It looks like a neat technological solution for Intel. I wonder how far behind IBM and other are to working production worthy silicon? Kudos to Intel! I too would like to see ARM use this technology (or similar) but more than that I am looking forward to the dual/quad 1G to 3G ARM processors.. someday.
My understanding is that the fins can be paralleled for more current switching capacity, and in the original announcement IAG said that they could swing the process to minimize the off-state leakage by 10x, which IMHO would be phenomenal.
If it's true that so less cost difference is left between the planar and tri-gate, the tri-gate should be the must-be process in 20nm or 14 nm generation, and TSMC and IBM will keep on catching up Intel in the tri-gate process, w/ non-SOI wafer.
It is much like the RISC vs CISC battle of the 1980's. The result was CISC won out in the PC and server market, while RISC took the embedded market. The key deciding factor was software, rather than hardware. The key difference 30 years later remains software. However, this time Microsoft has leveled the playing field with ARM Windows, not to mention the existing Linux compatibility.
Blog Doing Math in FPGAs Tom Burke 15 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 ...