Problem with VPM-B in master

[Rick Walsh]

On 21 August 2015 at 00:30, Dirk Hohndel <dirk at hohndel.org> wrote:

> On Thu, Aug 20, 2015 at 09:31:30PM +1000, Rick Walsh wrote:
> > Hi,
> >
> > VPM-B is producing different plans in the latest master.  I have tracked
> it
> > down to commit 72806e42bc23ab7c7f2a639a1efee017388b53bc Replace global
> > in_planner variable by helper function.  I don't believe this commit was
> > supposed to change the resulting plans.
>
> Well, the patch was incomplete and Robert sent the second half yesterday,
> but I didn't get around to apply that - I just did.
>
> Can you please retest with the latest master that I just pushed?
>

Thanks Dirk and Robert, it worked much better in my three minutes of
testing as I ate breakfast this morning.

However, I did get an error message in the red box at the base of the
screen when testing the 100 m / 10 min dive with 18/45 bottom gas.  Along
the lines of the cubic equation has a complex solution.  Sorry, I had no
time to track down the cause.  If I were a betting man, I might put my
money on it being related to there being 45% He and 37% N2, i.e. fHe >
fN2.  If that's the case, we might need to pretend fN2 - fHe = 0 in the
equation determining the exponential decay parameter, or otherwise put a
lower bound on the terms in the cubic equation.

I think this is leading us to ignore the CVA iterations that produce deeper
first stops (from memory, our comparison profiles from other software has
the first stop at ~54m compared  to our 42m), and come up with a less than
ideal profile.

Cheers,

Rick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20150821/439037ba/attachment.html>