Related
Options

Fenix 6 Series - 20.86 Public Beta

Julianna_ODQA
over 3 years ago

Hello Fenix 6 customers,

Our next beta version for the Fenix 6 series is now available! 

Please email the Garmin Outdoor Quality Team if you find issues in software after installing this update: [email protected]

If you’ve participated in the Fenix 6 Beta Program previously, you’ll notice a few changes from how we have managed this in the past. And further down the line, we’ll have some additional improvements to the Beta Program (more on that later!) Notably, this forum post is now the primary resource for reading the change log notes – they are not available on another webpage. (There will be more options for viewing change logs after we implement the additional improvements to our beta program.)

The links to download the software update are listed below. Select the link specific to your watch model. Please note that your computer will download a .zip file when you open this link. This zip contains the software update, as well as a folder with the file necessary for backdating software, and instructions for this process if you choose to do so. Remember, backdating software will reset all settings.

Use these links to download the zip for your watch to update to version 20.86:

Change Log for version 20.86 (changes since 20.85):

  • Added prompt to reboot if GPS is unresponsive.
  • Fixed an issue that could prevent device from booting up properly.
  • Added a new tip to notify the user that the Wrist Heart Rate Sensor is disabled. This tip will be displayed whenever accessing a heartrate-related glance while the wrist heartrate sensor is turned off on the device.
  • Updated name of the Endurance Horseback activity to be Horseback.
  • Fixed Run/Walk detection algorithm to properly calculate running & walking times.
  • Fixed an intermittent issue where the device would report inaccurate activity distance data.
  • Fixed intermittent shutdowns that would occur in the Strength activity.

Instructions for updating software:

  1. Download the .zip specific to your computer, using the links above.
  2. Connect the watch to your computer.
  3. Copy the included gupdate.gcd file (from the System_x.xx folder) to the Primary\Garmin folder on your watch.
  4. If there is a Sensor Hub_v*** folder, place the gupXXXX.gcd file into the \GARMIN\REMOTESW folder of the device. Please note, not all Beta releases will include an update.
  5. If there is a ANT_BT_BLE_v*** folder, place the gupXXXX.gcd file into the \GARMIN\REMOTESW folder of the device. Please note, not all Beta releases will include an update.
  6. Disconnect the watch from the computer and it will prompt you to update software.

After updating, your watch will be running the following software versions. These can be confirmed on the Menu – System – About pages. New versions are indicated in red.

System Software: 20.86

GPS: 5.30
Sensor Hub: 18.00 *If you have been participating in the Beta Program, you may already have this sw
ANT/BLE/BT (Pro models): 6.15 *If you have been participating in the Beta Program, you may already have this sw
ANT/BLE (non-Pro models): 3.01
Wi-Fi (Pro models only): 2.60

Please note, the beta updates released on these forums are not compatible for APAC region devices.

  • 0 over 3 years ago

    Garmin Express seems to think that I'm on 20.60 following updating to this beta (from 20.82). Looking like a string wasn't updated somewhere.

  • 0 over 3 years ago in reply to Tom Bell

    The heart rate zones is correct. Max HR had been automatically reduced by 1, probably because of a birthday earlier. 

    This problem with the mail Training Effect must have been something else. Will look at it more on the next run. I have had a long running break over the winter and trying to get going again. 

  • 0 over 3 years ago in reply to Tom Bell

    That's normal, Garmin Express shows the latest production release on the main page. If you open up "Tools & Content" in Express you will se the real version number.

  • 0 over 3 years ago

    So even by the odd sleep things going on generally since 20.8x today has been weird! Another 'false' nap has been detected.

    So...original wake time was fine...then wake time 'moved' to after my morning run...and now that turned into a nap...

    'sleep' with no REM detected again

  • 0 over 3 years ago in reply to gspinoza

    On my side I got this crazy behavior with the 20.85. In 35 minutes the altitude increased by 100 meters instead of decreasing by 15 meters.

    The first 4 minutes the altitude was correct and then becomes crazy.

    In the past with previous release like 20.50 or 20.82 I got similar behavior with linear altitude increase but with an error limited to +/- 20 meters, equivalent at 2 hPa per hour variation while the air pressure is stable or variation lower than 0.5 hPa per hour.

    I reported the problem to [email protected]. Not sure it has been taken into account as we don't have any acknoledge..

    Sometimes, the altimeter is working well during activities.

    Watch: Fenix 6X Sapphire.

  • 0 over 3 years ago in reply to 5135968

    Odd, seems it continue re-analyzing the data, and as the HR line progresses it changes its mind on when a sleep started and ended... looks more concerned with the "shape" of the hd curve than with absolute values (like, sleep is a relatively low and stable HR period which comes after a less stable and higher value period) - and the relative values are re-evaluated while the curve grows longer...

    I am day after day more curious to understand how they calculate this, it's reminding me of my graduation study where I was  doing numerical analyses of the brighness level of evaporating fluids, and... well.. to get to the bottom of the study in time I was tweekeng a bit the algorithm to make it more suitable to the specific set of images of that specific day Smiley

  • 0 over 3 years ago in reply to Frankysour

    Yes I agree - each time I have reported false sleep detection it has always occured during the post activity recovery dip in HR.

    I think every example I have posted shows this pattern, the dip in HR triggers the start of a sleep event. The revised algorithm on the surface appears to be a naiive implementation, or too sensitive to be charitable ;)

    Overnight sleep WAS ok until it was overwritten by the false nap. 

  • 0 over 3 years ago in reply to 5135968

    Maybe in this moment Garmin needs to get the data that you are seeing, but for the future a possible suggestion could be to have the "additional sleep" deactivatable by user (unsure how easy this would be.

    Surely they need to consider twerking the current algorithms for various metric calculation so that they only consider the main sleep period detected (or even better, consider the added naps as additional sleep, not replacement sleep)

    Still i can't help but find all this story very interesting!

  • 0 over 3 years ago in reply to Frankysour

    Maybe it should offer the option to accept or reject the 'extra sleep' similiar to how threshold HR detection works?

    But really your solution or my solution is just hiding the issue with the current false positive sleep detection. Plus that would indicate lack of confidence in the algorithm which isn't a good look.

    I did give Garmin permission to access my profile last week - it appears they are interested in my HR data, I think the dip in HR and the start of the false sleep events correlate too much to be a coincidence.

  • Former Member
    0 Former Member over 3 years ago in reply to Frankysour

    Sounds interesting

    Frankysour said:
    I am day after day more curious to understand how they calculate this

    So am I. This paper from FirstBeat explains a lot of it.

    Years ago I made a DIY 'sleep scoring' algorithm based on Monitoring FIT files from my Fenix 3. As far as I recall it needed 12 hours of data and 2 passes over that data to sort of guesstimate plausible start and end of sleep moments, and then calculate how much time between these moments was spent at various activity levels. It wasn't much off compared to a cheap activity tracker. Detecting daytime naps was plain impossible that way.

    But that's very different from working based on (for a large part) R-R intervals as FirstBeat does, and that's probably why they can (attempt to) detect daytime naps.