Socket-AM2 Performance Preview
Without major architectural changes to the new AM2 CPUs, we wanted a quick and easy way to showcase the performance differences between AM2 and Socket-939. What we've got is a massive table below with all of our usual CPU benchmarks and their results for the same CPU in both Socket-939 and AM2 varieties, and the performance benefit offered by AM2:
| Benchmark | Socket-939 (DDR-400) | Socket-AM2 (DDR2-800) | % Advantage (Socket-AM2) |
| PC WorldBench 5 | 115 | 115 | 0% |
| Business Winstone 2004 | 23.3 | 23.2 | -0.4% |
| Multimedia Winstone 2004 | 38.4 | 38.9 | 1.3% |
| SYSMark 2004 | 220 | 224 | 1.8% |
| ICC SYSMark 2004 | 282 | 286 | 1.4% |
| OP SYSMark 2004 | 171 | 175 | 2.3% |
| 3dsmax 7 | 2.38 | 2.38 | 0% |
| Adobe Premier Pro 1.5 (Export w/ Adobe Media Encoder) | 130 s | 128 s | 1.5% |
| Adobe Photoshop CS2 | 210.6 s | 210.3 s | 0.1% |
| DivX 6.1 | 11.6 fps | 12.0 fps | 3.4% |
| WME9 | 35.2 fps | 35.6 fps | 1.1% |
| Quicktime 7.0.4 (H.264) | 3.63 min | 3.63 min | 0% |
| iTunes 6.0.1.4 (MP3) | 43 s | 43 s | 0% |
| Quake 4 - 10x7 (SMP) | 111.3 fps | 117.4 fps | 5.5% |
| Call of Duty 2 - 10x7 | 59.3 fps | 60.1 fps | 1.3% |
| F.E.A.R. - 10x7 | 92 fps | 94 fps | 2.1% |
| Multitasking Test (LAME + WME + Anti Virus + Zip) | 216.3 s | 213.4 s | 1.4% |
| ScienceMark 2.0 (Bandwidth) | 5007 MB/s | 6805 MB/s | 36% |
| ScienceMark 2.0 (Latency 512-byte stride) | 53.83 ns | 49.77 ns | 7.5% |
We'll start at the bottom of the table and go up from there. Rev F processors feature a 128-bit DDR2-800 memory controller, which works out to offer a peak theoretical bandwidth to/from memory of 12.8GB/s. As you can expect, that's twice the bandwidth of Rev E CPUs' 128-bit DDR-400 controller at 6.4GB/s. Thus to see a 36% increase in memory bandwidth according to ScienceMark is to be expected, albeit a bit on the low side. The old DDR-400 memory controller is able to deliver 5GB/s out of a maximum of 6.4GB/s, but now we're only seeing 6.8GB/s out of a maximum of 12.8GB/s with AM2. This however is a huge step for AMD, as it is the first spin of the Rev F silicon that we've been able to see such a significant advantage in theoretical memory bandwidth over previous DDR-400 cores.
What's even more important than the increase in memory bandwidth is that access latency has been reduced by 7.5% over the DDR-400 memory controller in the Rev E cores. Lower latency and more bandwidth means that, at bare minimum, performance won't go down. At least, not perceptibly: .4% slower in one test that has a 1-2% variability is nothing to worry about.
It also doesn't guaranee that performance will go up, as you can see from the results above. If we only count the overall SYSMark score and leave out the synthetic tests, the real world performance advantage averages out to a little under 1.3%. There are some special cases such as Quake 4 and DivX were performance goes up fairly reasonably, which can be expected since both of those tasks are fairly bandwidth intensive and make good use of both cores. However similar benchmarks, such as F.E.A.R. and Windows Media Encoder 9 show lower improvements, so it is very dependent on the specific application and workload.
It's important to note that until recently, AM2 samples were not able to produce scores even on par with Socket-939, so the fact that we're seeing a performance increase at all is a major step from where we were just a couple of months ago. The real question is, is this all we get?
