Zeev,
I work in embedded systems and MRAM could be huge if it worked the way I want it to, since right now we usually store the code in flash and uncompress it into DRAM to run the code, and would greatly benefit from having one memory device that is fast and nonvolatile. However, in order for this to happen we must be able to read from/write to it essentially an unlimited number of times; anything less and I don't see how it's usable in the majority of applications. Unless manufacturers time it so the MRAM dies right after the warranty runs out :)
As for Monday, I guess we'll see what effect CY's statement that "MRAM development is taking slightly longer than expected" will have. I'm wondering just how much longer "slightly longer" really is. That's probably more important than the fact that CY sold all their NVEC shares.
Market Data 
Markets 
AI
