Final 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.

Memory Scaling with Overclocking
POST A COMMENT

76 Comments

View All Comments

  • Rajinder Gill - Friday, July 29, 2011 - link

    LOL!

    Caught with your pants down there sir.
    Reply
  • Isaac the k - Tuesday, August 02, 2011 - link

    Memory above DDR3-1333 is plentiful and cheap.
    Sandy Bridge benefits from the bump up to 1600.
    Mobo's all can handle the higher spec'ed memory.

    WHY IS INTEL SPEC'ING THE NEW I7's FOR 1333 INSTEAD OF 1600 BY DEFAULT?

    I don't want to have to futz with my BIOS just to get my memory to run at STOCK timings!
    Why the HELL are they crippling the basic utility of their chips??

    I grant you, it isn't a major increase in performance, but why set the ceiling so low? why not set it higher and let the mobo manufacturers choose what their boards are compatible with???
    Reply
  • shriganesh - Wednesday, August 24, 2011 - link

    This article is thoughtful and very good news for mid-end to high-end system builders :)
    There's absolutely no need to pay extra for those faster and better memory modules!!
    Reply
  • ryedizzel - Tuesday, September 20, 2011 - link

    Thank you very much for this article. I have been memory shopping for a couple days now and debating on different memory speeds vs. latency vs. price. I wish I had the lab setup to test them all. Thanks again! Reply
  • James D - Monday, November 28, 2011 - link

    On each speed type (1333, 1600, 1866 etc.) you used only slightly different timings! For example for speed 1333 Mhz you had:
    1) 7-7-7-18
    2) 8-8-8-18
    3) 9-9-9-18
    In this case of course you won't get much different results! Don't you know that changing CL and other timings you can change tRAS timings which is the minimum number of clock cycles needed to access a certain row of data in RAM between the data request and the precharge command.
    OF COURSE you won't get big difference if you didn't change tRAS! If you lower all other timings then most likely you can decrease tRAS which also will increase performance.

    Please rebench all these with slightly different all types of timings. It means a lot for results.
    Reply
  • poohbear - Saturday, December 17, 2011 - link

    Thanks for running the gaming benchmarks @ 1920x1080 to show us practical results (ie there were none!). i hate it when they run these benchmarks @ 800x600 or some nonsensical low resolution to show us a difference that we really could care less about in 2011. Reply

Log in

Don't have an account? Sign up now