This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

How to get the timestamp to millisecond accuracy

Former Member
Former Member

When I export the data to .csv from the Garmin Connect Web app, I have the timestamp with an accuracy of 1sec. For some analysis, I need a better accuracy (for instance millisecond, but 0.1 second would already be enough). How can I get such a more accurate timestamp?

Thank you very much!

  • Copy the activity numeric ID from the URL line, when you are on the screen of the concerned activity, and then call the following URL in the browser:

    https://connect.garmin.com/proxy/download-service/export/csv/activity/NNNNNN (where the NNNNNN is the numeric id of the activity)

    In this way, unike with the export through the menu of the activity, you will download the CSV file with splits containing non-rounded values. All lap times will be including milliseconds (thousands of a second)

  • Former Member
    0 Former Member over 5 years ago in reply to trux

    Thanks a lot for such a fast answer; that's almost what I am looking for. But is there a way to have the same precision concerning the time at which the activity started ? Something like: 2020/03/21 @ 9:45:28.547.

    Thank you very much.

  • But is there a way to have the same precision concerning the time at which the activity started ?

    Yes, of course, all the times in the CSV file, generated in this manner, are shown with the millisecond precision, including the first starting point.

  • Former Member
    0 Former Member over 5 years ago in reply to trux

    In fact, I just obtain a summary of the activity, with only its total duration, not the exact starting time, and without the details during the activity itself. How can I get the detailed data?

  • If you speak about the CSV export, it only exports splits (laps data). If you want all keypoints, including the full time stamps, use the TCX of GPX file formats instead. 

  • Former Member
    0 Former Member over 5 years ago in reply to trux

    Thanks. Does it depend on the device? With this method, I indeed get non-rounded values for most data, but the time does not contain milliseconds. I get the UNIX time (e.g. 1565941036) and the date (2019-08-16T07:37:16Z), which are precise within a second. Should I parametrize my device in some way? Thank again!

  • Does it depend on the device?

    I do not think so.

    The raw data (https://connect.garmin.com/modern/proxy/activity-service/activity/NNNNNNNN/details) contains timestamps in milliseconds (i.e. 1583511491000), but all time entries are rounded to full seconds (last three digits are always 000). The only unrounded times including the exact milliseconds are the already mentioned lap times.

  • Former Member
    0 Former Member over 5 years ago in reply to trux

    Thank you for your answer.

    Indeed, I get all times rounded to full seconds. But, as lap times include the exact millisecond, they must be calculated knowing the exact starting time, precise within a millisecond. Do you know if there is a way to get this exact starting time?

  • Do you know if there is a way to get this exact starting time?

    The times are delivered in this way by the watch. I do not think it is stored anywhere with higher accuracy. If it is somewhere, then it could be only in the original activity FIT file on the watch (in the //GARMIN/Activity/ folder), but I bet it is stored in the same way too. Anyway it is a binary format, so you need a tool for reading it - you'll find some on the web.

    Just curious, why do you need to know the starting time of an activity accurate to the milliseconds?

  • Former Member
    0 Former Member over 5 years ago in reply to trux

    Thank you. 

    I wanted to know the exact starting time because I am working on a research project on aerodynamism in cycling. I am studying the variation of the aerodynamic force on a cyclist, moving behind another, with the distance between them. I have a distance sensor and I need to synchronize it with the powermeter. That's why I need to know the exact starting time of the powermeter and the distance sensor. Thanks again for your help!