News Headlines
- Tue, May 21
- Microsoft's Next-Generation Xbox Reveal: Liveblog and Discussion
- Xbox One officially announced, here are the facts
- EA teasing a new Need for Speed with image of a police supercar
- Check out the Call of Duty: Ghosts reveal trailer and Infinity Ward's behind-the-scenes tech video
- Xbox One games confirmed thus far, major publishers like EA, Square Enix and Ubisoft on board
New Articles
Related Articles
Power Consumption

The low-power Athlon II-powered system managed to beat even the Clarkdale, and by a nice margin. There are also some other interesting numbers though; the highest-clocked Athlon II does significantly better than its slower predecessor, thanks to the C3 revision of the silicon.
Here are the low-power processors in a 785G board, the ASUS M4A785TD-V EVO, running the IGP. The Core i5-661, also running its IGP, is included.

Great. The Athlon II X4 low-power with a 785G idles almost as low as Clarkdale, however a load it is a full 28W lower! Underclocking it was not as beneficial as it did on the Phenom II X4 910e. It did reduce its power usage another 3W, bringing it 31W lower than Clarkdale.
Conclusion
In the previous article about the Athlon II X4 635, it was concluded that in gaming, the X4 635 had quite a hard time beating the Core i3-530 in the same price range. With today's launch however, the X4 635 is the new sub-$100 Athlon II X4, meaning that its direct competitor becomes the Pentium G6950. In rendering, its four physical cores were already giving it a great advantage. In gaming this time, it now beats the Pentium G6950 in four tests out of seven. So overall, the X4 635 at its new decreased price is now definitely a great value, either for rendering, gaming or general computing. That is the greatest thing about today's launch.
The C3 revision, this time applied to the Propus die, has made another few miracles. A 100MHz speed bump at the same time as decreased power consumption is very welcome. The new Athlon II flagship wins against both the Pentium G6950 and the Core i3-530 in all rendering and general computing benchmarks. It only loses against the Core i5-661 in WinRAR and Cinebench R10 in these six tests. As for gaming, it is not quite the same story; out of seven tests, it could not beat it one single time, but that was expected since they are not quite in the same price range. It did beat its similarly priced competior in two benchmarks though. So the same conclusion at the time of the launch of the previous flagship can be made; for rendering, folding or anything of this kind, the 3GHz Athlon II is definitely the better bet. For gaming however, the $3 more expensive Core i3-530 will score better.
The other tested subject in this article was the low-power Athlon II X4 610e. If one wants the best performance below the 45W mark, this processor is the right choice. In fact, looking at its amazingly low power usage, Intel provides only two solutions that have a better power consumption. The performance is just not comparable though; these are the single-core Celeron and Atom platform. The Pentium G6950, being a 73W TDP part, should fall around the 110W mark at load, maybe a tad less, which is still close to AMD's offerings. That being said, it outruns the 2.4Ghz quad-core in only two benchmarks out of six in rendering and general computing. As for gaming, it wins four tests out of seven. In other words, the Pentium offers a bit higher power usage, a tad better performance in gaming, but a definitely lower performance in general computing and rendering. The price tag is not quite the same though; there is almost a $60 difference. Motherboards that can house Athlon IIs can be had for more than $20 less than a H55 board though, so that evens out a bit. So really it depends on the needs of the user. The low-power Athlon II, like its brothers, really shines in rendering and general computing at the 45W mark.

- Comment on this article (2)
- check out our other Socket AM3 CPUs articles
- read more AMD Athlon II X4 640 reviews
- visit our Socket AM3 CPUs section

Noticed that the higher res gaming benchmarks tended to even out across the board. Was that because it was GPU bound? Usually the 800x600 puts the full weight on the cpu as the GPU is processing the bare minimum.