Dive planner VPM-B problems

[Stefan Fuchs]

Hello Willem,

please find a few comments from my side on top of your two mails from
today about this topic.

Am 16.10.2017 um 15:31 schrieb Willem Ferguson:
> Attached two images of a mixed gas dive.
>
> Plan1 is a first step of a multigas dive plan and the planner places
> some points above the ceiling. There are three issues to note:
>
> 1) The planner places the last number of points during which the
> violation occurs. If I change the conservatism to either 2 or to 4,
> then the violation disappears.
>
> 2) If I change the mix of the intermediate gas to 27/27, the violation
> disappears.
>
The feeling I had all the time during the last weeks is that the deco
ceiling calculation for the ceiling displayed in the profile (inside and
outside the planner) still has some bugs.
What is interesting is that it seems that the waypoints generated by the
planner are correct. Minimum two reasons why I say this:
- Test did always pass.
- Your example from your second mail comparing current master with
4.6.3.xx and also my tests show that planner generated waypoints didn't
change

So for the moment I guess you and me provided more than enough examples
for Rick and Robert and the only thing we can do now is hope that they
find some time to search for the bug(s).
At the same moment I wouldn't see the situation too negative because as
I'm saying the planner results seem to be correct.

>
> Two suggestions, not at stake at all for V4.7 (see Plan2):
>
> 1) I would like to insert a top row in the dive points table to
> accommodate a travel gas (27/27 in this case) and at this stage I
> cannot. It would be really helpful if there was a way to do this. As
> it is at present, I would have to delete the whole dive points table
> and start from scratch.
>
Isn't this already possible? Insert a new waypoint at the end and then
set the value for runtime to a value lower than runtime of your first
table row. This will move the table line to the first position. And then
you can enter your travel gas.

> 2) The minimum gas need is meaningless in this situation and I suggest
> that this calculation is only done for single cylinder dives.
>
Yes, you are right. This is what is also written in the documentation :-)
The minimum gas results are useless for some more complex cylinder
setups like your example where you have a travel gas for ascend.
But limiting the minimum gas calculation for single cylinder dives is
not the correct approach. There are many multi cylinder setups where
todays minimum gas calculation works perfectly.
For example this - for me rather complex one - works and is also what I
need for myself:
Descend with 1st deco gas 50/15
Switch to 80cuft bottom stage 18/45 at ~15m
Dive until bottom stage empty at e.g. 60m
Switch to back gas 18/45 24l and continue at 60m
Then start ascend and do deco with 50/15 and Oxy
Minimum gas will be calculated perfectly for the last bottom datapoint
at 60m when you are breathing the 18/45 from the back gas.

Could you provide me examples (xml and setups) of dives where you think
todays minimum gas implementation is useless?
I would like to check if I can improve something or maybe even force the
calculation off for such setups.

BTW: You could be one of the best candidates for testing my latest
changes around cylinder handling in the planner. It would be great to
hear from your side if this improves or breaks s.th. for you. Robert
merged the changes already but the PR was this one:
https://github.com/Subsurface-divelog/subsurface/pull/663

Best regards
Stefan

-- 

Stefan Fuchs
E-Mail: sfuchs at gmx.de <mailto:sfuchs at gmx.de>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20171016/eac0599d/attachment.html>