AMD To Fab Again?
There is talk that Intel is deliberately delaying introducing its 22nm process because AMD cannot get enough 28nm processors from its foundry partners.
It is possible that Intel’s market share is rising to the point where it is embarrassingly close to the level at which it can be accused of operating a monopoly.
This could explain why Intel recently stopped offering its sales data to TSMC. The less the world knows about the extent of Intel’s market share, the better for Intel.
Intel has offered a reason for not putting 22nm product on the market in April, as planned.
Intel says stocks of notebooks at its PC customers need to be cleared before it can put its new ultra-books, for which the 22nm processors are designed, onto the market.
Another reason could be that the process is not be yielding.
But if, for whatever reason, Intel has to wait before running 22nm wafers in volume while, at the same time, it would want to keep running 22nm wafers in pre-production quantities, then what product could be better for debugging the lines than a nice, regular-structured product like an FPGA?
Hence, possibly, the fab deals with FPGA-makers Tabula and Achronix.
For AMD, with a great product but no high volume 28nm process on which to run it, the situation is frustrating.
AMD first encountered 28nm problems at its GloFo partner-foundry. Now, it seems, there are 28nm problems at its TSMC foundry-partner.
This is all putting pressure back on AMD to give up being fabless and go back to operating fabs again.
How on earth can it afford to do that?
Well governments are pretty good about putting up money for fabs, while Nvidia, Altera, Xilinx, Qualcomm and others may be getting sufficiently tee-d off with flaky 28nm availability to consider a wafer fab consortium.