Continuation of incomplete runs

[RobertRidgway]

I've been running into times where the time allocated to completing the computation is insufficient and the run is aborted. Adding the ability for Exocross to dump to a file during computation and allowing Exocross to continue from an incomplete file would be quite beneficial to have.

[Trovemaster]

It is a good idea, I will think about implementing the ability to dump and restart.
BTW, you tell me what exactly you are trying to compute I might try to advice on a strategy to speed the calculations up. For example, you can try to generate a temperature dependent set of super-lines and then use them to produce temperature/pressure dependent cross sections. These two steps together are significantly faster than a direct line bu line generation of cross sections. See, e.g. example for POKAZATEL at https://github.com/Trovemaster/exocross/wiki/Task-3:-New-water-line-list-POKAZATEL-and-super-lines.

Cheers,
Sergey

[RobertRidgway]

Hi,

I'm computing opacities for a range of pressures and temperatures using the updated NH3 line-list ExoMol put out recently.

[Trovemaster]

Hi Robert,

In case you would like to try using the super lines to generate the CoYuTe cross sections, I have uploaded the NH3 super lines for a set of temperatures T = 100,200,...,2000 K to exomol, http://exomol.com/data/molecules/NH3/14N-1H3/CoYuTe/
See also examples of ExoCross input files to use or generate super lines on the same exomol page.

I hope this can help.

Cheers,
Sergey

[RobertRidgway]

Hi,

I'm looking at the manuals documentation on Super-lines. I was planning on using the .broad files from the ByTe line-list. How much of a change in the calculated opacities would you expect would not being able to use the state-dependent broadening parameters?

[Trovemaster]

This depends on the application and something you would like to test. For opacities we don't see significant changes.