Great find @retiredyoung. Very informative article but as you...

Great find @retiredyoung.

Very informative article but as you say it is still guessing and their best guess is wrong if you believe what Micron/Intel told people on that launch video.

From your article:

"I'd say CMOx is more likely that CBRAM, but less likely that PRAM. My money is still on PRAM. But if I had to have a backup bet, I'd agree with "mgsporer" that this a plausible alternative technology given the cross licensing agreement."

The author thinks PRAM is the most likely bet for the 3DxPoint memory.

From your article:

"So we've come this far. We now know that Intel and Micron are almost assuredly using PRAM. But what is PRAM actually? Well PRAM -- as the name implies -- stores data in the form of a phase change to a tiny atomic-level structure. Specifically, a piece of special glass known as chalcogenide glass (chalcogenide refers to elements in Group 16, such as sulfur, tellurium, and Tellurium)."

If you listen to the launch video below in the question and answers session (32min 00sec) it is clear they are not talking about phase change memory. They get a direct question on phase change memory and say it is very different to phase change but they admit Micron "itself has some experience with [phase change] in the past".

I'll have a re-read and make some more comments later but the one thing that stood out was the fact that Micron has form when it comes to patents.

From your article:

"Micron was sued by Rambus over various memory technologies around 2011-2012. Micron and fellow defendants managed to invalidate some key Rambus patents, effectively winning the case. But in the end many of the defendants -- including Micron -- settled with Rambus and agreed to pay lesser licensing fees to make Rambus go away (more or less)."

Many of those patents listed in that article relate to the computer architecture around the memory cell. Weebit's main patents go to the very core of the memory cell and none of the patents in that article as far as I can see are addressing the material (memory cell) itself. I'll have a closer look but I don't think that article is revealing any of the core patents. It would seem likely if you believe your article that 3Dxpoint might use selectors (diodes) based on the work of Nishant Chandra.

From your article:

On the flip side Intel did hire Nishant Chandra, who developed among other things a manufacturing method for "top-down etched silicon nanopillar Schottky diodes" and "a process for bottom-up fabrication of germanium nanowire diodes during his PhD work. But Chandra is working at a process engineer in Hillsboro, Oreg. Intel and Micron are reportedly designing and manufacturing Xpoint at their joint facility in Utah, as noted by The BBC. So that's one strike against that lead.

The selectors are part of the memory "sandwich" layer and provide the switching voltage to write and erase the memory from what I understand. As mentioned in my post yesterday the tantalum (V) oxide memories that professor Tour is working on don't seem to need selectors. I'm not sure if this is the same for the silicone oxide memory. The complexity and cost without this extra "selector" layer would be very much reduced one would imagine. We need to find out if the silicone oxide memory also provides this advantage. Got to go back and read some of the patents again.

Eshmun