Sandy Bridge Memory Scaling: Choosing the Best DDR3
by Jared Bell on July 25, 2011 1:55 AM ESTFinal Words
I think we confirmed what we pretty much knew all along: Sandy Bridge's improved memory controller has all but eliminated the need for extreme memory bandwidth, at least for this architecture. It's only when you get down to DDR3-1333 that you see a minor performance penalty. The sweet spot appears to be at DDR3-1600, where you will see a minor performance increase over DDR3-1333 with only a slight increase in cost. The performance increase gained by going up to DDR3-1866 or DDR3-2133 isn't nearly as pronounced.
As a corollary, we've seen that some applications do react differently to higher memory speeds than others. The compression and video encoding tests benefited the most from the increased memory bandwidth while the overall synthetic benchmark and 3D rendering test did not. If your primary concern is gaming, you’ll want to consider investing in more GPU power instead of a faster system memory; likewise, a faster CPU will be far more useful than more memory performance for most applications. Outside of chasing ORB chart placement, memory is one of the components least likely to play a significant role in performance.
We also found that memory bandwidth does scale with CPU clock speed; however, it still doesn't translate into any meaningful real-world performance. The sweet spot still appears to be DDR3-1600. All of the extra performance gained by overclocking almost certainly comes from the CPU overclock itself and not from the extra memory bandwidth.
Finally, although the effects of low latency memory can be seen in our bandwidth tests, they don't show any real world advantage over their higher latency (ahem, cheaper) counterparts. None of the real-world tests performed showed any reason to prefer low latency over raw speed.
Even though there's merely a $34 price difference between the fastest and slowest memory tested today, I still don't believe there's any value in the more expensive memory kits on the Sandy Bridge platform. Once you have enough bandwidth (DDR3-1600 at a small $9-$10 price premium), there's just not enough of a performance increase beyond that to justify the additional cost, even when it's only $34 between 4GB kits. Once you jump to the 8GB kits, the price difference for CL9 DDR3-1600 is a mere $8, but it becomes much more pronounced at $92 to move to DDR3-2133. We simply can’t justify such a price difference based on our testing.
Of course, testing with Sandy Bridge doesn't necessarily say anything about other platforms. It's possible that AMD's Llano and Bulldozer platforms will benefit more from higher bandwidth and/or better latency memory, but we'll save that article for another day. Also, we've shown that performance scaling on integrated graphics solutions can benefit, particularly higher performance IGPs like Llano. Ultimately, it's up to you to choose what's best for your particular situation, and we hope this article will help you make better-informed decisions.
76 Comments
View All Comments
dfjgkheu - Tuesday, July 26, 2011 - link
believe you will love it.====( www )( bestniceshoes)( c o m ) ====
The website whol esale for many kinds of fashion shoes,
like the ni ke, jor dan, pra da, also including the jea ns,shirts,bags,hat and the decorations.
All the products are free ship ping, and the the price is competitive,
and also can accept the pay pal payment.
,after the payment, can ship within short time
vailr - Monday, July 25, 2011 - link
No discussion of differing voltages?A quick check for DDR3 at Newegg shows:
G.SKILL ECO Series 4GB (2 x 2GB) 240-Pin DDR3 SDRAM DDR3 1600
@ 1.35 volts & Cas Latency: 7
vs.
G.SKILL Ripjaws X Series 4GB (2 x 2GB) 240-Pin DDR3 SDRAM DDR3 1600
@ 1.50 volts & Cas Latency: 6
A more thorough consideration of these two DDR3 modules might be interesting.
For virtually the same money, aren't most people going to seek out DDR3 with the lowest possible CAS latency number, and also combined with the lowest possible voltage design?
I know that: I wouldn't consider buying any DDR3 memory modules with a (nominal) CAS latency higher than 7.
JarredWalton - Monday, July 25, 2011 - link
Just as we didn't test with ten different modules (for ease of testing), we didn't use different voltage memory. Whether your RAM is 1.5V or 1.35V, at the same timings and speed the performance should be identical (less than a 0.5% difference). And we did look at the effect of lower latency RAM; sure, at the same price buy lower latency and higher bandwidth RAM, but prices aren't the same, particularly on 2x4GB kits.Tchamber - Monday, July 25, 2011 - link
I'd like to see how these tests stack up against the tripple channel nehalem i7's.duploxxx - Monday, July 25, 2011 - link
compare with what an EOL platform? it was alreay known that there is no added value with memory speed testing on these systems, just like the previous gen., 1366 is dead testing has been done in the pastThis test just showed that it is a lot of wasted money and time investigated in this.
They better take the time and investigate further into Liano memory speed, something that really does scale with memory.
Finally - Monday, July 25, 2011 - link
It's already done, see Computerbase...JarredWalton - Monday, July 25, 2011 - link
We've done it as well for graphics applications:http://www.anandtech.com/show/4476/amd-a83850-revi...
We haven't done the application testing with different DDR3 on Llano, however.
banwell - Monday, July 25, 2011 - link
You can also get a nice 'free' bump in performance at 1600 by switching to 1T. Something the better quality memory will be able to do easily.AssBall - Monday, July 25, 2011 - link
I'm not sure why they didn't test 1T . It is a memory scaling article after all. Anyway TechReport did and their conclusions are about the same, I.E. unless you are overclocking and running synthetic benchmarks, it doesn't really matter.compudaze - Monday, July 25, 2011 - link
Lowering the command rate from 2T to 1T at DDR3-1600 doesn't necessarily mean you can do the same at DDR3-2133. Not all memory modules, CPU's and motherboards are creased equal. Testing all configurations at 2T kept the results comparable.