Roland, thank you very much for the information.
My student (Beatrice Giudici, in cc to this email) told me that this is the same WENO-Z that is used by Bernuzzi in his BAM code ( https://ui.adsabs.harvard.edu/abs/2016PhRvD..94f4062B/abstract). She is going to do some tests, starting with the Balsara tests and compare it with MP5 and WENO. We will then probably move to oscillating TOV and then BNS (maybe doing some of the tests that Roberto did in his paper with WENO).
We will keep you posted. I just wanted to check what the status was. I added Roberto in this discussion because of his work with WENO and GRHydro, so I was wondering if he had a look also at WENO-Z.
Thanks, Bruno
Il giorno gio 16 gen 2020 alle ore 19:30 Haas, Roland rhaas@illinois.edu ha scritto:
Hello Bruno, Roberto,
I just discovered that GRHydro supports also WENO-Z (a different implementation of the WENO 5th order reconstruction scheme).
I have a master student currently looking at high order schemes and I
admit
that I thought that GRHydro had only WENO5 implemented (recon_method="weno") since this is the scheme that was mentioned and tested in the GRHydro paper (which instead does not mention WENO-Z which
I
assume was probably implemented after the paper).
Has the WENO-Z implementation also been tested? Is someone using it?
It has been years since this was really tested (it was first implemented in December 2012 / January 2013). I believe Christian Reisswig and I have to take blame for implementing any bugs in it.
I am no longer quite sure if how much testing was done. I looked through old emails but cannot tell from them whether only WENO (plus changes from Sasha Tchekhovskoy's WHAM paper: https://arxiv.org/abs/0704.2608) which were described in the GRHydro MHD paper (Moesta, Mundim et al.) were tested or also the WENO-z implementation.
Worse, I do not even have a reference from the flavor of WENO-z that was implemented :-(. Could be https://www.sciencedirect.com/science/article/pii/S0021999107005232 Borges et al. JCompPhys Volume 227, Issue 6, 1 March 2008, Pages 3191-3211 or something completely different.
So, while I believe that there was effort to test it, there is no longer a reproducible way to check what was tested and how well it performed.
Yours, Roland
-- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://pgp.mit.edu .