spice up your life with WINSpice

You can download WINSpice from this link: http://www.ousetech.co.uk/winspice2/
Here's the code for the inverter circuit.

* Title : CIRCUIT2
Vin 1 0 PULSE(0 5 0PS .1PS .1PS 20PS 40PS)
Vcc 3 0 5V
M1 2 1 3 3 modp W=2U L=250N AD=100N AS=100N
M2 2 1 0 0 modn W=1U L=500N AD=100N AS=100N
.model modp pmos(LEVEL=2)
.model modn nmos(LEVEL=2)
.OPTIONS DEFL=250N
*.DC Vin 0 5 .1
.TRAN 1PS .05NS
.control
.endc
.end

I wonder if my conclusion that changing the width and length of the transistor has no effect as long as their ratios remain the same is correct. I also put in the report that when the width of the PMOS is increased, the delay from input fall to output rise increases. And when the width of the NMOS is increase, the delay from input rise to output fall now increases. The length on the other hand changes the time it takes for the output to change state. It's a bit difficult to explain in words so here's an exaggerated waveform (with the length of the PMOS very much greater than the NMOS):


For the case when the length of the NMOS is larger, the curve will be seen on output fall.