On 18/12/2014 07:21, Joshua Joseph wrote:
On Thu, Dec 18, 2014 at 7:23 AM, Willem Ferguson
<[email protected]
<mailto:[email protected]>> wrote:
On 18/12/2014 03:27, Tomaz Canabrava wrote:
Tomaz,
If it's possible at all, we need to complete the CCR UI. Would you
be prepared to help, please? I have been messing around with
private branches of the code that are rapidly ageing but my
understanding of the Qt code is pathetic. We need:
a) Graph of setpoint data when the pO2 is shown on the profile.
b) A graph that displays all the data from the individual oxygen
sensors with respect to the setpoint.
I would so much like to get this part of the development behind us
in order to get on with the other CCR systems.
I think I would be able to help with this.
Extension of the present code to accommodate APD logs will not be
cosmetic, but very easy, now that a basic infrastructure exists:
problem there is 2 parallel dive computers.
The APD development is not likely to have UI consequences, or, if
at all, minor consequences in the manner of selecting APD logs in
the import panel.
Kind regards,
Willem
--
Thanks,
Joshua
*
*
Joshua,
Thank you very much. The steps in UI development I foresee are as follows:
1) Display a graph of the setpoint information on top of the (green)
oxygen partial pressure graph. I attach a profile of a Poseidon dive,
following an experimental and totally incomplete patch that I was
working with. The setpoint graph is in red. Under separate cover I will
send you correspondence about this, including the experimental patch for
what it's worth. IMPORTANT: the setpoint graph should only be shown when
the display of oxygen partial pressure in activated in the toolbar to
the left of the dive profile, and obviously only for CCR dives. For CCR
dives, the oxygen partial pressure is more or less the mean value of all
the oxygen sensors in the system.
2) An oxygen sensor analysis panel. This more complex. I attach a mockup
diagram that I have used in previous correspondence: the use of black as
a colour for the setpoint information here appears not to work well.
This analysis uses the space on the dive profile normally used for gas
partial pressures. It shows the information for each of the individual
oxygen sensors in the system (usually 2 or 3 sensors) with the setpoint
data as a comparison. The idea is that one should be able to check
visually whether any of the individual sensors is drifting upwards or
downwards or whether any one of the sensors fluctuates more than it
should. It is a qualitative assessment and is likely to be complemented
with more statistical information (that will not be shown on the dive
profile, but in the Extra Data tab). I will also forward to you previous
correspondence on this matter.
As an aside and so that you know this ahead of time, there is a catch in
the above. A diver may switch from CCR to open circuit (or bailout)
during the dive. The question is whether one should show the oxygen
sensor data while the diver is on bailout. Probably not. During a dive,
one can determine whether a diver is on CCR or bailout by looking at the
ID of the cylinder that is currently in use, available in the plot_data
structures. This probably does not need any immediate action yet, but it
is useful to keep this in mind when writing UI code that may need some
adaptation later on once when the bailout issues have been finalised in
the underlying code and data storage structures.
3) Specifically for the Poseidon (MkVI and Se7en) dive logs that record
cylinder pressures throughout the dives, the positioning of the cylinder
IDs and the positioning of text labels indicating start- and
end-pressures on the dive profile can be improved. Again I will forward
a hack on this code that may or may not be useful. For CCR, there are
two cylinder pressure graphs (oxygen and diluent). At the start of the
dive, diluent may have a higher cylinder pressure than the oxygen, but
at the end of the dive the dive it may be the other way round with
oxygen having a higher cylinder pressure than diluent. So, the cylinder
IDs and pressures need to be indicated in a way so that the text IDs and
pressures are always clear and do not overlap with each other or with
the graphs themselves. I attach an example of what it should look like.
I hope this gives some indication of what may be required. What do you
think?
Kind regards,
willem
_______________________________________________
subsurface mailing list
[email protected]
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
