You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please, join our community today!
UMC will start producing 14nm FinFET in the first half 2017, however they only plan on 2000 wafer capacity initially. It seems to be starting so low because they haven't found much customer demand for it yet. Maybe, it's because type of customers that go to UMC want cost effective solutions and 14nm is not cheap?
UMC will start producing 14nm FinFET in the first half 2017, however they only plan on 2000 wafer capacity initially. It seems to be starting so low because they haven't found much customer demand for it yet. Maybe, it's because type of customers that go to UMC want cost effective solutions and 14nm is not cheap?
I will check on this next week when I am in Taiwan but I see some serious challenges ahead. UMC is primarily a second source for TSMC. All of their processes are "T like" which means you can take a TSMC design to UMC without change.
It has not been confirmed that UMC 14nm will be "T like", in fact, I have been told it is not.
TSMC is in their third generation FinFET process (16FFC) which is very low cost and capacity will not be a problem now that Apple and others are moving to 10nm.
UMC also skipped 20nm so 14nm will be their first double patterning process as well. UMC does have some big customers and powerful partners (ARM) so I'm sure it will get done but the question is when? 1H 2017 seems a bit early to me.
Yeah, so no chance of "second source" for 14nm - that's a good point.
So far everything that UMC said would happen, has happened, so I don't see any reason to doubt them when they say it'll be in pilot production in 1H 2017.
I don't know about the origonal schedule.. that was a long time ago. More recently, for example the 28nm HPC was delivered exactly when they said it would be.
