I'm putting together a jupyter notebook to assist with gm/id sizing. Someone want to take a look and provide feedback?

@User this looks good to me, I love a Jupyter notebook. Something I want to look at after submission is making a Python class that can be used to query pre-characterised gm/Id data (and EKV inversion factor). A slow but one-off simulation to grab all operating point data over temperature, voltages and corners including noise parameters - and offset ideally. Once that data is stored as file a Python class can easily query it. Then a nice interactive GUI should be possible for people to easily play around with bias conditions and see how it affects the performance of devices. The other thing the object would allow is generative design like BAG does. The design equations for lots of popular circuits are well known so by querying the characterised devices automated sizing should be possible. And then optimization in a reasonable time becomes much more feasible when an iteration only involves querying some Python data structure rather than booting up NGSpice, running the sim, extracting the result etc. This is merely me restating (part of) the BAG philosophy but I would be interested in seeing it in a more portable, self contained Python class which people can easily understand and contribute to. Maybe this is something we should collaborate on?

Sounds good. I'd planned on something more like that after the first shuttle, I just put this together quickly now. Thanks for looking at it

Nice one, we can discuss it further in a few weeks