It's on my todo list to look into this. But at the moment I'm a bit overwhelmed
with bug reports, so it's going to take a bit longer.
Most likely the problem is that a tank switch sample also includes a pressure
value. That's where we get that initial pressure from. I suspect there might be
some bits to indicate there is no sensor attached, and thus no (valid) pressure
value.
Jef
On 11-11-16 01:35, Adric Norris wrote:
For what it's worth, the visible issue *appears* to be caused by a bogus
initial pressure reading from the first sample after the air source was
switched. Applying the following diff to the XML and re-importing the dive
seems to get everything looking back to normal.
$ diff -u test.xml test-fixed.xml
--- test.xml 2016-11-06 08:07:47.592116207 -0600
+++ test-fixed.xml 2016-11-10 18:26:17.323748662 -0600
@@ -182,7 +182,7 @@
<sample time='41:30 min' depth='10.954 m' pressure='48.953 bar' />
<sample time='41:45 min' depth='10.649 m' />
<sample time='42:00 min' depth='10.516 m' pressure='47.919 bar' />
- <sample time='42:15 min' depth='10.287 m' pressure='336.464 bar'
sensor='1' />
+ <sample time='42:15 min' depth='10.287 m' sensor='1' />
<sample time='42:30 min' depth='9.754 m' />
<sample time='42:45 min' depth='10.363 m' />
<sample time='43:00 min' depth='10.516 m' />
As mentioned previously, there was no transmitter present on the pony
regulator, so I don't believe it should have included the pressure
attribute in any event. No idea how it managed to get there in the first
case.
On Sun, Nov 6, 2016 at 9:09 AM, Adric Norris <[email protected]>
wrote:
I've encountered a bit of an oddity with reported air pressure on my most
recent dive, which I believe can only be explained by a bug of some sort...
no idea if it resides in my Oceanic VT4.1 dive computer, Subsurface, or
libdivecomputer. Specifically, the graph displays an impossible ending
pressure. I've confirmed that my computer has the current firmware
<http://www.oceanicworldwide.com/us/support/software/> (2B03) loaded, for
what that's worth, and I'm using the Linux appimage of Subsurface
4.5.95-25-g02389c95cb59.
My dive computer supports four air sources, although prior to the dive in
question I had never made use of more than one. On this occasion, however,
I was practicing deployment and use of a 19cuft pony bottle near the end of
the dive... since I was expecting to use it for around 10 minutes, I
decided to switch my computer to the second air source as well. In
hindsight this wasn't really necessary, as the pony regulator doesn't
currently have a transmitter, but it seemed reasonable at the time. Both
tanks were filled with ordinary air, so there was no difference in the air
mix configured for each source on my dive computer, and had a starting
pressure of approximately 3000 PSI (200 bar). The main tank (the one with
the transmitter) ended with about 700 PSI (48 bar).
Air usage on the graph looks normal up until the time the air source was
switched, at which point it displays a pressure of 4880 PSI (336 bar) for
the remainder of the dive. Although it makes sense that it would show a
constant pressure (due to the lack of a transmitter), I don't understand
how it could possibly come up with the displayed value (especially knowing
the ending value of the main tank).
I've made the following files available at https://drive.google.com/open?
id=0B-3AX-51D0csZ3I1dFNkR2FUYmM:
- *subsurface-libdivecomputer-dump.bin* - libdivecomputer dumpfile
covering around 70 dives (since I had to force a re-download), but the dive
in question is the very last item.
- *subsurface-libdivecomputer-dump.log* - libdivecomputer logfile
covering the same "import" where the dumpfile was captured.
- *subsurface.log* - libdivecomputer logfile, covering import of only
the dive in question (although I still had to force a re-download).
- *subsurface-terminal.out* - Output produced by running Subsurface
from the terminal.
- *text.xml* - Subsurface XML file containing just the dive in
question. Please note that it was saved immediately after import, with no
manual changes having been made.
On my real logbook, saved to cloud storage, I did manually add a second
tank with the relevant capacity and start/end pressures. This didn't make
any apparent difference to the graph, however.
At this point I have *not* tried to reproduce the issue under the latest
official release (4.5.6). I'll be happy to do so if you like, however, so
please let me know.
Thanx!
--
"In the beginning the Universe was created. This has made a lot of people
very angry and been widely regarded as a bad move." -Douglas Adams
_______________________________________________
subsurface mailing list
[email protected]
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
_______________________________________________
subsurface mailing list
[email protected]
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
