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

Bug in the interval workout screen with "pace as target": for the first minute the watch displays in the interval screen a slower pace when starting from a stationary state

SETTINGS: Firmware beta 2.64. Runscribe footpod paired and connected with the watch as "source of speed =always" and "source of distance =always".

EDIT: The initial thread was wrong, so I edited the title and this is the correct issue.

When the footpod is set as source of “pace” and “distance”, the watch gives in all screen instant pace, lap pace and distance with source footpod. I tested this writing an absurd value of 150 as calibration factor, end everything is ok.

The issue has nothing to do with gps pace or foopod pace, but it has another cause, and the issue is present always, when using as source of pace and distance “gps” and when using as source of pace and distance the “footpod”.

I tried to execute the workout and I continued running in the rest periods, without stopping, and everything was fine.

In the interval screen the pace was always near the correct value.

The problem is when I start from a stationary state.

Starting from a stationary state, for the first minute the watch gives wrong pace value in the interval screen, slower than the real value. Probably the watch displays a pace that is an average of X seconds, and for the first minute it takes in account even the seconds before starting the lap, when speed was = 0 (because I was stationary). The pace shown in the workout screen is slower than instant pace and than lap pace (with lap pace I mean the average pace from the beginning of the interval). The pace displayed in the interval screen is probably an average value of seconds that don’t begin with the start of the interval, like lap pace, but contains also some seconds of the previous rest period.

I’m going to change the title of this thread in: “Bug of the firmware 2.64 with watch FR945: for the first minute the watch displays in the interval screen a slower pace when starting from a stationary state”.

EDIT: This was the old post,  I don't erase it, but I discovered that the real issue is what I described above and not what I was thinking when I wrote this post.

OLD POST:  When doing an interval workout with pace as a target, in the  interval screen with the "pace gauge" the watch shows not the footpod pace but gps pace.In another page I had "pace field" and this was displaying the correct pace of the footpod (for example 3:45/km). When I start the interval, for the first 10 seconds approximately in the interval screen is displayed "pace of footpod" for example 3:45. But after about 10 seconds, the watch alarm beeps and the watch displays the gps pace, for example 5/km. The watch then continues beeping and the speed shown changes slowly toward the real speed.

For example   I set today an audio alarm when outside the desired pace (larget from  3:40/km to 3:50/km), but with gps pace it's not useful. For short intervals of 400 m, for example, if I'm running at 3:45 /km pace (instant pace of Runscribe footpod), gps pace especially at the beginning in the first 100 meters is very far  from true value (for example 5:00/km), then at 200 m is about 4:30/km. at 300 meters is 4:00/km and it's near the true value only at the end of the interval.

I started the intervals from a stationary state.

In the other thread I opened ("Bug in the interval workout screen with "pace" as target. The watch doesn't use instant pace.") the conclusion was that  Garmin has always used  in the interval screen "step (lap)  pace" instead of "instant pace"  by design. But if I set in the watch "speed and distance always from footpod", if the watch uses gps pace this is a bug for sure.  

The interval workout screen is very nice for the graphic and colors. It reminds me what step to do and what iteration to do, and how much time I have to rest. But now I can't use the audio alarms and I can't use the "pace gauge".  I hope that Garmin will fix it.

If someone can confirm this behavior, I ask Roland to put this in the list of bugs.

  • @Phl0w the advantage of interval screen are the pace audio alerts. Of course for short intervals of 400m now the audio alerts are wrong, so Garmin should fix this as soon as possible. In my FR945 I can get pace audio alerts only if there is a "workout". Otherwise I can have only distance, time and Heart rate alerts, but not pace alerts

  • There are no pace alerts anymore in the 945? Another reason I stick with my trusted 935.

  • I have pace alerts as configurable alert on my running app, just checked.

  • It would be good if someone from Garmin could take a look into the code and explain what is happening and why it is like it is. Otherwise we will be speculating the next years how the algorithm works.

  • fischest you are right, I was remembering bad, in the FR945 there are also normal pace alerts. I'm seeing that in the normal pace alert there is only one value for hight alert and only one value for low alert. If there is the need to differentiate between various alerts and various paces (for example warm up, run, cool down), one needs to use the "my workout"function.

  • Yes, never understood why one would use these basic pace alerts. I am doing structured workouts (including pace alerts) on about 90% of my runs, even the steady runs.

  • @fishest I agree with your analysis that you made in the other post and that is probably true, I report it below:

    "Post of fischest 9 days ago: E.g. I am 100% sure that the pace alarm is based on a rolling average of X seconds and X does not start at an interval. Means: I am always slow when I start into a fast interval as the rolling average is constructed out of the jogging interval before combined with the fast started interval. BUT: it is my assumption, no explanation from Garmin."

    But I don't need that Garmin confirms that this is what is happening right now and that your explanation is true (by the way, I believe it's true). I don't need to know exactly how their algorithm works, they could use various tricks to make it works best. I would like that Garmin admits that now there is a problem, that they know that the current algorithm is not working good and that they will improve it.

  • When I have done some Workouts with pace as target (ie 8 mile run, pace between 8 min and 9 min), that data screen that shows your current pace gets "stuck" on a pace after a while and is no longer accurate.  Basically just  eventually freezes on a pace for the rest of the run.  Has anyone else expereicned this?

  • Mr Jakey, you could read this other post:

    Bug in the interval workout screen with "pace" as target. The watch doesn't use instant pace.

    The conclusion of that post is that probably Garmin doesn't use instant pace (what you call "current pace") by design. For the "audio pace alert" and in the "workout screen in the pace gauge (tachymeter)" probably Garmin is using "lap pace", or "step pace" or an "X seconds rolling average" (this is the hypothesis of user fischest), i.e. a pace that is an average. Maybe what you are experiencing is not that the pace freezes for the rest of the run, maybe your speed is quite stable and the "lap/step/rolling pace" is stable also. 

    If this is the real explanation of your issue, it's quite impossible that Garmin changes this because fishest reports that Garmin uses this  by design from always (from the FR910XT).

  • Yes, I am quite sure that this is the algorithm. Did today some short and fast intervals with jog in-between and I am always starting "too slow" in the fast laps. When I do steady runs as structured workouts (e.g. 20x1km@pace), a new lap does not start with a "too slow" or "too fast" alert.

    On my very fast intervals, I can not really look on the watch, but what I see though, is that the arrow always starts from left (too slow), when I go from jog to run and then it moves right.