Aiming to become the global leader in chip-scale photonic solutions by deploying Optical Interposer technology to enable the seamless integration of electronics and photonics for a broad range of vertical market applications

Free
Message: speed/node size

See if my mathematics is correct:

Assume the speed of a silicon CMOS processor at 40 nM is x,

then the speed of a POET processor is 5x – that is it is 5 times faster.

Using a compounding formula for a 20% increase of speed for each node, then over 4 nodes and compounding the result:

Silicon = (x + 20% of x)4 = (6x/5)4 = 2.07x. This means you double the speed over a 4 node reduction in size.

With POET the initial speed at 40 nM is 5x and using the compounding formula for a 20% increase you get:

POET = (5x + 20%of 5x)4 = (6x)4 = 1296x.

According to me, if you reduce a Silicon processor through 4 nodes, you double its speed and if you do the same to a POET processor you get 1296 times faster.

That means when the POET processor’s nodal size is reduced through 4 nodes it is more than 600 times faster than a Silicon chip. Now that is impressive. Add to that the superior clock speeds, the superior transistor performance, light transfer, and the substantially reduced power consumption. Remember a 40nM POET processor aims to have the same speed as the best CMOS FINFET processors on the market now.

Disruptive, I suggest, is an understatement. Don’t worry about the share price today. More than happy to be corrected.

David

Share
New Message
Please login to post a reply