GPS coordinate problems

[Miika Turkia]

On Fri, Jan 22, 2016 at 4:37 PM, Dirk Hohndel <dirk at hohndel.org> wrote:
>
>> On Jan 21, 2016, at 11:17 PM, Willem Ferguson <willemferguson at zoology.up.ac.za> wrote:
>>
>> There are problems in the way the the GPS fixes are recorded in Subsurface-mobile
>>
>> Subsurface-Mobile              Companion app
>> 11:20  -27.5339 32.6842 |  11:20  -27.533912  32.68416
>> 11:30  -27.5471 32.6696 |  11:34  -27.530935  32.688359
>> 11:44  -27.5312 32.6878 |  11:59 -27.531462  32.68737
>>
>>
>> Above are coordinates for a specific dive recorded by Subsurface-mobile as well as  by the Companion app. The coordinates of the Companion app are supported by an independent Garmin GPS. Maybe Dirk is right in that there are rounding errors in the mobile application, but this is not the most important problem. There are some "rogue" measurements, e.g. the 11:30 measurement of Subsurface-mobile in the above sequence.
>
> IIRC you ran both on the same device, right?
> Subsurface-mobile is indeed truncating the values (I just looked at the code again and went "crap").

Too much information :D

> But that doesn't explain why it get's different coordinates than the companion app.
>
> I'll try to run both on a device today to see if I can reproduce and understand what happened. I assume you never went this far inland and this truly is bogus, correct?

I would say that the timestamps explain different coordinates. Only
the first line is at roughly same time, and assuming the device is not
completely stationary, the differences can be easily explained by
that.

miika