Quote Originally Posted by marten_larsson View Post
Maybe not:
http://www.planet3dnow.de/vbulletin/...#content_start

Also, what I found out reading Anandtech's review was that the original Phenom used a CnQ that clocked down each core but now the new Phenom uses the same state for all the cores based on the fact that the thread is moved between cores all the time. Why, oh why did they do that? Really, the problem here isn't AMD's but Microsoft's. I understand that there were some bugs with this also but the idea was better than what they have now... (Which is also present in i7 if I remember correctly.)
I don't know if it is Microsoft's fault or is the AMD design flawed. I have a Turion X2 Ultra having that independant core technology, and when I'm playing a game that uses 1 core in total or sometimes 1.5, either one of the core or both is running at half its speed so the game is very choppy. To avoid this, I have to set affinity to one core only which fixes the core to maximum frequency.
I think there is still room for improvement on that design, especially in how Vista manages threads. So for now, since PII is not for mobile, AMD avoids the huge penalty hit with the technique. But the technology is still there if the core isnt used at maximum potential.