Spencer Bryngelson comments

Results 234 comments of


                                            Spencer Bryngelson

Operator splitting, adaptive time stepping, and other fixes for mixing layer and bubbles

Note: I found this https://github.com/MFlowCode/MFC/issues/390

Delta GPUs results give NaNs in the viscous sub-grid bubble benchmark case whereas Phoenix GPUs do not

Update: This issue is associated with `parallel_IO = 'T'`. I witnessed it again on a Rogues Gallery GH200 chip w/ NVHPC 24.1.

Delta GPUs results give NaNs in the viscous sub-grid bubble benchmark case whereas Phoenix GPUs do not

I'm not sure if this is still "broken" or not.

Dead code

Possibly. Can you go back in the git history (perhaps looking at `MFC-develop` repo as well) and see _when_ they were used? I assume, at some point, they were.

Dead code

Note on this issue: VSCode seems to be pretty good at finding variables that are declared multiple times (once globally and again locally) and finding _unused_ variables. You will need...

Dead code

@anandrdbz remove `_flat` names in variables.

Equation of state is not DRY

I think `m_variable_conversion` is a more natural home for this kind of thing.

@anshgupta1234 Check out these lines in `m_cbc`: https://github.com/MFlowCode/MFC/blob/d80602abfc6c1528708963032ef8cd44b3f26622/src/simulation/m_cbc.fpp#L786-L853 I think these can probably be factored out completely. One thing we didn't initially notice is how pervasive the computation of `c`,...

Equation of state is not DRY

I realize this is becoming an omnibus "issue". Let's merge your pull request when you have all of the main variables conversion stuff done, then do speed of sound `c`...