Differences in the new solver

[K20shores]

@SancyW pointed out differences in the result of her configuraitons between the micm-powered music box and the camp-powered music box

Below are some screenshots and configurations showing difference. New is on the left, old on the right

C1

configuration: config_0522_C1.zip

C2

configuration: config_0621_C2.zip

C3

configuration: config_0621_C3_perfect.zip

C4

configuration: config_0704_C4.zip

C5

configuration: config_C5_0730.zip

[K20shores]

I can recreate Sancy's differences using version v3.1.0 of the api and v2.4.0 of the client.

The emissions rate are really different. For some reason, the configuration I downloaded from the site in the old version adds an M to the reactants for the emission of CO. The difference in magnitude of the integrated reaction rate between the micm and camp versions is 10^2.

[SancyW]

Thanks! @K20shores Sounds like you've found the key problem! What about the reaction rate? Has it also changed? Is it related to the reaction CO + OH + M? I also added an M here.

[K20shores]

@SancyW, yes, that could be another difference. Adding an M would change the integrated reaction rate for that arrhenius reaction

[SancyW]

@K20shores That makes sense. A butterfly effect. Complicated non-linear impacts.

[K20shores]

The C1 configuration linked didn't work with the newest version of music box. Attached is a zipped folder of configuraitons that all work with the new one, for posterity

new.zip