Since Toshiba doesn't make DRAM (or want to play in that market either) and it is highly unlikely for MRAM to replace NAND or HDD, I wonder if Toshiba (and Hynix) are striving to create a new application for MRAM. It seems hard to do without a customer, or when most customers are accustomed to just Flash and DRAM.
Will perpendicular MJT have better scaling characteristics than other approaches? Can it be implemented with multiple layers and multiple bits per cell? These are the types of things I'd be interested in reading about.
Technology, materials, process knowledge improves over time and a revival of MRAM technology (more than current suppliers) with spin-transfer torque magneto-resistance would be welcomed in the market. But let's give the DRAM and Flash technology a big applause since these two technologies have carried the ball for many years.
Blog Doing Math in FPGAs Tom Burke 23 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 ...