DRAM faces a not very early death

It’s too early to declare that DRAM (dynamic random access memory) is dead or even dying, but the writing is on the wall and it eventually will be displaced.

So thinks senior semiconductor analyst Malcolm Penn, writing in the latest issue of Future Horizon’s chip bulletin.

Says Penn: “All of these new advancements are in various stages of development and it will be some time before many of them are in a position to challenge current mainstream memories for design slots in PCs and consumer products.”

Molecules, he says, even big molecules, are small compared to today’s memory cell architecture. “At its simplest, the addition of an electric charge through electrons can make the molecule take a negative charge and an absence of electrons can lead to a positively charged molecule.”

But the ordinary connection circuitry needed poses some problems. Molecules called Oligo Phenlylene Ethynylene Molecules (OPEs) can switch but the switching isn’t predictable yet.

Research on this type of molecular memory right now is in academic labs or in the labs of large corporations, says Penn. HP, Penn State University (no relation) and ZettaCore have all formally announced they’re working to solve the problems. NASA supports research for non volatile computer memory. And Rice University, says Penn, is working on graphene memory with potential cell sizes near to 10 nanometers

According to Future Horizons, Unity Semiconductor will produce passive re-writeable cross point memory array that doesn’t need transistors for a memory cell and it will bring a 64 gigabit product to market by the end of next year. The memory is a non volatile, multi-layer, multi-cell memory with a write speed five times faster than mainstream NAND.

Future Horizons is here .