Closed cnrrobertson closed 2 years ago
Indeed, at first glance seems like https://github.com/FourierFlows/FourierFlows.jl/blob/c6c041c60ff02854d76c91880224c039b9e8e216/src/timesteppers.jl#L374 is redundant.
Open a PR -- that'd be great!
Great. See #315
Thanks for an excellent package. I've really enjoyed using it and appreciate all the work you've put into it.
I noticed while comparing the ETDRK4 implementation to my own that there is an extra substep here.
Specifically, it seems that only one of
ETDRK4substep12!(ts.sol₁, ts.exp½Ldt, sol, ts.ζ, ts.N₁)
or@. ts.sol₁ = ts.exp½Ldt * sol + ts.ζ * ts.N₁
is needed. Removing either passes the tests.Happy to submit a PR if its helpful.