[FIXED PATCH 3/2] gas model: add polynomials for Z factors of oxygen/nitrogen/helium

[Linus Torvalds]

On Wed, Mar 2, 2016 at 5:43 PM, Linus Torvalds
<torvalds at linux-foundation.org> wrote:
>
>
> I worry a *bit* about the fact that the least-square coefficients were
> built from the tables that only go up to 4000 psi (275 bar), and I don't
> know how well the quintic approximation behaves above that.

Ok, I just made Wolfram alpha plot the Oxygen curve, and yeah, it
looks reasonable to 300 bar. It starts looking worse at 400, and by
500 it's going down towards 1.0 again.

I think it's fine as-is for now. As long as we're doing ok in the
1-300 bar range, we're fine. I think 300 bar is very unusual even in
Europe, just because filling at those pressures is a pain.

Having a curve that goes all the way to 500 bar (like the air one we
have - that polynomial looks good all the way) would be nicer, of
course. I think that would be a "further tweak" and the patches are ok
as-is for now.

                Linus