Memory Latency: No Integrated Memory Controller Necessary

One thing we pointed out in our earlier preview coverage of Intel's Core 2 Extreme is that the new processors have extremely low latency memory access, despite relying on the same memory controller as the previous generation of Intel CPUs.

Without an on-die memory controller Intel's Core 2 processor must use the memory controller in its chipset, which currently means the 128-bit DDR2 memory controller in either Intel's 965 or 975X chipset. The confusing thing is that although the Core 2 processors use the same memory controller as the old NetBurst processors, memory latency has been improved tremendously:

Memory Latency Comparison

Intel's Core 2 processors now offer even quicker memory access than AMD's Athlon 64 X2, without resorting to an on-die memory controller. While Intel will eventually add one, the fact of the matter is that it's simply not necessary for competitive memory performance today thanks to Intel's revamped architecture. Update:As many astute readers have pointed out, Core 2's prefetchers are able to work their magic with ScienceMark 2.0, which results in the significant memory latency advantage over AMD's Athlon 64 FX-62. This advantage will not always exist; where it doesn't, AMD will continue to have lower latency memory access and where it does, Intel can gain performance advantages similar to what ScienceMark 2.0 shows.

 CPU Everest
CPU-Z 1.35 (8192KB, 128-byte stride) CPU-Z 1.35 (8192KB, 64-byte stride)
AMD Athlon 64 FX-62 (2.8GHz) 45.9 ns 43.2 ns 19.3 ns
Intel Core 2 Extreme X6800 (2.93GHz) 59.8 ns 52.8 ns 10.9 ns

 

Because of the Core 2's intelligent prefetchers doing too good of a job with ScienceMark memory latency test, we wanted to also showcase situations where Core 2 would feature slower memory accesses than the AMD platform with its integrated memory controller. Everest's results are more in line with what we'd expect to see, with the FX-62 offering over 23% faster memory accesses than the X6800. CPU-Z's latency tool also reported somewhat similar findings, with an 18% performance advantage due to AMD's integrated memory controller. CPU-Z also provided us with numbers that showcase how well Core 2 can perform if its prefetchers are able to "guess" correctly; at lower strides the Core 2 Extreme manages faster memory access than the FX-62. The 128-byte stride numbers are indicative of what will happen if the pre-fetchers are not able to get the Core 2 the data it needs, when it needs it, while the 64-byte numbers show you what can happen when things go well.

L2 Cache: 4MB or 2MB? FSB Bottlenecks: Is 1333MHz Necessary?
POST A COMMENT

200 Comments

View All Comments

  • arachimklepeto - Tuesday, July 25, 2006 - link

    And what about noise Core 2 Duo fan(decibels)? Reply
  • bmaamba - Tuesday, July 25, 2006 - link

    Hi,
    Acc. to Toms hardware, for EIST to work, setting in Control panel has to be changed from "desktop" to "portable/laptop".AT guys, was this done? If not, how about putting it in the "Power consumed" graphs?(acc. to Tom(if i rem. right), least power in this mode is about 25watts by core 2 duo!!!).Also anyone knowledgeable, is this setting available in Linux?
    Also, how about putting XP X2 3800+ EE in the encoding benchmarks (along with core 2 duo 6300)?
    Thanks
    Ed
    PS.Price and power consumed when idle are v. imp. to me.
    Reply
  • herkulease - Thursday, July 20, 2006 - link

    Unless I missed it what are temps like on these.

    Reply
  • Justin Case - Monday, July 17, 2006 - link

    What the heck is a "composite score"...? What are the units? How about giving us rendering times (you know, minutes, seconds) and render settings, so the numbers actually mean something...? Reply
  • rahvin - Monday, July 17, 2006 - link

    Where's a good 64bit comparison on Linux and a LAMP stack run at 64bit? There hasn't been a serious linux server benchmark posted. Reply
  • BikeDude - Sunday, July 16, 2006 - link

    I'd love to see some timings from a C++ compiler or two... Looks like I'll have to revise our standard developer PC configuration.

    --
    Rune
    Reply
  • kmmatney - Sunday, July 16, 2006 - link

    "Jarred that would be great to see. The E6300 and X2 3800+ seem close, but the final AMD pricing and the overclocking potential of each could really make either the clear winner for performance per dollar in the midrange segment."

    Yes - this is the test that most people want to see. I';m sure a lot of people are like me, and don't much care about any processors over $200. We want to see that the low end can do!! The AMD X2 3800+ is going to be even lower priced than the E6300, so there may be a good battle at teh low cost end.
    Reply
  • aznskickass - Sunday, July 16, 2006 - link

    Battle? What battle? The war is over my friend. ;)

    The E6300 wins hands down vs X2 3800+, even more so once both are overclocked:

    http://xbitlabs.com/articles/cpu/display/core2duo-...">http://xbitlabs.com/articles/cpu/display/core2duo-...
    Reply
  • Jeff7181 - Saturday, July 15, 2006 - link

    Would have been nice to see a Core Duo CPU in there too just for comparison for those of us with laptops who might considering spending $200 on a Merom if it would increase performance 10-20% over a Yonah with the same power consumption. Reply
  • IntelUser2000 - Thursday, July 20, 2006 - link

    quote:

    Would have been nice to see a Core Duo CPU in there too just for comparison for those of us with laptops who might considering spending $200 on a Merom if it would increase performance 10-20% over a Yonah with the same power consumption.


    Link: http://www.trustedreviews.com/article.aspx?art=316...">http://www.trustedreviews.com/article.aspx?art=316...
    http://www.hardware.fr/articles/623-10/intel-core-...">http://www.hardware.fr/articles/623-10/intel-core-...

    Core 2 Duo E6400 2.13GHz is approximately 15% faster than Core Duo T2600 2.13GHz, in addition to the fact that 4MB cache versions are 3% in average faster, it looks estimation of 10-20% faster per clock than Yonah is right, even with the 2MB cache version.
    Reply

Log in

Don't have an account? Sign up now