If it’s sub 10nm, it’ll be ‘good enough’ for common electronics. Cash registers, parking meters, iot and little sensors. Not all chips in cars have to be the latest tech; you don’t need the climate control or ECU to be on the bleeding edge, but it still takes up wafer & space.
Ehh, the 100nm process is over 20 years old at this point, like older than the core2duo processors. Even ignoring power efficiency, the physical size of the chip and the amount of silicon it takes would be concerning.
Since die area is squared, you can make 100 times more processors using a 10nm fab with the same amount of silicon instead of the 100nm process
If it’s sub 10nm, it’ll be ‘good enough’ for common electronics. Cash registers, parking meters, iot and little sensors. Not all chips in cars have to be the latest tech; you don’t need the climate control or ECU to be on the bleeding edge, but it still takes up wafer & space.
No, 100nm is good enough for that.
Ehh, the 100nm process is over 20 years old at this point, like older than the core2duo processors. Even ignoring power efficiency, the physical size of the chip and the amount of silicon it takes would be concerning.
Since die area is squared, you can make 100 times more processors using a 10nm fab with the same amount of silicon instead of the 100nm process
I mean, microcontrollers, stand now, are commonly done in 100 or 80 nm.
Smaller is more expensive and less robust too.