16/14nm First generation Yield ramps was kicked off in 2014 and should be GREAT by now in 2017 especially for GF who licensed it from Samsung. The game there is no longer Yield but how to manage price pressures / cost management etc.
It is no longer certain what smaller metal pitch / fin pitch and gate density gets us today.
In the past , smaller meant better characteristics HOWEVER with the demonstration of TSMC's 16nm beating Samsung's 14nm in performances despite being having bigger dimensions, the benefits of scaling is no longer evident from a performance standpoint.
From a consumer point of view, I would prefer my battery life to last longer & have faster Applications speed .... and if a big clunky node does the trick , I would go with that vs the company with the smallest pitch , fin , gate etc.
P.S. I just want to watch videos the whole day if possible on my handphone. ( consumer whining )
I agree with you. Intel 10 nm is best, the name has no relation to what it is anymore. I do think glofo 14 nm yields are probably decent. You know why? It's because the fell on their face and broke it when when they tried it themselves. Then they had to tuck their pitiful tail between their legs and go pay Samsung for the process they are running. So if you go buy a process from someone that knows what they are doing you are going to be ok. It's just an exercise in copying. Now they are once again forging out on their own. I doubt if they will do anything impressive here. They will be very late and TSMC will have already captured the lions share of the best 7 nm accounts. Glofo is bleeding cash badly and only being held up by the mubudala. They probably also got some cash from the kungfu deal but who knows how much. I think they are in big trouble though and mubu wants to flush it down the toilet.
Intels 10nm offering is better than every foundry's offering on 7nm. You have to know that these node names are mere marketing gimmicks now. contacted metal pitch, fin pitch and gate density are not the right indicators for a more superior process. As to what the GF yields at 14nm are, I can assure you they are very competitive.
TSMC has and will continue to ram the Globalfoundries face into the pavement. As a matter of fact, I just read that Qualcomm has switched from Samsung to TSMC at the 7nm node. Samsung foundry is going to be in desperate condition after that. Time is rapidly running out on the glofo. Just go read the glassdoor reviews for that sewer. I was reading there that the 7nm was a complete fantasy. But I'd really be interested to know about who has actually made the biggest mistake of their life to trust the glofo disaster at the 7nm node.