The number of bugs in a chip is relatively proportional to the number of transistors - Bob Colwell, former Intel chief architect
SAN FRANCISCO: CHIPMAKER Intel said it can't bet on extreme ultraviolet lithography for its upcoming 10nm process node.
Intel recently invested billions in ASML to fund research into 450mm wafers and extreme ultraviolet lithography, but it's not banking on either technology being ready for its 10nm process node. Mike Bohr, director of process architecture and integration at Intel said extreme ultraviolet lithography is a technology that the company is extremely interested in but it simply isn't ready.
Bohr said extreme ultraviolet lithography is "not quite ready for manufacturing or meeting cost goals". He also said Intel is "not betting on extreme ultraviolet for 10nm", which is expected to tip up around the 2016 timeframe.
As for Intel's research into 450mm wafers, Bohr said that the larger wafers are somewhere between four and five years away, once again making the technology unlikely to hit in time for Intel's 10nm process node.
Bohr confirmed that Intel's 10nm process node can use the firm's Tri-gate transistors but added that the firm will need to introduce incremental technology on top of Tri-gate for 10nm. With Intel not being able to rely on extreme ultraviolet lithography, Bohr confirmed that Intel already has immersion lithography working at 10nm, but said it is not clear whether that will provide economically viable chips.
What Bohr's comments strongly suggest is that if Intel is to maintain its goal of making ever cheaper transistors with the 10nm process node, it needs to look away from extreme ultraviolet lithography and 450mm wafers for the time being, unless ASML produces a minor miracle. µ
Sign up for INQbot – a weekly roundup of the best from the INQ