Related
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

Pace During Workout is Incorrect

runner42
over 2 years ago

The calculation of the instant pace in version 7.11 deviates from the actual speed in about 0.5 min/km to 1 min/km.

  • 0 over 1 year ago in reply to Garmin-Kevin

    Support seems hung up on me doing their work for them and choosing not to pass my explanations and data to engineering.

    It's clear to me that the algorithm is simplistic and uses time of arm swing over anything else.  If I T-rex run my sprints it looks more as expected.  No one sprints that way. 
    To reproduce the issue - I do some interval type workout - anaerobic or sprints.
    Now that I have run some slow runs on treadmill the slow parts seem to have corrected.
    5. The sprints are very off.  I have had cyclist next to be say I am going 18mph but instant pace says 5+ min/my (should be 3:30).  My natural sprint my arm swing increases, stride increases and arms and legs go faster.  If I shorten my arm swing (natural I can obtain numbers that are closer to actual sprint though still probably not correct)
    6. reproduce: do a sprint or anaerobic exercise 
    7. measured speed by other factors like cyclist next to me, see the lap times measure significantly different than the instant pace is showing.  I have moved my arms in different ways and noticed that instant pace is sensitive to that most of all and source of issues. 
    Obviously instant pace was not tested with actual runners outside 
  • 0 over 1 year ago in reply to 6949985

    Are you all having an issue with auto-pause as well, which is probably the same bug? Instant pace not detecting that you're stopped and therefore not pausing? Seems all related to me and easy to reproduce.

  • 0 over 1 year ago in reply to Garmin-Kevin

    Dear Kevin,

    Today I had an interval work out and the bug clearly showed up again.

    I had 5*800m repeats at an average pace of 4:22-4:25/km, with 400m recovery in between. I can assure you that I was running at a very constant speed during each interval, but as you can see from the screenshot the recorded instant pace is always below the interval average pace until the very end.

    It's worth noting that during each recovery, on the other hand, the instant pace is always higher then the average.

    GPS+ glonass, HRM-Pro, open field with cloudy sky.

    Please feel free to access all my data, unfortunately I do not have the possibility to provide you the kind of details that you requested by email.

  • 0 over 1 year ago in reply to Pierangelo75

    Hey , Did you run with your phone on you during this activity?
    If you ran without your phone then it confirms something. Because I can't recall a run where this happened to me while running with my phone on.
    Every time I faced this issue was when I ran without my phone. Thinking

  • 0 over 1 year ago in reply to Shiva

    Hi Shiva,

    I always run without the phone

  • 0 over 1 year ago in reply to Pierangelo75

    I'm also using GPS + Glonass and a HRM Pro.
    If you are repeatedly facing this instant pace bug during your workout, can you do your next few workouts with your phone on you. Either on a flipbelt, or even just holding in your hand so that we can confirm if this issue occurs more frequently when not running with a phone?
    Yesterday, during a 10k run with the phone I realised that all the times I faced this issue was when I ran without my phone so after the run, I disconnected my phone and tried to do a short run, but the pace showed up correctly.
    What has your experience been like? Do you face this bug on all your workouts or only some of them?

  • 0 over 1 year ago

    https://connect.garmin.com/modern/activity/10044908505?share_unique_id=8

    My 10K race took 3 minutes to find my pace. Always the same for me around 3 minutes. GPS + GLONASS. My anerobic training affect is 30 it should be 400s as my high is 1600 and low 2600. I give permission to access my connect to help others fix this bug. 

  • 0 over 1 year ago in reply to Shiva

    I'm always running with my phone

  • 0 over 1 year ago

    I tried running with my phone (and no foot pod) today as suggested by Shiva.

    https://connect.garmin.com/modern/activity/10095956781

    Unfortunately, the results suggest the phone does nothing to help.  I would have been quite surprised if it did, but thought it was worth a try.  My best work-around is still to use foot pod for instantaneous pace, rather than GPS (see my previous posts).

    Overall, the data I have seen suggests that instantaneous pace based on GPS is acting like a moving average over a fairly long time period (60-90 seconds).  Note that in my run intervals in the above activity, I'm trying to maintain a constant pace of roughly 9 min/mile during the run, and I slow to more like 11 min pace during the rest intervals.  The saw tooth-shaped would suggest that I was ramping up speed in the run interval and ramping down speed in the rest interval.  But this was definitely not the case.  Notice also how long it takes to notice that I was walking at the start of each of the last two recovery periods.  Hence my moving average theory.  Other things could be playing into this algorithm, such as arm swing (was noted by someone) or cadence.  Either way, the algorithm is much, much too slow, and certainly slower than the last release.

    It seems plausible to me that instantaneous pace issues are at the root of many observed issues:

    1. Saw-tooth graphs of instantaneous pace

    2. Max pace that is lower than average pace

    3. Alerts during workouts

    4. Slow auto-pause

     

    Since Garmin is suggesting that these issues are hard to reproduce, but several of us seem to reproduce them easily, I would ask the following:

    KEY QESTION: Is anyone able to run short intervals (say 1-2 min) at a constant pace that is faster (by say 2 min/mile) than your rest interval, with GPS as the source of instantaneous pace, and has ACCURATE instantaneous pace?  If so, I would love to compare notes with you and see if we can figure out what setting or environmental factor is different between us.  If not, I would suggest this problem is not hard to reproduce.

  • 0 over 1 year ago in reply to Garmin-Kevin

    Exactly the same behaviour that I experienced yesterday, here the link to my activity:

    https://connect.garmin.com/modern/activity/10093692791

    I can't believe that it's so hard to reproduce. By the way, as somebody already posted here, this bug was not present in the previous firmware. Here is an example from my past activities, same hardware (FR945LTE + HRM-PRO), same settings, same location, firmware 5.07

    https://connect.garmin.com/modern/activity/8934792730