[PATCH] Re: Heat map and deco model name

Rick Walsh rickmwalsh at gmail.com
Tue Sep 13 02:50:07 PDT 2016


On 13 Sep 2016 19:23, "Anton Lundin" <glance at acc.umu.se> wrote:
>
> On 13 September, 2016 - Robert Helling wrote:
>
> > Hi,
> >
> > > On 13.09.2016, at 10:56, Anton Lundin <glance at acc.umu.se> wrote:
> > >
> > > Fantastic. I really like this.
> > >
> > > Would be fun to hear what Simon has to say about this. Anyone who
would
> > > like to dig up a email adress and drop him a note?
> >
> > indeed, this color scaling is much much better (on vacation, I could
not access the video).
> >
> > Dirk, please apply all four patches (two from me and two from Rick)
plus this one which slightly cleans up the code.
> >
> >
> >
> > I was also thinking about showing this to Simon Mitchel (once this is
in a daily build).
> >
>
> I tested this with the VPM code in the planner, and something doesn't
> look right there. There is a lott more on gasing with the same profile.
>
> I compared 20/85 vs. VPM-B +3 on a 18/55 / 50% / 100%, 20min @ 70m dive.
>
> With this settings the both algorithms generate about the same profile,
> but the heatmap for the VPM dive looks way "better" with way less
> super-saturation. Something smells fishy here.
>
>
Yes, I noticed the same.  It's completely wrong for vpm.  To get it to
work, we need to calculate Buhlmann "percentage" values, which isn't done
for vpm at the moment.  Adding it shouldn't be too difficult.

For interest's sake, you can create a vpm plan, save it, go back to the
planner to set the deco mode back to Buhlmann, then close the planner and
view the saved vpm dive....

Another oddity I noticed were the plans used in the navy study in the
presentation.  Both the non-deep stop and deep stop plans were for 30min at
170' (52m), for a total dive time of ~210min all on air. First stop on the
non-deep plan was at 40' (12m). To approximate that plan, I needed to set
gf to something like 150/60.  Very wierd plan.

Rick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20160913/0587f1b0/attachment.html>


More information about the subsurface mailing list