Wilco (Wilco.Dijkstra.delete@this.ntlworld.com) on February 5, 2015 3:27 am wrote:
> Michael S (already5chosen.delete@this.yahoo.com) on February 5, 2015 2:03 am wrote:
> > Wilco (Wilco.Dijkstra.delete@this.ntlworld.com) on February 4, 2015 2:59 pm wrote:
> > > David Kanter (dkanter.delete@this.realworldtech.com) on February 4, 2015 10:56 am wrote:
> > race.
> > > >
> > > > 1. I don't think beating a power-constrained 28nm/20nm TSMC
> > > > processor using a 16FF+ processor is at all impressive.
> > >
> > > Just beating is not impressive, but beating by a huge factor is. These are
> > > far bigger gains than Intel got when moving to their FinFet process.
> > >
> >
> > ARM claims 3.5x increase from 28nm A15 to 16nm FF A72
> > In Intel terms it will be like comparing 32nm Arrandale vs 14 nm Broadwell (Core-M).

> Remember that 16nm FF uses a 20nm process, so from 28nm to 16nm FF is just one full
> process node + FinFet. Ie. it is exactly equivalent to Intel 32nm to 22nm FF.

What you mean to say is that 16nm FF uses a 20nm metal interconnect stack and 16nm FinFETs.

In terms of density, TSMC is about half-way between Intel's 14nm and 22nm. But in terms of performance it is equivalent to 22nm.

I agree this bears mention, but it's not going to change. Intel is ahead of TSMC in process technology, and TSMC isn't really catching up. Samsung might be narrowing the gap, but we'll have to wait until later this year to see for sure.

Right now, my expectation is that TSMC 16FF+ devices will ship to consumers in 1H16, meaning 2Q16. That's a long ways behind when Intel started shipping 14nm chips in volumes >1M/month.

David