Comment (by Roland Haas):

I ran the parfile (making sure to ask for 4th order finite differencing for the hydro source terms) using ET_Proca + Meudon_Bin_NS version eec2b89a "Exact: fix integer dvision for Goedel spacetime" of einsteininitialdata. This produces results (pre-merger) that are indistinguishable from results produced using the code originally run on Zwicky. There is a difference post-merger but given the huge difference running with or without -Ofast made I am not concerned about that. Both runs used the same compilation options (using -O2). I will run one more run using the master version of Meudon_Bin_NS to check what differences become visible. Commit 1b9d6a3 "Meudon_Bin_NS: using EOS_Omni" of einsteininitialdata is known to change the ID on the grid a bit (see https://bitbucket.org/einsteintoolkit/tickets/issues/2119/the-binary-neutron-star-gallery-example#comment-50396585) but mostly in the post-merger region.

‌

--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2119/the-binary-neutron-star-gallery-example