Facebook
Twitter
RSS
107 Comments
View All Comments
AdamK47 3DS - Monday, April 10, 2006 - link
Conroe WILL perform better than AM2 no matter how much spin people try to put on it.bob661 - Monday, April 10, 2006 - link
How can you say this when there are exactly NO shipping Conroe parts? I THINK that Conroe will be competitive but even that opinion is speculation at best. YOU won't know shit about Conroe until it shows up at Newegg!Shintai - Monday, April 10, 2006 - link
If your logic applies, then he wont know anything about Conroe until he got it at home and working. So get over it, even the ES samples out in the wild kicks AMD so hard. The only question was if AM2 would bring extra performance to compete against Conroe, and it surely didn´t. Conroe prices also leaves AMD in the utter dust along with performance. 300$ Conroe E6600 chip or a 1200$ FX62? And the E6600 will be faster in most situations. You gotta be some extreme hardcore fanboi not to go Conroe.bob661 - Monday, April 10, 2006 - link
There's nothing to get over, asswipe. You nor he knows for sure how Conroe performs, period! You can fanboi me all you want. Facts are facts. When they ship and there are 3rd party benchmarks on 3rd party machines tested, then we'll all know for sure how they'll perform. Puff, puff, pass man, puff, puff pass.Shintai - Monday, April 10, 2006 - link
http://vic.expreview.com/read.php?1">http://vic.expreview.com/read.php?1http://vic.expreview.com/read.php?2">http://vic.expreview.com/read.php?2
http://vic.expreview.com/read.php?3">http://vic.expreview.com/read.php?3
http://vic.expreview.com/read.php?4">http://vic.expreview.com/read.php?4
Now cry me a river again and keep whining over something new.
AnnonymousCoward - Monday, April 10, 2006 - link
That's the crappiest website ever! I think a modem is hosting it. Plus it's retarded how the pictures shrink after they're loaded.Furen - Monday, April 10, 2006 - link
I wouldn't compare the pricing quite yet. AMD's AM2 pricing is for May 5th while Conroe's is for launch day... whenever that happens (and before you say June/July, Intel itself said that it would launch as close to the beginning of Q3 as possible but did not commit to an early Q3 launch).Another thing, I have yet to see an E6600 being tested in "most situations", so until I see so, I'll say that your assumption that this is true is a bit irresponsible and fanboyish. Especially so if you consider that neither of the two CPUs (AM2 K8s and Conroes) can be bought quite yet, so convincing someone that one is a better deal is a bit premature. Personally, I think that AMD is going to get its ass handed to it by Conroe but I wouldn't go our of my way to swear it.
redbone75 - Monday, April 10, 2006 - link
Wow. Only twenty posts until someone put up something about Conroe performance. In case you haven't been reading AM2 news or even read the article, AM2 is supposed to launch in June and here we are in April with a preview that, for the most part, is pretty disappointing for anyone that had high hopes for a Conroe challenger. This is just like when Intel migrated to DDR2: it wasn't really necessary but it will give AMD experience for when they can really use it. However, I'm thinking it won't take AMD as long to see more noticeable performance gains with DDR2 than Intel. Regarding Conroe, any way you call it the chip is going to kick some serious @$$, especially if Intel doesn't have any problems ramping up the clock speed. Also, Conroe is still several months away, so unlike AMD, Intel still has some time to tweak the chip for even more performance for its scheduled launch time (won't say date because there is none yet), so who is to say that the current performance claims are bogus. Even if they are now, which I seriously doubt, there just might be enough time for Intel to live up to the hype anyway.MrKaz - Monday, April 10, 2006 - link
I'm not sure,but I think anandtech tested it at 533Mhz again, or
the processor is locked at 533Mhz.
Why did anandtech do the test only at 800Mhz?
Why didn’t test 533Mhz and 667Mhz DDR2 modules?
Because looking at those numbers:
-DDR2 533 will achive less bandwidth than DDR400?
-The latency of DDR2 is lower than DDR1?!?!
-Is the processor already full, so doesn’t need more bandwidth, and only at (theorical) 4GHz and beyond will use it?
defter - Monday, April 10, 2006 - link
At higher speeds yes. Of course 3-4-3 @ 400MHz (DDR2-800) will offer lower latency than 2-2-2 @ 200MHz (DDR1-400).