Related
Options

Issues with Step Counting, Sleep Tracking, and Calorie Burn on Fenix 8

MaciejHassak
10 months ago

Hello,

After several weeks of testing, I still have a few issues, and interestingly, my friend who owns the same watch is experiencing similar problems, so I believe this might be a general issue.

  1. Step Counting during Car Rides: During a three-hour car ride, my watch recorded over 700 steps. This seems like a significant error, and my friend has reported the same issue. This is important to me as I pay close attention to my daily activity, and such inaccuracies are concerning.

  2. Sleep Tracking: The sleep tracking feature seems inaccurate. For instance, even when I’m awake for over an hour and a half in the morning but still in bed, the watch doesn't register me as awake until I physically get out of bed. This is frustrating, as sleep tracking is a key feature for me.

  3. Daily Calorie Burn: On rest days (without training), my watch shows that I’ve burned only around 1900 kcal. This seems incorrect considering my weight (72 kg) and fairly active daily routine. I believe the watch is underestimating my calorie expenditure on non-training days.

  4. Step Counting Sensitivity: The watch seems overly sensitive when counting steps. Even when I walk around my small apartment, it feels like the step count is too high. Also, I’ve noticed that even sitting and frequently moving my arm seems to add steps.

I hope Garmin focuses on addressing these issues, especially the step counting, which appears to be too sensitive.

  • 0 5 months ago in reply to mcbit

    Through many Fenix, Epix, Instincts and Venus of the last 6 years, I never encountered any major issues with step counting...ever.

  • 0 5 months ago in reply to Garmin-Laurie

    Thanks for the details, but relying solely on wrist movement is not an ideal way to measure steps. From what I’ve observed, the step count is fairly accurate when actually walking or when steps are logged as part of an activity like "Walking." However, the issue is that steps aren’t an activity—they are a fundamental metric that should reflect actual movement from point A to point B, not just wrist motion.

    Repetitive wrist movements, such as washing hands, brushing teeth, typing, or even shifting in bed, seem to be triggering false step counts. This is why I woke up with over 2,000 steps already logged.

    It seems like the step detection algorithm could be refined to filter out these false positives while still accurately recognizing real steps. Other wearables appear to do this more effectively, possibly by incorporating additional sensor data, better movement modeling, or machine learning improvements.

    The issue here isn’t just about explaining how Garmin tracks steps—it’s about recognizing a problem that affects accuracy and looking at ways to improve it. Rather than pointing to general support articles, would be great if this feedback was actually passed on to the engineering team. Step tracking is a core function of a fitness watch, and improving its accuracy would make the product more reliable and valuable to users.

    Is this something Garmin is working on, or are there any planned updates to address step count accuracy?  The problem is incredibly simple to reproduce .