Summary: There's No Rush

No matter how much the industry and the media are discussing a possible successor to Flash, nothing changes the fact that it is far from being replaced. For companies and users this means Flash is today a safe investment.

There is no lack of potential successors. In addition to MRAM and OUM, other candidates include the reasonably scalable FRAM (FeRAM), Polymer memory (PFRAM), PCRAM, Conductive Bridge RAM (CBRAM), Organic RAM (ORAM) and most recently Nanotube RAM (NRAM). However, nothing has been decided yet. The Nanomarkets consulting firm characterizes the current situation as "an open race with a slight advantage for MRAM".

The fact that Flash is "good enough" for its current areas of application and is still undergoing rapid development leaves possible successors with only a slight chance of grabbing market share away from the leader of the pack. MRAM and the others will only get a chance to broadly attack Flash's market dominance when the progress of development of Flash slows as feature sizes drop below 50 nanometers. However, that is unlikely to happen before 2008, and until then, the use of MRAM will be restricted to niche markets. For its part, Intel will surely not torpedo its flourishing Flash business with OUM. Instead, it will probably wait for an opportunity to position its new technology effectively against competing memories.

Several years will pass before MRAM and OUM can catch up to Flash in unit counts and sales. After all, Flash itself needed a good five years to outstrip EPROM.

At this point in time, no semiconductor producer can make an accurate prediction of which technology will eventually gain precedence over Flash. Still, the companies are sure that any successful technology ideally has to add the advantages of Flash to those of volatile memories (DRAM, SRAM). This means combining the advantages of high speed and non-volatility with reasonable production costs. And here, without a doubt, MRAM currently has the best hand at the table.