Rumor has it that fast DDR3 memory kits help uncork the performance of AMD's Bulldozer architecture. We grab four of the latest kits sporting data rates as high as 2800 MT/s with the goal of finding out. One contender rises to the top in this round-up.
The introduction of AMD's Bulldozer architecture was marked by a handful of notable improvements and a collection of setbacks. The flagship model with eight integer cores didn't perform as well as enthusiasts were hoping it would. And yet, we took some comfort in the suggestion that operating system scheduling fixes might eventually rectify some of the challenges that the FX family encountered with Windows 7. Better still, the new architecture featured overclocking advancements that could yield improved performance in the hands of an experienced tinkerer, including its higher DRAM multiplier.
We’ve moderated enough of these “speed versus latency” debates that we’re skeptical of any claim that an enthusiast-oriented desktop processor needs faster RAM. Of course, we know that ramping up bandwidth has a positive effect on integrated graphics engines. But we've also received a number of requests suggesting that AMD's FX-series chips stand to benefit significantly from memory running at higher data rates.
In order to put these claims to the test, we pulled out one of our best 990FX-based motherboards, an AMD FX-8150 processor, and invited every memory vendor to send in its best AMD-focused kit in for today's round-up. At the end of the day, four manufacturers chose to participate.

We were all a little surprised to find that AMD's FX processors were overclocking memory at rates well above what we managed to achieve with Intel's Sandy Bridge-based CPUs.
| 8 GB Dual-Channel DDR3 Rated Settings | |||||
|---|---|---|---|---|---|
| Rated Capacity | Channel Organization | Data Rate (MHz) | Timings | Voltage | |
| G.Skill Ripjaws X F3-17000CL9D-8GBXM | 8 GB | 2 x 4 GB | 2133 | 9-11-10-28 | 1.65 V |
| Kingston HyperX KHX2400C11D3K4/8GX | 8 GB | 4 x 2 GB | 2400 | 11-13-11-30 | 1.65 V |
| Super Talent Quadra WQ213UX16G | 16 GB | 4 x 4 GB | 2133 | 11-11-11-30 | 1.65 V |
| Team Group Xtreem LV TXD38192M2800HC11RDC-L | 8 GB | 2 x 4 GB | 2800 | 11-14-14-31 | 1.65 V |
| Reference DDR3-1600 CAS 9 8 GB Dual Channel | 8 GB | 2 x 4 GB | 1600 | 9-9-9-28 | 1.50 V |
You might have noticed that Super Talent’s kit consists of four 4 GB modules, yet only two are shown in the photo. The firm offers these specific modules two ways: as its WQ213UB4G single-module package and WQ213UX16G quad-channel kits. Super Talent chose to send its quad-channel kit rather than two single-channel packages. AMD’s Socket AM3+ supports dual-channel mode, and using only two modules allowed us to retain a constant 8 GB for all tests.
Alternatively, Kingston thought its 2 GB modules would be the best match for AMD’s memory controller, compelling us to use all four modules to retain a constant 8 GB for all tests.
Finally, we added a set of our old high-end test modules to represent the DDR3-1600 CAS 9 baseline, since the specific modules we hand in mind actually ran at the desired default settings. It’s nice, after all, to be able to set up our memory with just a few keystrokes.
The first question that comes to mind is: if all of these modules appear to be Intel-specific, why does AMD-optimized memory still exist?
- Does Faster RAM Improve The Performance Of AMD FX CPUs?
- Does AMD-Optimized Memory Still Exist?
- G.Skill Ripjaws X F3-17000CL9D-8GBXM
- Kingston HyperX KHX2400C11D3K4/8GX
- Super Talent Quadra Series WQ213UB4G
- Team Xtreem LV TXD38192M2800HC11RDC-L
- Test Setup And Benchmarks
- Overclocking And Under-Latency Results
- DDR3-1600 Performance
- DDR3-1866 Performance
- DDR3-2133 Performance
- DDR3-2400 Performance
- Making A Case For High-Speed RAM
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.
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?
and do include more tests in the review.
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.
and do include more tests in the review.
Replace the last word with "performance" crash.
Good article which gets answers a couple of questions people have been asking - well done !!
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.
Thanks
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.
It's like putting a diamond in an aluminum ring. It makes no sense.
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.
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.
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.
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?
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