I can only report what the executives say.
You say no Intel 14-nm parts will ship in 2013 but the end of the 2013 is 12 months away. We will have to wait and see.
And is it a case of Intel can't make 14-nm chips or is choosing to slow down the process introduction because of market issues it sees around uncertainty for its favored computer form factors?
Pulling the engineers home and claiming no 14nm slip is like the other Intel manager saying
"foundry is dead" while Intel is trying to get into foundry business
Many source and data points suggest Intel 14nm is 6 months to 1 year late. No parts will ship in 2013 (original plan on paper)
You may have put your finger on it.
And the issue remains that the longer Leixlip is only running a 65-nm process and does not get P1272 the more chance senior Intel management has to change their mind.
That said Intel Ireland has won a number of manufacturing excellence awards within Intel and did management to beat out Intel Israel for the chance to run P1272. So probably Leixlip deserves its "chance to shine."
There seems to be a shift at Intel from CPUs for PCs to SoCs for Mobiles. 14 nm process may have to be reoptimized for SoCs ( analog, I/O,.. ). This might take some time. Perhaps not much point in keeping 600 Irish stuck in the Desert ( AZ ) during the Holidays while all this goes on.
Not sure about that.
The workers there know that work is pressing on to finish the re-commissioning of Fab 14, which will probably just be labeled as part of Fab 24.
But what they don't know so well is how an incoming CEO at Intel will regard Intel Ireland.
One way that what Eamonn Sinnott said makes sense is if there is NO timetable for when Leixlip starts running the 14-nm FinFET process.
It is just there to follow on when the two U.S. foundries are maxxed out.
In which case, sending a significant part of a work force in training home before the training is complete cannot be related to a delay as without a timetable you cannot have a delay.
And even if Intel started to have second thoughts about ramping 14-nm at Leixlip they would not stop the fab shell commissioning work. Having started they more or less have to finish the work even if they choose to delay putting equipment in.
Blog Doing Math in FPGAs Tom Burke 2 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...