If 20nm "was that simple many others would have done it by now."
Done it with what? 4DS has the only area based RERAM tech out there. Everyone else is working with filamentary RERAM - it has developed a suite of patents that protect what it has developed.
It's taken the technology to full PCMO and has got it to fab at 60 nm on commercial equipment - it has commercialised the tech. To take it to the shelves - it does not have the resources or a strategic goal in terms of integration that would be required - it's always going to be an Infineon, NVIDIA, Micron or Samsung that gives that to you.
20 nm has been indicated as the industry benchmark - the company took the decision that the additional upside/reward for getting the tech to that node size, was worth the time/development cost.
We don't know where the issue lies, that is true, but we can look at the problems with PL2 and PL3 in the past for guidance.
Getting the array fabbed and functional has been achievable - it's been getting it to work with the overlying control layers as a cohesive, controllable memory that has been the challenge, whether that is an etch issue or as recently intimated by Ting, a limitation of the IMEC platform.
"If it was that simple, many others would have done it by now"
And yes, 4DS would not exist, the need and market opportunity would not exist. Nothing that is worthwhile is easy.
Add to My Watchlist
What is My Watchlist?
