Carlos Castillo Passi

Hi sorry for the delay, I was at ISMRM in Singapore. Can you give a little more detail to reproduce the problem? In theory, as @gsahonero says, the FOV is...

As you said, there are some preprocessing steps before the simularion that are not accounted for in the reported time. As `run_sime_time_iter` is the actual simulation I think is fair...

We could just remove that line, maybe it doesn't help. If you do a `for` loop of `simulate`'s, and there is no memory problems without without that line, I would...

@gsahonero can you report if this did not generate any problems? > If you do a for loop of simulate's, and there is no memory problems without without that line,...

Where would you put it in the docs? in the docstring for `simulate`?

Sounds like a good first PR :smile: (You can do it directly on GitHub by pressing the :pencil2: icon) https://github.com/JuliaHealth/KomaMRI.jl/blob/master/KomaMRICore/src/simulation/SimulatorCore.jl

Hi @Tooine, first of all, thanks for using Koma :smile:! That definitely sounds like a problem. Thanks for attaching the sequence so we can investigate. Are you using the newest...

I forgot to answer this: > Maybe, there might be a way to modify the implementation to separate the precompilation from the simulation, in which case the sequence preprocessing could...

Thanks @curtcorum for all the effort!! This is similar to what I would have tried. I will try to take a further look asap.

Yeah I agree that the GC line is not worth checking. I will profile your sequence during the weekend. (Btw for this I recommend to use @profview in VSCode instead...