few companies using Intel notably as foundry for 22nm do not have any advantages using Intel 22nm over TSMC/Samsung bulk 28nm?
Means if foundry product requires a full range of transistors (low leakage,RF, quality analog etc.) for a mobile 22nm SOC, that is not offered even for internal intel product until end of 2013. Mobile SOC intel foundry customers thus at a disadvantage compared to tsmc and other foundry.
I can't think of any way that Intel's 22nm process integration is cost competitive against the foundry cost. Unless the finFET yields at the foundries are really bad ( a possibility since it would be their first generation)
Globalfoundries roadmaps are fantasies as befits a company that is associated with AMD. There are no shipping G-F 28nm products yet so talking about 20nm as if it was a shipping product too is ridiculous. G-F like Samsung has to change from gate-first to gate-last yet which is a whole new technology/yield learning curve. Both these companies rely on IBM process development which has proved inferior to Intel's over decades. PD-SOI, gate-first HKMG were all costly mistakes.
As the next step to accelerate the development and commercialization of EUV lithography tools, ASML Holding (ASML) has made a bid to acquire light sources provider Cymer (CYMI). When ASML roped in Intel (INTC), Samsung and Taiwan Semiconductor Manufacturing Company (TSM) earlier this year by selling a 23% stake in the company, we did wonder how this cash would be used. Since Cymer has been collaborating closely with ASML over the past year, the deal make sense.
The EUV Roadmap
ASML provided some details in its press release. The company currently has 6 EUV systems (NXE: 3100) in beta, with satisfactory resolution performance at 22nm. However, the successor system, NXE: 3300B, which is intended for 14nm production currently requires closer integration with light sources to drive targeted levels of efficiency.
Joint laboratory testing of the NXE: 3300B by Cymer and ASML shows ability to process 18 wafers per hour. The companies intend to take this to 69 wafers per hour for 2014 chip production.
For this purpose, ASML intends to have 11 NXE: 3300B systems installed at customers for testing in 2013, with production-ready shipments at targeted efficiencies starting in 2014. The company has received four commitments so far and expects another four to eight in the next six months.
Now do you folks get it now?
here is more facts:
GlobalFoundries has added a 10nm finFET process to its roadmap and expanded its technology platform offerings. The foundry vendor plans to go from 20nm planar in 2013, to 14nm finFET in 2014, to 10nm finFET in 2015, and 7nm finFET in 2017.
At the SOI Consortium’s event at IEDM, Jeff Watt, a fellow at Altera, presented an evaluation and benchmark of planar fully depleted SOI technology. A simulation showed that 20nm FD-SOI provided a 5X reduction in power over 28nm bulk, Watt said. However, Altera has not made a commitment to SOI for FPGAs and is currently evaluating the technology, he said. “We are looking at all options,” he said. For 20nm, Altera plans to use a bulk technology at TSMC. At 14nm, the FPGA house will likely go with bulk finFETs at TSMC. However, Altera is also exploring SOI.