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 report] Incorrect Training Effect (FW 4.6)

There is a clear bug in Training Effect calculation. If you disconnect you HR monitor during activity (in my case battery drain) and if last HR reading was very high, then watch would assume till the end of this activity you have the same high HR reading and will calculate Training Effect based on this high HR.

Here is an example of activity. After my HR battery drain, I have been looking at TE increase on the watch itself in the last four hours of my run; you may also notice very low time in Z4 and Z5. Internal OHR was turned off during this activity https://connect.garmin.com/modern/activity/4529303506

One can easily reproduce this bug - start activity, get high HR reading from external HR (run fast), disconnect external HR, look at the TE datafiled on the watch while activity still not ended.

Top Replies

Aurelie over 5 years ago in reply to jmto +2

4.06 was marked as “4.60” in GCW...

All Replies

0 jmto over 5 years ago

There is no FW 4.6. There was 4.06 beta and then 4.10 release.

0 Former Member over 5 years ago

dpupkov said:
Internal OHR was turned off during this activity

Are you certain about this? Looking at your HR graph it looks like you had a steady reading from start to about the 13:10 mark like one would see wearing a strap. You then have dead space from 13:10 to about 14:32 were I assume you disconnected the strap. From the 14:42 mark to your finish you have a more erratic graph with a couple of spikes one may see using the WHR sensor. Are you sure this isn't what happened, or are you saying the graph from the 14:32 mark to finish was randomly thrown in?

0 Aurelie over 5 years ago in reply to jmto

4.06 was marked as “4.60” in GCW...

0 dpupkov over 5 years ago in reply to Former Member

Yes, I'm sure. WHR was turned off during the whole race. I turned it off before start, and turned on after the finish. During first half of the race I had Polar OH1 connected; between 14:42 and 24:52 I had another external HR connected - Scosche.

What I'm pointing out in this report - is area between 25:00 and 29:00 where I had no HR connected (neither external, not OWHR); one can clearly see there are no HR points on the graph. Although, during this time algorithm assumed I had my HR ~175 (last known HR datapoint) and was consistently increasing both Aerobic and Anaerobic Training Effect for the duration of ~4 hours.

0 dpupkov over 5 years ago in reply to jmto

In Garmin Connect it is clearly indicated as "Software: 4.6.0.0". Anyhow, one could easily reproduce and verify the issue. Sorry, but I do not understand your purpose of your comment here.

0 jmto over 5 years ago in reply to dpupkov

Purpose was to ask was your version as it wasn't clear to me. I didn't install the 4.06 beta, and I didn't have any idea, that GCW would show it to be 4.6.0.0. To me it's still 4.06 Beta.

0 mcalista over 5 years ago

If we strip this back to the basics, the situation here is that if you lose HR data, the watch assumes (for the purposes of TE at least) that you continue at the same effort level.

Given the options (ending the calculation or assuming all remaining HR data is at some completely arbitary level), this seems like a reasonable default assumption to make.

It just happens that in the particular situation you mention, you disconnected the HR sensor while at a high effort level.

0 jim_m_58 over 5 years ago in reply to jmto

It depends on where you look. at times it will show 4.6, as it ignores the leading zero when it's displayed.

Another thing is the last digit in the normal format is important. If the last digit is 0, it's production (in general)

so, 4.06, the last didgit is non-zero (a beta), while 4.10 (zero) is production That's why specifying the last 2 digits in a FW version matters.

0 jeomorris over 5 years ago

I have found that there is a bug here. Or at least the watch does something I think it shouldn’t.

last week I was running and had this situation where my HRM was giving values close to my max HR (185) even though I was jogging. I stopped the run after a mile and discarded the run. However although I cannot see the activity in Garmin connect (web and mobile) it still counted the activity in my training load (and correctly said it was anaerobic, which it would have been if I was running at max HR). The time my HR was falsely reporting as high is still there though in my daily HR data

I think this is because it is stored on the watch and processed there - so to fix this the watch should discount and forget that activity at all.