momchil-flex

Yep sounds great!

Looks good to me. Maybe for those who may still wonder why this choice: imagine that there's an axis in the direction of propagation and the beam center is at...

I guess this could be purely frontend then? Actually it can even be done as an example script rather than integrating in our code? Like in your optimization loop, after...

Regarding caching, note that this is unfortunately currently turned off for all autograd tasks because of #1906

> It's only turned off for server-side autograd task though right? which isn't default yet. True. > Regarding https://github.com/flexcompute/tidy3d/issues/1906, maybe one easy approach is just to store the post normalization...

> @momchil-flex I want to make sure this doesn't fall through the cracks for 2.8 since probably it will be highly requested.. do you think you'll find time to work...

This issue is about showing the maximum compute time remaining, i.e. once the time stepping starts we should be able to predict how long it's going to take if it...

Off the top of my head this is probably something to do with 2D simulations still being assigned a non-zero `dl` along the 0-dim, while PMLs are just plotted with...

The time-domain is basically the field profile at the central-frequency times the time dependence of the `source_time`. So the field profile is only correct at the central frequency, but it...

Hi @lucas-flexcompute this this fall through the cracks? I see there's some dangling commit. A user just asked about it.