Richard Cownie (tich.delete@this.pobox.com) on February 2, 2013 5:05 pm wrote:
> Ricardo B (ricardo.b.delete@this.xxxxxx.xx) on February 2, 2013 4:05 pm wrote:
>
> > No, they can do that integration for anything they can develop and fabricate
> > in house, provided it generates enough revenue to sustain the R&D.
> > Ie, they can't do that integration for a ARM core from ARM, but they can do it if they develop their own.
>
> They can tune the circuit and microarchitecture to the process for anything,
> with enough effort. They can't simultaneously tune the process to be optimal
> for two different purposes. So it would be unreasonable to expect the same
> level of global optimization that they can apply to their main business, x86.

You can't have an optimal process for everything, but you can have more than one variant (which everyone usually has).

Intel has historically managed to have competitive low power processes; they have competitive 32nm SoC process and they've just shown a competitive 22nm process variant.
And given the their current focus on power efficient x86, the trade-of between their low-power/SoC process and their main high performance process is smaller than ever.


> Other than Apple and Samsung, those other companies are barely making a profit.
> So yes, they'll buy their SoC's from a variety of suppliers, but I think not
> at margins that would look attractive to Intel if the products have to come from
> near-leading-edge fabs. They can't afford it.

Do you have profit margin figures for Samsung and say, LG's phone/tablet business?

Also, margins are not a fixed mater.
In the end of the day, what defines them is how much your client is willing to pay and much it costs you.
Intel's margins in the x86 business because they manage to deliver a win-win: better products (performance, power) with lower production costs (area, yield) than the competition.
If Intel can deliver the same double-whammy for smartphone SoCs, then they'll be able to get higher margins than current SoC manufacturers.