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?
Comments Locked

202 Comments

View All Comments

  • Josh7289 - Friday, July 14, 2006 - link

    When I go to college next year, I know what I'll be buying. :)
  • Pirks - Friday, July 14, 2006 - link

    quote:

    When I go to college next year, I know what I'll be buying. :)
    When I see AMD price cuts next month, I know what your face'll look like. :)
  • theoryzero - Friday, July 14, 2006 - link

    This review (and the many others from other review sites) is interesting in that it confirms Intel's bold claims made back in March.

    However, with that out of the way, what I really want to know is which chipset/MoBo to go with? Is springing for high-end memory worth it on Core2? Any plans for a motherboard review in the very near future? Given the Abit AB9 preview and recent articles on DDR2-1000 memory I kind of expected this stuff to show up after the NDA was lifted too.
  • Wesley Fink - Friday, July 14, 2006 - link

    Conroe starts shipping July 27th - the original launch date. Intel moved the launch date back two weeks because OEMs like Dell and HP were hot to start advertising and shipping Conroe systems. I suspect Intel was also trying to salvage their NDA. Good news is hard to contain and two more weeks of NDA would have likely resulted in more info leaks that Intel did not want to turn into a flood.

    You will be able to find answers at AnandTech to your basic Conroe motherboard and memory questions early next week.
  • theoryzero - Friday, July 14, 2006 - link

    Sounds good, thanks!
  • bob661 - Friday, July 14, 2006 - link

    quote:

    You will be able to find answers at AnandTech to your basic Conroe motherboard and memory questions early next week.
    Thanks.
  • Howard - Friday, July 14, 2006 - link

    I would have loved to see minimum FPS as well as average FPS. The review was great otherwise.
  • JarredWalton - Friday, July 14, 2006 - link

    While we could report those scores, we didn't feel we should at least on Oblivion. The reason for that is because the Oblivion runs were manually tested with FRAPS, and the results aren't perfectly comparable between runs. Anyway, here are some of the numbers, but recognize that the margin of error is going to be larger than what you would see with automated timedemos:

    Oblivion Bruma:
    Core E6300: 32-53.7-77
    Core X6800: 47-78.6-117
    AMD 3800+: 27-47.0-72
    AMD FX-62: 38-62.5-94

    Oblivion Dungeon:
    Core E6300: 39-81.3-211
    Core X6800: 57-106.5-214
    AMD 3800+: 35-72.0-189
    AMD FX-62: 46-89.9-211

    F.E.A.R.:
    Core E6300: 34-92-224
    Core X6800: 41-118-310
    AMD 3800+: 28-83-212
    AMD FX-62: 38-101-247

    Rise of Legends:
    Core E6300: 8-68.2-137
    Core X6800: 45-120.5-216
    AMD 3800+: 5-52.1-115
    AMD FX-62: 25-78.4-144

    Minimum frame rates in the Rise of Legends benchmark are zero on most of the Pentium D systems, so again we take those with something of a grain of salt. Regardless of how you look at it, though, the Core 2 Duo processors consistently outperform their AMD equivalents in gaming results. We will be looking at additional games next week in our motherboard reviews.
  • nowuadmit - Friday, July 14, 2006 - link

    First off, when u guys reviewed a 486/33 for the first time... it was also "the fastest piece of silicon the world has ever seen"... then later, when u saw the P2-550... THAT was the "fastest piece of silicon... blah blah".

    Well its nice that intel has made a really expensive and barely available new chip, so they can finally match (or even exceed by a few points) AMD's reasonably priced previous generation. So now u can say they have the fastest, and if u want to spend thousands on a CPU that is only slightly faster than something that costs hundreds of dollars, HEY BE MY GUEST! there have been worse wastes of money! (in theory anyways, since the avg american wont be able to actually find any of these in the local stores for quite some time, if intel follows their previous habits).

    Anyhow the point of my comment... i think this is a good thing. i mean, anything that actually causes an intel fanboi to admit finally that intel has been reaming your butts for 5 years..

    while the people who simply buy whatever CPU makes the most sense have been enjoying getting great quality nookie from AMD! SO i present to you, this Technical Quote of the Year, spoken on behalf of all intel fanbois with half a brain!

    Technical Quote of the Year:

    "...power hungry, poor performing, non-competitive garbage (sorry guys, it's the truth) that Intel has been shoving down our throats for the greater part of the past 5 years."

    Congradulations on your honesty, finally.
  • MrKaz - Friday, July 14, 2006 - link

    Yes that’s truth.

    80% of Intel crap, only 20% is good. The Pentium M.

    So they say 5 years ago until now Intel was “all” crap. I think Northwood was/is very good.

    My maximum price is 100$ (or 120$ with cooler) for a CPU. Only AMD as an honest performing CPU for those prices.

Log in

Don't have an account? Sign up now