[mesa-users] mesa sermon: mesh_delta_coeffs and varcontrol_target are the 1st things to vary

[Bill Paxton]

Hi,

My mesa sermon of the week is to remind people to check the effects of varying 
the number of grid points and the size of timesteps in addition to varying the physics.
Even though mesa/star produces an answer, it doesn't mean that it won't happily produce
a quite different answer for a different resolution in time/space.  So if you find something
that looks a bit strange (or a lot strange), check how it responds to a factor of 2 or so
increase in number of grid points and number of steps before making up a story
to explain it physically (or, heaven forbid, deciding it is a bug in mesa).

mesh_delta_coeffs and varcontrol_target are the 1st things to vary.
If the weirdness is still there at higher resolution, then, and only then, consider
varying the physics to look for an explanation.   Don't decide you've found a great
new thing (or even a great new bug in mesa) until you've checked to see if the result
is still present when you look more closely by using more cells and more timesteps.

Thanks,
Bill