Tomás Chor

> > > Does it have to do with output? Do you know what the time-step is when this happens?I'm wondering if it has to do with using a very...

Ha! It seems @glwagner has cracked it! Top figure is the pressure std and bottom figure is `model.clock.last_Δt`:  I still don't fully understand why this seems to affect the...

> > The difference would be either zero or very small by construction. > > The difference is already very small. It's machine epsilon. > > We need the difference...

> To start working on this I think we need an MWE. Mabye that's easy, just a simulation with constant time-step and output on TimeInterval which should, in theory, work...

> Ok, here's a multi-pronged strategy to address this: > > 1. Rather than computing the next time step directly based on the schedules, compute the nearest _action time_ (either...

@DamienIrving thank you very much the your interest and invitation. Yes I have considered it, and I am already writing a paper on Pymicra (albeit kind of slow). However, I...

> I think for sophisticated research Checkpointing is common, but for simpler classroom and LES applications the checkpointer is used less. After all, probably the most simulations are actually run...

@liuchihl that's awesome. Is there anything relating to `AveragedTimeInterval` that doesn't work after this PR? Also, based on what you said I'm marking this PR as ready for review. (i.e....

@liuchihl I see there are some unaddressed comments, but this PR otherwise looks to be ready, no? Do you need help? Let's not let this nice PR go stale!

> I suggest a **two-step approach**: > ### A. Enable `NetCDFWriter` for tripolar grids and z-star coordinates > > * This step ensures `NetCDFWriter` can handle these grids, **without** writing...