#1795: PITTNullCode results depend on number of processors ------------------------------------+--------------------------------------- Reporter: rhaas | Owner: Type: defect | Status: confirmed Priority: major | Milestone: Component: EinsteinToolkit thorn | Version: development version Resolution: | Keywords: PITTNullCode ------------------------------------+---------------------------------------
Comment (by rhaas):
Christian Reisswig suggested to change the NullInterp options to the default ones like so: {{{ #Interpolation-------------------------- #NullInterp::interpolation_order = 4 #NullInterp::stereo_patch_type = "circle" #NullInterp::deriv_accuracy = 4 }}} which indeed reduces the differences between the 2 process test run and the 1 process test run, but does not make them go away.