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
Furen - Monday, April 10, 2006 - link
Hardly a "common rumor", it's an Inquirer rumor that has been given a lot of circulation but I have yet to see another publication confirm it.jones377 - Monday, April 10, 2006 - link
The bandwidth numbers are better now but STILL dissapointing. Intel will probably get about the same or better number from a 1333MHz FSB using DDR2 667. In the past, AMD always got a little better bandwidth out of the same speed modules. Are AMD sandbagging or do they have problems with their DDR2 controller?tk109 - Monday, April 10, 2006 - link
So they basicly have no answer to Conroe...rqle - Monday, April 10, 2006 - link
yes we do, we still have that "previous generation vs. new generation" quote we can always used. LOL. just like AMD64 vs northwood/presscott.But really though, i assume the amd crowd should be a little smarter and get over that quote.
phaxmohdem - Monday, April 10, 2006 - link
I get a bit confused as to what generation is what, and where you draw the line at a generation change for instance:AMD -> Intel
486 -> 486 <--Gen 4
K5/K6 -> Pentium <--Gen 5/6 vs. Gen 5
K6-2/3 -> Pentium II <-- Gen 6
K7 Athlon -> Pentium III <--Gen 7 vs. Gen 6?
K7 Athlon XP -> Pentium 4 Willamette/Northwood <--Gen 7
K8 A64 -> Pentium 4 Prescott <-- Gen 8 vs. Gen 8??
K8 A64 -> Conroe/Core <-- Gen 8 vs. Gen 9?
Perhaps someone can shed some light on this.
Shintai - Monday, April 10, 2006 - link
How did Prescott become Gen 8? Its a northwood with 64bitConroe is Gen 8. So your table should say K8 (Gen8) vs Conroe (Gen8)
Furen - Monday, April 10, 2006 - link
LOL, how do you define a "generation"? What kind of changes indicate that you're talking about a new generation? The K8 is remarkably similar to a K7, so is it a newer generation or just an improvement on the same one? Does it even matter?The K8 has been on the market for close to three years now, so it's completely understandable that a brand new architecture will give it a run for its money. I'll say this, though, both the K8 and the Prescott came out at around the same time, so considering both the same generation is reasonable.