Hi Tim,

Your suggestion makes sense, I think. Certainly a possibility :-)

It could also explain why different compiles do different things ...

Thanks for the thoughts,

Mike Andrews.

"Tim Attwood" <tim### [at] comcastnet> wrote:
> Since the box is very thin in the z direction,
> scaling it by a very small number in the z is
> creating a near-zero, which passes the scale
> by zero check, but still is generating a divide
> by zero elsewhere inside the code, probably,
> I think, well maybe.  scale <Scl,Scl,1> works
> fine though.

Post a reply to this message