Intel's Core 2 Extreme & Core 2 Duo: The Empire Strikes Back
by Anand Lal Shimpi on July 14, 2006 12:00 AM EST- Posted in
- CPUs
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:
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.
Facebook
Twitter
RSS
202 Comments
View All Comments
Calin - Friday, July 14, 2006 - link
This is a bit more complicated - you could buy a $1000 FX-62, or you could buy a $316 Core2Duo, then a $150+ mainboard. If you want to run SLI, you are out of luck right now - but things might change in the immediate future. If you have NVidia SLI, you must go to Crossfire (at this moment).But anyway, looks like AMD can not compete in the top
Regs - Friday, July 14, 2006 - link
Since the P6!!!! Makes me think if AMD actually cares about improving performance on their processors. Maybe they should scrap the Fab in New York and make a research facility instead. Start hiring interns from MIT. Do something! lol.I admit, even though I enjoyed AMD having the performance crown, It was a period of limited choice and limited performance gain. Who here on the free market care about 100MHz increaments? They went from a 110nm to 90nm with no performance benifit - they went from single core to the dual core X2's with no performance benifit -- they went from DDR to DDR2 with no performance improvement -- now they are going to 65nm which they also made clear they will make no changes to increase performance. AMD has really dropped the ball and they deserve what they get. I don't know why anyone, including over clockers, would want to be a AMD fan boy at the momment.
CKDragon - Friday, July 14, 2006 - link
AMD went from single core to dual core with no performance benefit?Maybe on Planet Troll...
Regs - Friday, July 14, 2006 - link
The X2 improved performance only on specific suites of software. Can you say the same about Conroe? I mean I was really able to crank up the rez in oblivion after I upgraded to an X2 *rolls eyes*.JarredWalton - Friday, July 14, 2006 - link
The performance increases you're seeing in most games on Core 2 Duo come entirely from the better architecture, not from dual processor cores. We just can't test single core performance on Core 2 because such chips don't exist and they won't until Conroe-L ships (in about a year judging by road maps -- it looks like Intel and their partners want to have time to clear out all of their NetBurst inventory first).Regs - Friday, July 14, 2006 - link
Yes, I completely agree. The only difference on the X2 compared to the single cores was encoding. Not unless you do own a 10-thousand dollar server for well...server use.Calin - Friday, July 14, 2006 - link
Being a fanboy is like a religion - you don't change your religion overnight.AMD cares about selling expensive processors. As long as the P4 was the opposition (especially after the Northwood days), AMD was king of hill, and sold its processors at whatever prices the market would pay. Now, Intel took that place. I hope this will change with K8L, as this will bring even lower prices for even better processors.
Also, AMD was unable to produce enough processors, so they sold most of it for a premium. As for the move to 90nm, they got some extra frequency headroom, and lower power consumption. This also reduced their costs (too bad the cost reduction wasn't really passed to customers).
If their move from single to dual core brought no performance benefit, tell that to companies buying dual core opterons for thousands dollars apiece.
segagenesis - Friday, July 14, 2006 - link
Good lord. You might as well throw up the GAME OVER and TILT signs for AMD right now. Although I wouldnt want them to disappear from competition (I dont want us to return to expensive Intel cpu's at the same time) there isnt much I see in this article that gives AMD any advantage at the moment over Intel. Sooner or later this was bound to happen from Intel though, the Athlon 64 made a similar situation against Pentium 4 making it look pretty obsolete comparitavely at the time.Now assuming the prices that AMD plans to drop to are correct, perhaps they can remain compeditive for building a budget system vs. Core 2 as I would not recommend a new Pentium 4 at this point to anyone...
That reminds me of the good ol days over overclocking the Celeron A...
dice1111 - Wednesday, July 19, 2006 - link
Ahhh, yes. My old Celeron A (still overclocked and in use). I was so happy about overclocking back then. Please Intel, let me get that taste of nostalgia!!!mobutu - Friday, July 14, 2006 - link
I'd really like to upgrade to Conroe but I don't want the Intel chipset on motherboard.Jarred, Wesley, do you know (estimate) when you'll have a review with final 590 reference board and when we can expect motherboards with 590 Intel edtn to be available?
Thanks in advance guys. Great Conroe review.