Gear star formation

No need to look at it, I will assign it to you Matthieu when it will be ready.

TODO:

• Cleanup comments
• implement total mass stars

Just one quick thing: You cannot pass the xp to the density_iact funtion. It's absolutely forbidden.

added 5 commits

• 8e2bbca3 - GEAR:Remove xp in density
• 96e80fcf - GEAR: more cleaning
• 7c47e3b6 - Format
• a8ad0270 - GEAR: star formation fully implemented
• 99c997f7 - GEAR: SFR working correctly

Compare with previous version

(Very low priority)

I have done a few modifications of the API, can you check it and tell me if you are ok with it? I will do the EAGLE / None modifications afterwards.

The main one is for the star_formation_logger, I have implemented a way to keep in memory the information (e.g. total mass of stars created).

The star formation is linked with the pressure floor, therefore I am planning to remove the star_formation structure from the part once both are implemented.

Can you explain the rational of changing the memory logger in that way? I am not against it but just wondering what information was missing before. You could get the total stellar mass formed just by summing the mass column.

I am not missing any information, but it was not possible to integrate a quantity through time inside SWIFT.

As we are simulating a single dwarf galaxy at a time, it is nice to be able to do a quick tail of the file and obtain the total stellar mass without reading the whole file and without any script.

If you are not happy with it, it is not important at all to have this feature. I just think that we should provide it if we provide a logger.

added 2 commits

• eb32472f - GEAR: add star_formation_history_accumulator
• 577dbe6d - GEAR: update comment

Compare with previous version

I have implemented the new structure and checked the code with MPI. If you agree with this code, I will update EAGLE and none before merging.

added 87 commits

• 577dbe6d...e63b81a3 - 77 commits from branch master
• f4900da4 - Apply patch from github merge request #13 (closed)
• 2ace1f38 - Add GEAR stars
• e6ed5f8a - format
• 68380482 - GEAR:Remove xp in density
• d2f8c943 - GEAR: more cleaning
• 93ee46fb - Format
• 21036419 - GEAR: star formation fully implemented
• 3af19353 - GEAR: SFR working correctly
• b837cbc9 - GEAR: add star_formation_history_accumulator
• 1076fe31 - GEAR: update comment

Compare with previous version

unmarked as a Work In Progress

marked as a Work In Progress

added 1 commit

• 48feaee3 - GEAR: add parameters in parameter_example.yml

Compare with previous version

I am generally fine with the GEAR SF logger storing the total mass formed. It does not impede on the other schemes. However, I think you should be careful not to fool yourself. Stars lose mass over the course of their lifetime so the number you read from the logger will never be a good thing to look at. Snapshots (since they include the mass loss) will give you a better information.

Thanks for the warning. We are never using the logger to compute a proper star formation rate, we always use snapshots.

The logger will only be used in order to quickly see the progress of the simulation.