIE -- Google for yourself or look at one of my (1000's) of RAM posts. I of all folks here have seen and know plenty especially when it comes to RAM -- name another 'here @ TH' that knows more. I can only 'wish' one size fits all when it comes to RAM -- it clearly does not! Adjust the OC, CAS Timings and Voltages i.e. take some time and I'm confident unless those (2) I looked a in some way 'damaged' (possibility) can & will work -- try raising the CAS if Voltage fails -- play around in the BIOS. I am not looking to debate you, I linked B&W the info on the kits.
    No, you didn't link the info on "the" kits, because "the" kits are the nonexistant kits you speak of. You linked to "these" kits, which, like any other performance RAM, are Intel optimized.

    You can't sucessfully SPD an XMP value for 1.65V, because SPD has no voltage info. If you did, it still wouldn't be JEDEC approved.

    Maybe you should read Page 1: All major manufacturers got invited, all were told this was an AMD article, and all but four of these companies said something like "Sorry, we don't have any memory that's a good match to the article" (that response is implied for companies that refused to respond). That should have put an end to your argument, yet here we are having this discussion.

    I'm not trying to be crass, but the stuff I'm currently going over with you is stuff I went over on my own before testing even began.
  • jaquith
    I don't know what you're talking about <or> you simply didn't read what I said. Existent or not the links are to the RAM you have placed in your article, and IF you don't like what's there take it up with the Vendors. I assume you told the folks for the 'free' RAM that you were testing on the AMD platform and IMO there was a terrible miscommunication.

    Before I was referring to kits with 'matching' JEDEC / XMP info like shown below (Example of a DDR3-1600 kit with duplicate profiles for DDR3-1600):


    I don't need a headache this AM...it's nothing I care too much about, and I was researching 'why' the (2) 'sets' failed @ Rated and at least 'I' get it...
  • Crashman
    216041 said:
    I don't know what you're talking about <or> you simply didn't read what I said. Existent or not the links are to the RAM you have placed in your article, and IF you don't like what's there take it up with the Vendors. I assume you told the folks for the 'free' RAM that you were testing on the AMD platform and IMO there was a terrible miscommunication. Before I was referring to kits with 'matching' JEDEC / XMP info like shown below (Example of a DDR3-1600 kit with duplicate profiles for DDR3-1600): http://aphnetworks.com/review/g_skill_sniper_f3_12800cl9d_8gbsr2_2x4gb/cpuz.png I don't need a headache this AM...it's nothing I care too much about, and I was researching 'why' the (2) 'sets' failed @ Rated and at least 'I' get it...
    You're showing me a JEDEC DDR3-1600 CAS9? Here I thought we were talking about performance RAM.

    Since we're talking about two different things, there's a good chance that we're not actually in disagreement.

    As for the 1.28V thing, well, I actually left the JEDEC updates for low-voltage RAM out of this discussion of non-low-voltage RAM. Since I'm a stickler for details, I apologize for skipping over that detail.