esrever :
hcl123 :
esrever :
28nm steamroller is so late. Its been 1.5 years since 7970 on 28nm came out at tsmc and we still aren't even close to steamroller. Hopefully Glofo can can keep production up and give decent yields.
That is consequence of all the turmoil with GloFo and AMD getting out of that partnership. At the time of Dirk Meyer i suspect GloFO with AMD direction was pushing hard SOI tech for 32 and 28nm. A new management was imposed by the principal owners of GloFo and all went in reverse. The epilogue of the drama finished last year with AMD getting out of GloFo entirely. But the last management of GloFo also went bust, its was all about bulk "mantra" and SoC even in the underwear, being the superior SOI tech even regarded as inferior... yet it didn't managed o get one single top IDM implementer of the ARM SoC armada, for the their *supposed* superior HKMG bulk high performance 28nm process ( or the low power one for that matter).
Qualcomm was there, tested, tasted, saw it, saw what they did to AMD... and said goodbye...
I think GloFo re-gained a lot more respect for AMD, since the notice was clear that they could even lose the only top IDM they had left, and no prospects of gaining another one in short order. That is why the old 32nm PD-SOI was revamped almost 15% and Richland happened... that is why the FD-SOI partnership with STMicro was possible... with all that, is even possible that the FX/server variants of Steamy is 28nm PD-SOI.
With all this "bulk" dance decisions(even at AMD) and SOI improvements, its no wonder things got delayed.
That makes a lot of sense but I don't expect SOI to be viable going forward especially going into finfets and trying to reduce power consumption. I thought both steamroller and excavater will be on bulk. I do wonder if they can even get much performance gains from 32nm SOI to 28nm bulk.
What is not viable is "plain planar bulk", 20nm is pretty much the dead end. There is a reason why Intel in his eagerness to be first choose finfet, and complicating for complicating was not the reason. From now on Fully Depleted techs will be the norm, it can be accomplished by UltraThin Body and Box(UTBB) or FD-SOI in short, or by finfet techs.
Just read some news http://www.advancedsubstratenews.com/
There are pretty good studies and heavy weights. IBM as example is clearly showing that Finfet on SOI is quite better than Finfet on bulk, and they don't do it because of "fanboyism" or propaganda, IBM doesn't sell or has stakes in SOI wafer suppliers.
FD-SOI, like with STMicro with its 4 core ARM chip at 3Ghz demo, at most probably yet less than 5W(which is already too much for smarthphones), is showing that FD-SOI is not only better for low power but can be also better for high performance. Cherry on top of cake is that FD-SOI can be less expensive to fab than finfet on inexpensive "bulk" substracts (specialty below the 28nm size) and provide quite better yields (much less variability with transistors).
I suspect, (betting on it) that the next high performance process of IBM (the one that delivers "commercial" chips above 5Ghz) will be FD-SOI.
Finfet can be the only option below 10nm because that thin channel bodies for FD-SOI will not scale further, but then it will be finfet on SOI not bulk, specially for high performance, if not for all, since finfet of bulk can reach the dead end below 10nm to, even for low power offerings.
Intel pushes hard the bulk mantra because it doesn't want to depend on others for anything, including wafer suppliers, and their requirements for volume is quite above anyone else. But i believe nothing prevents intel to license pertinent techs and do their own SOI wafers, which i believe will happen eventually.
Foundries just love "bulk" because it means higher profits for themselves... not better solutions for costumers.