2 ms·
Faster, maybe? I recall seeing a graph towards the end of the planar transistor days (maybe around 22 nm?) that FO4 speed peaked around 45 nm. Overall clockspee
by addaon 5y ago
Faster, maybe? I recall seeing a graph towards the end of the planar transistor days (maybe around 22 nm?) that FO4 speed peaked around 45 nm. Overall clockspeeds continued to improve because of lower power usage and thermal limits, but in isolation a smaller transistor is not necessarily faster; less capacitance and inductance, but also more resistance.
- addaon 5y agoCan't find a source for that recollection, and http://vcl.ece.ucdavis.edu/pubs/2017.02.VLSIintegration.TechScale/VLSI-Scaling-Stillmaker.pdf http://vcl.ece.ucdavis.edu/pubs/2017.02.VLSIintegration.Tech... seems to contradict it, so take with a supersized grain of salt.
- JJMcJ 5y ago> Faster I was thinking since the early days of the transistor, which really began around 1947/48. Or even since the beginning of the microprocessor era. Let's just say it would be hard to make a modern CPU with discrete transistors. Or vacuum tubes. Or relays.