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
poohbear - Wednesday, April 12, 2006 - link
>> 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.>>>>Huh!? And what does this drivel has to do with whether AM2 or S939 is the best choice for a new computer build in the next few months?
what im saying w/ that "drivel", is that ppl are gonna have their s939 for a long time to come, the fact that lots of avg folks still use athlon xps and have agp, is proof of the previous gen staying around a whole lot longer than the manufacturers would care for. you call it "drivel", i call it proof. whatever, it's just another internet argument.
yes, if u're buying a new system, and u're an avg user, then a s939 will suit your needs fine for the next 2-3 years. If the AM2 is the same price as the s939 when it's released, then ofcourse the choice is obvious, aside from it being a new port and the s939 being a mature and stable port. about the EOL stuff, the Athlon xp has been EOL for a long time, but we can all find those chips easily for dirt cheap, EOL doesnt mean jack all if u know where to find them. thank u, and here's to another 2 years on my s939 rig.:)
Brunnis - Tuesday, April 11, 2006 - link
>> nope it makes perfect sense. i wouldnt hold my breath for the am2 is what im saying.No, you made it seem like Anand recommended AM2 for people that already own reasonably fast S939 systems. The fact is that Anand did the exact opposite. What Anand said is that there's no point in buyin a S939 system for those who have older computers, which makes perfect sense. Don't you agree?
>> im sorry but a 5% increase doesnt justify ditching my s939 and opty 144.
Yeah? And who told you that you should? It wasn't me and it certainly wasn't Anand. I was talking about people with older systems, as was Anand, yet you fail to comprehend what either of us wrote. Way to go.
>> and what are u talking about limited life? w/ dualcores available on the s939 they're gonna be around well into 2008.
So, if I were in the market for a new system you'd recommend me to buy a complete S939 system insted of AM2? AM2 will likely get 65nm K8L CPUs and possibly quad cores. I'm sure S939 is great and all, but it probably won't have much faster CPUs released for it and it's therefore pretty dead. Notice that I'm still not recommending S939 owners to switch on the launch of AM2, or that current 939 systems will be slow compared to the initial AM2 systems.
>> 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.
Huh!? And what does this drivel has to do with whether AM2 or S939 is the best choice for a new computer build in the next few months?
Gnarr - Monday, April 10, 2006 - link
Since we only see a 35% percentage increese in bandwith with AM2, we should see the same gains from DDR2-533 as DDR2-800.DDR2-533 in dualchannel offers 8528MBps, but 6800MBps seems to be the peak bandwith for AM2 as is.
Furen - Monday, April 10, 2006 - link
Yes, DDR2 533 and DDR2 667 benchmarks would have been useful, considering that these are the cheaper types of DDR2. Just because the max bandwidth for DDR2 533 is still higher than AMD's current bandwidth it doesn't mean that it'll achieve the same usable bandwidth unless AMD is currently being bottlenecked by a data link rather than by the mem controller's efficiency.highlandsun - Monday, April 10, 2006 - link
Eh... With the arrival of DDR3 looming for 2007 it makes no sense to me to build a DDR2-based system now. DDR is still about the same price as DDR2, and where DDR2-800 gets a tiny performance edge over DDR-400, that edge completely disappears with DDR-500, or even a mild degree of overclocking.Or at least, performance isn't sufficient motivation to switch, and price doesn't seem to be either. Perhaps power efficiency, since DDR2 is lower voltage.
Calin - Monday, April 10, 2006 - link
If we assume you are talking about an AMD system, then DDR3 won't be a possibility until maybe three years from now. They just moved out of the Socket 939, and into a new socket. Next socket you will see will be in more than a couple of years...Yes, performance and price are not reason to switch now. Power efficiency for sure, if you will buy a low power processor (from 89W TDP to a 35W TDP, that is some 50W of power in best case, and possibly 25W in the worst case).
What other reason to switch could you have? Access to more memory (maybe), I don't know how much memory those new processors could access (more than 4GB? current mainboards seem limited to 3GB). Availability of high performance memory - fast DDR memory will dry up sooner than later. Processors will dry up too.
JarredWalton - Monday, April 10, 2006 - link
Can you point to any DDR3 chipsets for CPUs on the horizon? Intel is sticking with DDR2 for now, and DDR3 is basically only being used in GPUs. It could be several years before we see DDR3 chipsets on motherboards - assuming we don't just skip that option altogether and go some other route.NullSubroutine - Monday, April 10, 2006 - link
ddr3 and gddr3 arent the same things.menting - Monday, April 10, 2006 - link
right now DDR3 is still slated to sample in mid 2007, and announced late 2007 / early 2008 timeframe.DrMrLordX - Monday, April 10, 2006 - link
Word is AM3 will be out by 2007, giving AM2 a product life of maybe one year, maybe less.http://forums.anandtech.com/messageview.cfm?catid=...">http://forums.anandtech.com/messageview.cfm?catid=...
Of course, this is nothing but rumor and speculation, but it's a common rumor. If this is even remotely true, AM2 is nothing but a stopgap measure.