Re: So saying AMD has no new core until 2009 and Intel does is not really fair until we hear more details of both NGA and K8L.
That's not really what I was saying. It's like this:
If you have enough design teams and can pipeline their design schedules, you ought to be able to offer an enhanced micro-architecture every 2 years, assuming you stagger the feature list. In AMD's case, they are introducing new cores every odd numbered year. In 1999, they introduced K7, followed by the Palomino enhanced K7 in 2001, followed by K8 in 2003, followed by the dual core enhanced K8 in 2005, and we now expect the "L" enhanced K8 in 2007. In their even numbered years, AMD tends to compact their new designs using new processes. In 2000, they moved to a 180nm process, in 2002 they moved to 130nm, followed by 90nm in 2004, and we now expect 65nm in 2006.
The only comment that I have shared with Chipguy is that based on this schedule, their most likely target date for a "new" core (not K8 derived) is 2009, and sure enough that aligns with Hester's comments and the 3-5 year CPU development pipeline.
And as I already pointed out, this says nothing about AMD's competitiveness in the mean time. As a matter of fact, I agree with your comments about not knowing the potential benefits of NGA and K8L, which could influence the performance crown either way in 2007.
News 
Market Data 
Discover 
