In a notebook - the processor is the least of concerns. The gpu and the battery life are paramount. The gpu is critical not only for graphics, but because the HD decoding is typically offloaded onto the gpu.

The cpu isn't a factor anymore.

The Republicans have made such a mess of the economy, it seems whatever Intel releases will fall on poor corporate ears and not contribute to any windfall of sales.
For the next couple of years, the VALUE segment should prevail, so AMD would do well to step up, and act accordingly.

I still don't see why the billion dollar OEMs don't have the sack to stand up to the bully. Haven't the seesaw prices\platforms done enough damage?

I've been holding off buying a laptop because of the significant shift in the performance/power ratio expected in the mobile Nehalem chips. I'd expect many, many other people are too. If the chips and then laptops slip by two or three months, it is very likely that people like me will just hold off purchasing until the mobile Nehalem based systems do ship.

If the laptop manufacturers think sales are bad now just wait until July/August/September if they push the next generation out. The sales during that time frame will be zero for any group of people that don't *HAVE* to buy.

July/August is the back-to-school season, which then feeds into 4Q holiday (consumer) and end-of-fiscal-year (corporate) sales. October, even September, is too late for that -- which of course is why the OEMs want it, so they can sell their existing inventory into the strongest demand. (Assuming there is any demand in '09, ie economic conditions have gotten better, not worse).

But Intel was going to miss that window with the mainstream Auburndale anyway; Clarksfield, being a quad with higher TDP, was going to be a bit of an niche chip after all -- much like the i7 vs the i5. Pushing back to October might just mean launching them one quarter apart instead of two, or even just launching both together. Or, yeah, maybe just delaying the whole thing and going to 32nm instead.

On the other hand, Intel could push back against the OEMs and stick to whatever schedule the feel like -- and if they slip, hey, they can say they're doing it for the OEMs.

I thought the official party line for Calpella was Q3. So they could probably slip to the last week of September and still be "on schedule".

My personal preference is to just skip 45nm mobile Nehalem completely and going directly to 32nm Westmere which is due Q4 anyways. Yes that does leave a bit of a gap, but I'm sure an aggressive launch on 32nm would make up for any drop in market share. 32nm is probably needed in terms of power consumption and cost to usher in mainstream mobile quad cores anyways. With vendors requesting a delay to Q4, this may well be a reasonable option.