Linus Torvalds (torvalds.delete@this.linux-foundation.org) on May 19, 2013 12:19 pm wrote:
> Michael S (already5chosen.delete@this.yahoo.com) on May 19, 2013 11:39 am wrote:
> >
> > Sounds very LRB. Not at all like a Silverthorne.
>
> So I have no real inside information, and I may have exaggerated things a bit, and it was
> all complete conjecture anyway (but conjecture that explains what we saw with Atom).
>
> But I'm interested: why do you think the original Atom core is a much better core than LRB?
>
> Yes, yes, LRB had the magical vector extensions and obviously the whole "lots of cores" thing, so it was
> very different and much more ambitious than Atom was, but I think they are similar in the sense that they
> started out as small projects by a couple of people, and took a lot of shortcuts due to that.
>
> And my real argument was that the reason the original Atom was in-order wasn't that it was a better
> and more power-efficient design, it was because it was a short-cut for a project that didn't have
> the resources to do anything more. Then it kind of stayed that way (much too long).

The thing is that Intel already knows how to do good, performant, power efficient cores. There is two decades of experience there, you don't *need* to go back to Pentium core.

They could have started with a pentium-m ish core, and probably had a much easier ability to take improved structures from later cores and even share their own improvements back to the high performance line.

I think the reason they did not do that is because management is petrified about cannibalizing their own high margin lines.