#2155: Fixed bug in calculation of conserved S vector in MHD Prim2Con ------------------------------------+--------------------------------- Reporter: hannah.klion@… | Owner: (none) Type: defect | Status: review Priority: minor | Milestone: Component: EinsteinToolkit thorn | Version: development version Resolution: | Keywords: ------------------------------------+--------------------------------- Changes (by anonymous):

* status: new => review

#2155: Fixed bug in calculation of conserved S vector in MHD Prim2Con ------------------------------------+--------------------------------- Reporter: hannah.klion@… | Owner: (none) Type: defect | Status: review Priority: major | Milestone: Component: EinsteinToolkit thorn | Version: development version Resolution: | Keywords: ------------------------------------+---------------------------------

Comment (by Roland Haas):

The F90 code uses: {{{ dsx = sdet * ((drho*(1+deps)+dpress+b2)*w*w * vlowx - & ab0*blowx) }}} and the papers (https://arxiv.org/abs/1304.5544 eq. 11) give: {{{ S_j = sqrt(detg) (ρh W v_j − α b_0 b_j) }}}

I have accepted the pull request and backported into the release code.

We should send out an announcement since this affected science runs using default Einstein Toolkit parameters since commit fe5578d1d28d813cee37e35df69572ddc2f0ef71 "GRHydro: enable C++ code by default" from Mon Feb 23 00:01:14 2015 (and all code using the cxx flavor of GRHydro's MHD code since its inception).