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.
107 Comments
View All Comments
tony215 - Wednesday, April 12, 2006 - link
Who in their right mind is going to fork-out hard earned cash for an AM2 mobo, CPU, & DDR2 if there's little to no gain in performance over AMD Socket 939. This review makes depressing reading for AMD fans looking to upgrade, it's obvious that AM2 has little to offer this year, I expect to see many jumping ship when Conroe is released.clairvoyant129 - Thursday, April 13, 2006 - link
Don't count on many people jumping ships. AMD has a pretty good enthusist fanbase (fanboys mostly). I know Im going to drop my FX-55 for the 2.67GHz Conroe but don't expect many to switch sides... besides, most of these fanboys are probably going to cook up some conspiracy theories to degrade Conroe.Calin - Thursday, April 13, 2006 - link
I might jump ship to AM2 from my Duron 600MHz processor. For me, a dual channel architecture coupled with cheap Sempron processor and onboard video would make a lot of sense - especially a low power Sempron, if such beasts could be found.I am waiting for the AM2 platform (processors, chipsets, mainboards)
tony215 - Thursday, April 13, 2006 - link
Calin, by jumping ship I meant switching from AMD to Intel. Let's see what June brings, if this review is anything to go by I don't expect a massive leap in performance from the AM2 platform over current socket 939, People with money in their pockets now aren't interested in what AM2 /DDR2 may develop into in the year 2007. If Conroe lives up to the hype of offering more bang for less bucks ... I'll make the switch, screw brand loyalty.tekkstore - Monday, April 17, 2006 - link
http://www.tekkstore.com">tekkstore.comphillock - Sunday, January 28, 2018 - link
https://visual.ly/community/Infographics/entertain...phillock - Sunday, January 28, 2018 - link
https://tinyurl.com/yaz5xfx7clairvoyant129 - Tuesday, April 11, 2006 - link
My god, that website has to be the most biased **** I've ever seen. Explain Victor Wong's resoluts on SuperPI 1M (21secs) with a $300 processor with an unsupported motherboard and BIOS. I've never seen such fanboyism in my life. You can't believe anything from that website.Anemone - Monday, April 10, 2006 - link
AM2 may not hold up against Conroe, but until you can buy them in a store, they are vaporware. Intel is late a LOT, so I'm not buying this "don't worry it'll be soon" stuff. Merom prediction: Jan 07. Conroe prediction Oct 06. That's a long, long time for AMD to sell superb chips to a lot of folks.$.02
mkruer - Monday, April 10, 2006 - link
Everyone is bitching how the Conroe is going to kick AM2 @$$ but the facts of the Conroe are starting to leak out. http://sharikou.blogspot.com/">http://sharikou.blogspot.com/ Does this illegitimate Conroe? No, but it does clarify why Conroe preformed so well in the benchmarked apps. and that Conroe is still more memory constrained then the AM2. The AM2 might just get 4MB of Cache per Core thanks to Z-RAM or at least the K8L server editions sure will. Will the real Conroe to step forward?