Does AM2's Performance Make Sense?
Assuming for a moment that the performance we're seeing here today is representative of what AMD will show off in 2 months, does it make sense? AMD has effectively doubled their memory bandwidth but they've seen virtually no increase in performance, other than in some very isolated situations.
If you'll remember back to the introduction of AMD's Revision E core, we did an article about how the new core brought support four new memory dividers allowing you to run at speeds up to DDR500 without overclocking your CPU or the rest of your system. In that article we looked at the overall performance benefit of DDR-500 over DDR-400 on a Socket-939 platform in a variety of situations. A recap of our performance results is below:
| Benchmark | Socket-939 (DDR-400) | Socket-939 (DDR-480) | % Advantage (DDR-480) |
| Multimedia Winstone 2004 | 41.9 | 42.7 | 2% |
| 3dsmax 6 | 2.78 | 2.80 | 1% |
| DivX 6.0 | 50.6 fps | 53.2 fps | 5% |
| WME9 | 4.22 fps | 4.28 fps | 1% |
| Quake 3 (10x7) | 121.9 fps | 127.2 fps | 4% |
| ScienceMark 2.0 (Bandwidth)* | 5378 MB/s | 5851 MB/s | 9% |
As you can see, given almost a 9% increase in memory bandwidth, we saw similarly small increases in overall performance. It would seem that the Athlon 64, at its current clock speeds, just simply isn't starved enough for memory bandwidth to benefit from an increase in bandwidth. You'll also see that the areas where faster DDR memory helped back then are pretty much the areas where DDR2-800 is showing gains today.
Based on our results from back then, if a 9% increase in memory bandwidth doesn't increase performance tremendously, then the 35% increase in bandwidth we see with DDR2-800 on AM2 shouldn't yield any more of a performance increase. Or simply put, yes, our AM2 performance numbers make sense.
Facebook
Twitter
RSS
107 Comments
View All Comments
DrZoidberg - Monday, April 10, 2006 - link
Yeah it is disapointing that DDR2 800 doesnt increase performance by much for AMD. I think we will only see nice 20%+ improvements when AMD moves to 65nm CPUS, smaller transistors less power higher clock speed. Too bad 65 nm seems like Quarter 4 at earliest, next year most likely.I do hope when Conroe is released AMD does big price cuts, cause their CPUS will no longer have performance crown so they no longer have excuse to have their X2 processors more expensive than Intel, so we should hopefully get X2 4400+ for $300, or X2 3800 for low $200s.
Shintai - Monday, April 10, 2006 - link
Even a 300$ 4400+ would be a bad buy. For 300$ You will get a 2.4Ghz Conroe that will be somewhat like an FX-62. So maybe a 200-250$ 4800+ and a 150-200$ 4400+AMD really needs some extremely aggresive pricecuts to be competitive.
abhaxus - Tuesday, April 11, 2006 - link
they don't need to make price cuts yet... when conroe comes out i'm sure they will drop the prices by quite a bit. as it stands, the X2s are by far the best chip on the market and have been for quite some time, and have been reasonably static on price for half a year now. This is the first time in a long time that i remember chips staying THAT static at high prices.That said... this review makes me worry for AMD. I hope they have something up their sleeve otherwise this generation will go very badly for them.
Sunrise089 - Monday, April 10, 2006 - link
20% Seriousely?I'm no CPU expert, but I cannot imagine that kind of gain. Pentium 4's moving to 65nm and 7900GPUs didnt see anywhere near those kinds of gains.
Furen - Monday, April 10, 2006 - link
I think a 20% clock speed increase is conceivable if a) AMD's 65nm shrink goes off well (let's assume a 10% increase due to this), and b) AMD's embedded germanium technique is 10% better than current DSL silicon. Of course, clocks woul not be 20% better until yields hit a decent point.I think that the main way we'll see AMD get closer to a performance parity with Intel will be through the various architectural tweaks in Rev G, though there WILL be some clock speed increase out of manufacturing,.
bob661 - Monday, April 10, 2006 - link
I think I will wait for the die shrink and just get a dual core and some ram for now. I've been trying to decide whether to wait or not to upgrade. I was thinking about waiting for the die shrink anyways.poohbear - Monday, April 10, 2006 - link
"better late than never" is the expression we all know, not "better early than never". wow, anandtech are really trying to sell this cpu in their "final words" section, even though it seems like a waste your money according to your performance tests. i think i'll stick w/ my s939 and just upgrade to a x2 cpu instead of a whole new socket.:/Brunnis - Monday, April 10, 2006 - link
You're just reiterating what Anand wrote. He said that there's no point for S939 owners to upgrade, but that AM2 is the natural socket of choice for those who don't already own an up to date system.Are you suggesting that those people should buy S939 parts instead, despite them having a very limited future and worse performance? That makes absolutely zero sense.
poohbear - Tuesday, April 11, 2006 - link
nope it makes perfect sense. i wouldnt hold my breath for the am2 is what im saying. im sorry but a 5% increase doesnt justify ditching my s939 and opty 144. and what are u talking about limited life? w/ dualcores available on the s939 they're gonna be around well into 2008. It's 2006 and there are still tons of people using athlon xps and agp, so plz drop your enthusiast perspective on the market, it's not realistic of what the avg person has.sp1nfer - Tuesday, April 11, 2006 - link
no, it's EOL (end of life) is Q4'06, with socket 754 holding out one year more. By the time you decide to go X2, with AM2 out and all, prices are going to be higher than AM2 counterparts. AMD said it themselves that prices for s939 will be increased near and on AM2 launch. I think Brunnis covered most of it.
It makes perfect sense.