beibei
Newbie level 2

I use AC and S-parameter simulate the ft of NMOS and PMOS in cadence spectre, two kinds of simulation results, then use ft = gm/(2 * pi * Cgg) verify , found simulation of NMOS and calculation results is same, calculation results of PMOS are much smaller than the simulation results, I can get gm, and Cgg using DC analysis. "ft=gm/2pi*Cgg" is suitable for PMOS?I also found that when PMOS is in saturation,calculation results and simulation results are different;when PMOS is in linear,calculation results and simulation results are same?Why?