All Flash processes use 'thick' oxides (compared to the oxides that would be used for logic circuits). This is for the simple fact that flash stores charge on a floating gate, and that oxide between the channel and the gate (as well as the control gate and the floating gate) cannot be too thin (else the precious charge will leak off, making taking the "non" out of non-volatile memory).
Intel's StrataFlash is complex, as it requires additional circuitry to verify the amount of charge stored on the gate (or, more technically, the threshold voltage shift). This offsets the benefit of storing multiple bits in a single cell. Cool idea, none the less. Data suggest Intel knows how to make money, so I can't believe there is no benefit in their design.
I'm surprised at the accusation of AMD's flash being "hopeless," as it was the primary money maker for AMD for many, many years. Could be that their technology does not scale well, however. I doubt the author of your quoted e-mail knows, either.
Whatever the case, I think that any flash company can develop a flash process to utilize thicker oxides, or dual-thickness oxides, or whatever is needed. I don't know anything about SSTI, but the message sounds like hype to me.
Steve |
