For the CMOS, it could roughly separate it into linear, sub-threshold and saturation region. Actually, which region has the highest and lowest gain and linearity?
Actually, could you recommend the book which could explain in detail about this?
Subthreshold has higher gain than saturation but the current is lower so limited linearity. Also biasing for subthreshold can be difficult because of Vth tolerances over the die.
"Subthreshold has higher gain than saturation but the current is lower so limited linearity. "
Is it so? As in my impression, the gain just changes more quickly than in saturation region, but not higher.
the slope of the gain is higher in subthreshold than than of in saturation. By definition, the gain is higher in subthreshold (I think) but one has to remember that signals in subthreshold region are really-2 tinly.
In subthreshold region the relation between Id And Vgs is exponential (like in case of Bipolar) and hence Maximum gain (due to bigger slope). But due to low currents its not feasible to use at subthreshold.