Date and time wrong on Uwatec Smartcom Import

[Stuart AlucardDr]

I am trying to get off the antiquated software that Uwatec provided with my Smartcom and really want to use Subsurface.  I converted my existing log without a problem - it looks great!

Then I tried to actually upload some new dives via the Infrared into SubSurface.  After the usual trials and tribulations with the interface I finally got them to import.  I did the dives in July 2021 but when I import the dives it says October 2020!

So I went back to the old software and brought those dives in and they imported with the correct date, so I concluded that it's not an issue with my dive computer.

Can anyone give me tips as to how I can import these with the correct date and time?

Thanks in advance

[Jef Driesen]

If you send a memory dump of your dive computer, I'll have a look. Simply enable
both the "Save libdivecomputer logfile" and "Save libdivecomputer dumpfile"
checkboxes in the subsurface download dialog. When finished, no dives will be
imported. Just send those two files.

In the past, the date/time parsing relied on the clock of the host system, but
this caused some issues when changing the clock on the dive computer. Therefore
we now use the fixed reference time (2000-01-01 00:00:00 UTC). This might result
is some small differences, but the time certainly shouldn't be off by a year.

Jef

[Guido Mennen]

I have the same issue with my Uwatec Smart Z,
Only a little bit worse ...
It imports dives correctly, but dates them in august 2010

The smartTRACK software imports at the correct date

I'll also send you my logfiles, hope there is something to learn from them.

Thanks,

Guido

[Guido Mennen]

And the logfiles ...

[Jef Driesen]

On 28/10/2021 13:37, Guido Mennen wrote:
> I have the same issue with my Uwatec Smart Z,
> Only a little bit worse ...
> It imports dives correctly, but dates them in august 2010
>
> The smartTRACK software imports at the correct date
> I'll also send you my logfiles, hope there is something to learn from them.

In the logfile you have:

Event: systime=1635420555, devtime=673467714

That's respectively the timestamp at the time of the download on the host system
(systime) and the dive computer (devtime). That corresponds to a reference time
(epoch = systime - devtime / 2) of:

2011-02-26T02:18:18 UTC (1298686698)

The expected epoch value for uwatec/scubapro dive computers (and the value used
in our parser) is:

2000-01-01 00:00:00 UTC (946684800)

Since there is such a huge difference of about 11 years (352001898 seconds), all
your date/time values are also completely off with the same amount!

I have no idea why the difference is so large. Some small differences are
expected due to clock drifts, etc. Typically these differences are at most a few
minutes. There are also many cases in my dataset where the difference is around
one hour (or a nice multiple of one hour), which I suspect are caused by users
changing the clock due to DST changes (or timezone changes). I have only one
other case in my dataset where the difference was also very large (about 1.5 year).

The reason why we switched to the fixed epoch is because of that user changeable
clock. Due to how the uwatec/scubapro dive computers maintain their time,
adjusting the clock also changes the date/time of all previously recorded dives.
So if you download dives after a clock change, the date/time of all earlier
dives would now be off.

I believe smarttrak uses the calculated systime/devtime instead of the fixed
epoch. So that's why it shows the correct time. I'm not sure about logtrak. Can
you try to download your dives with logtrak to see what dates you get there? Can
you also double check what dates your dive computer shows in it's internal
logbook, and it's current time? I suspect it will be off as well.

Jef