I've a question about sky130, magic and parasitic extraction/simulation. I was curious about the differences between the metal layers, wanted to look at capacitance (lower layers being closer to substrate), cost of vias vs that cap and the track's internal resistance, stuff like that. So I created a set of tracks that all go to identical resistors: an input, some trackage, a resistor, an output port, and a few VSS. Before trying to compare the effects between tracks, I looked at the effect of having the equivalent of a ground plane under everything or not (vss is connected to ports sunk in locali on both ends, with "gnd pour" or not). The top graph is my control, just simulated sky parts, the other three use parax spice extracted via magic Then I just clocked it with fast rise/fall and the graphs show what comes out (red trace is my input clock). The first set does show some rounding, but with the GND plane I'm getting over/undershoot (but no signs of ringing). So my questions are: • is it sensible to try and look at this with simulation? (i.e. is the model complete enough to mirror reality?) • would sim be able to show the cost of switching layers (vias) and difference in stray cap for tracks (does it even account for the substrate)? • what explains the waveform of the bottom two cases--is this ground bounce because of the high R locali, something else? Thanks!