first cut of dive planning

[Dirk Hohndel]

Jan Schubert <Jan.Schubert at GMX.li> writes:

> On 01/05/13 18:23, Dirk Hohndel wrote:
>>> I just played around and got some funny looking profiles (saw tooth is a
>>> good description for some of them :-). But I also managed to get a idea
>>> what is working and that looks promising.
>> I wonder what you did. Can you post? I want to make sure the algorithms
>> are sound...
>
> See screenshot attached, this seems still to use the definitions from
> planner.c and not the ones I entered in the UI (which seems to be gone
> also afterwards).

Did you read the part where I said that the UI isn't hooked up to
anything and ignores the input? Yeah, I meant that.

And yes, this looks terrifyingly bad. Even worse than what I was able to
produce. Negative ceilings are strange.

> I tried to play around with different segments (time and depth) in
> planner.c but quite often ended up in such saw tooth profiles. The
> screenshot is from just defining a single segment:
> plan_add_segment(&diveplan, 70000, 136000, 120, 700);

Ok, I'm heading out and won't get to it till later this evening, but
I'll use that to try and hunt down the bug...

/D