Quote:
Originally Posted by monir
Something less somber:
New Device Could Revolutionize Chips
We've reached the theoretical limitation a while back with how tiny individual transistor can be at chips making. This new innovation, memristor, is apparently going to take that limitation to an entirely new level by at least a factor of two. It also resolves the central issue that limits the transistors at chips making to a certain nano length: over-heating. Who is excited?  |
Either the article is not very scientific, or the chipmaker lacks some decent knowledge in biology. Memristors are not synapses for a simple reason... they cannot freely build up links to neighbouring memristors/synapses, which is a very important ability for the data/processing network in brains.
And the comparison of viruses 100 nm and todays transistors 40-50 nm and memristors 3nm was so stupid... the virus holds a whole RNA full of much more information then any of the listed devices (also RNA is 4-state while a single bit of information in transistors and memristors is usally limited to binary) on the other side it does not work like a synapsis in any way... so what was the purpose of the comparison to begin with?
The final sentence is what is basically the information that I was looking for in that article:
Quote:
|
In a phase-change memory, heat is used to shift a glassy material from an amorphous to a crystalline state and back. The switching speed of these systems is slower and requires more power, the H.P. scientists say.
|
It explains how memristors work (basically like a DVD-RW but more dense and not optical)... and it explain the trade-off: they will increase the storage density at the cost of speed.
I wonder why the noise in the rest of the article is needed.