Hi Folks:

Let us say I have a 2d case that takes x sec/cell/iteration to execute, based on bench marked tests. Is it a reasonable assumption that a 3d case will take the same amount of time ? Thanks for your answers.

Sam

Your 3D case presumably will have one additional equation to solve (W-momentum), right? If so then no, the 3D case will be more expensive per cell per iteration.

Your parameter is the correct one to look at. By the way, that was defined as the "grind time" by the T-3 group at Los Alamos in the late 60's. [At least, that's the only place I ever heard it used.]

Another factor to consider is that (for a transient problem at least), the stability limit on the time step will probably shrink as the third dimension is added.