Sebastian Micluța-Câmpeanu

I see... Do you think that the second part would better live in a separate package? In the case of GitHub & GitLab there are julia interfaces which would cover...

I've reimplemented progress logging without doing any stdout redirection. The implementation is now simpler and it should be more robust. I've also added progress for time based limits. The only...

Hmm, it could be due to the stepsize calibration for the symplectic solvers. I'll take a look and double check that.

I've re-ran the timestep calibration code https://github.com/SciML/SciMLBenchmarks.jl/blob/a7df7e7c0fe27c2915769312a8cbd27608c90043/benchmarks/NBodySimulator/liquid_argon.jmd#L123-L129 and I get ```julia 9-element Vector{Float64}: 1.0 2.9735980871351813 3.0005619459841046 3.0164920767484356 4.985990322613598 6.9410438162718835 8.011579310978117 18.053161246947383 35.93506077349945 ``` as opposed to the hardcoded ```julia c_symplectic...

In the past ~6 hours I've been running `run_benchmark!(results, t, symplectic_integrators, τs[1], c=c_symplectic)`, i.e. just compute the results for the first timesteps and it looks like the runtimes are larger....

It is possible to be over an hour, but I'm not sure, I'll have to run them and see. Should I try to make it shorter? What should be the...

Oh... That's unexpected.

Are there any logs available to see where it hangs / what happens?

I ran them locally and it looks like it's not getting stuck, but that the benchmarks take very long  I'll try to have a look at some concrete examples...

Yeah, there is was issue with interpolating high order solvers because the interpolating polynomial was 3rd order and would give bad energy conservation. So can't use saveats and we have...