Under Review
over 1 year ago

Wrong sunset times after GPS location on the watch times out?

Wrong sunset was 19:17 a month ago, then 19:15 now I get 19:14, that tells me the watch things it's on the other half of the planet, as days are getting longer where I am not shorter.

Happening on my F7, F6, F5 and someone else is seeing this on a Venu 2.

I remember seeing this a year ago, then it was fine for the most part of summer/autumn.

Also 3 different IQ faces not just one.

https://apps.garmin.com/de-DE/apps/f39a4383-7c2c-4823-beda-09837beb1e06
https://apps.garmin.com/de-DE/apps/6bb25b91-1c3f-44cb-a8ff-98d9760d34ff
https://apps.garmin.com/de-DE/apps/3c2f7725-e9f8-4470-94bf-d94bf2440b13

forums.garmin.com/.../connect-sending-the-wrong-location-from-the-phone-wrong-suntimes-and-location-for-weather-in-iq-watch-faces

https://forums.garmin.com/outdoor-recreation/outdoor-recreation/f/epix-2/326447/sunset-sunrise-after-a-while-wrong-on-non-stock-watchfaces/1586309#1586309

  • I think the location is coming from the phone if the last gps activity was to long ago or far away from my current position.

    This also does not explain that it jumps back to the correct position / time by merely scrolling my glances or opening the activity menu on the watch. And is off again after a few minutes when connected to my phone and moving, but stays correct when Bluetooth is disconnected, some false location data seems to be coming from the phone. 

    Could very well just be an issued with the Connect software sending a bad location or zero coordinates to the watch.

    Maybe I can get some input from the developers. 

  • This can happen if the developer is not checking for a valid position—the sunrise/sunset APIs rely on a position, and if the position the app has is bad (for example, if there's poor or not GPS reception), the position the app gets can be wrong.

    We'll do some checking to make sure there are no issues in Connect IQ, but generally this is something that the app developer should first attempt to address. you did the correct thing by initially contacting the app developers, but we're limited in what we can do with this report since we don't have any code snippets from the affected apps and are unable to fix the apps (assuming this problem is due to app bugs) since they are not apps maintained by Garmin.

  • Not sure that is related as it seems stock and IQ use different sources for their sun data.

    I can have watches that did not have a gps fix for a year show the correct sun data in stock faces and widget but incorrect in IQ faces after connecting to a phone.

    Currently the F7 beta has wrong data for stock and widget but the IQ faces show the correct times (asides from the issue described here).

  • I had also an issue with the sunset/sunrise times a few days ago: 

    The weekend before last we had the changeover to daylight saving time here. Immediately the next morning the sunrise/sunset times were still correct, but on Monday morning they were suddenly shifted by exactly two hours: sunset 17:54 instead of 19:54.

    These times were displayed in both the stock watchface and the sunset widget.

    Even manually selecting a different location still gave the wrong times. After a reboot was everything correct again.

  • I tried with the two IQ WFs to the left and to the right and never got this Problem with my Tac7  on 11.28 and never observed it before BUT I'm never connected with a smartphone. A friend of mine with FR955 who is connected to an Android 12 to get OWM data has exactly the same issue especially when moving around with the connected FR955.

    So I agree this isn't a problem of the Garmin device itself but with either the communication to the smartphone - It seems the location data used are the one from the smartphone and those mess up the location data at the device. Communication is done by Garmin Connect - what changed when it went from nogo to go and back to nogo during the last year is device software and GC version.

    Best guess: GC pushes the wrong location to the device with an earlier version, that was fixed later and now is wrong again. Old repository used to build the current version?