Wrist Heart Rate still broken

Something is really off with the wrist heart rate - I compared my HRM pro plus and the wrist readings from yesterday's workout:

I added the yellow markings where it really seems that the software does some corrections - for a good reason in the first case, for no reason in the second case. After about 10 minutes it's fine (close to the HRM) - why not in the beginning?

Garmin-Laurie, are the software engineers aware of these wrong wrist heart rates?

Orontion said:

firmware goes nuts when HR changes rapidly,

I can confirm this as for my testing scenario with two units one on 15.77 and one on 21.19  which I described in more detail here:

https://forums.garmin.com/outdoor-recreation/outdoor-recreation/f/fenix-7-series/413518/altimeter-barometer-not-accurate-anymore

As soon as heart rate is changing rapidly OHR readings go nuts to both sides and easily as much as 35bpm both sides. It takes some time (minutes) for the system on 21.19 to synchronize if the change was really rapid. This was not the case with the testing system on 15.77 and still is not the case on the everyday system with 15.77. Both units are used simultaneously, both with nylon loop straps which guarantee imho the best fit and thus the most reliable results.

When I'm done with the testing I'll post a summary. So far it doesn't look good for 21.19. Of course I factory reset the test unit after update for good measure.

That's an awesome test you're running there - thanks a lot for your efforts!!

Will you share / link your test results in this thread or which should I follow in order to not miss them?

Hats off for such a thorough testing!

I was limited by only one Series 7 unit and just been fiddling with different firmwares on one watch.

Can confirm - issue still exists for me on 21.19.

As a side note - I noticed exact same bug in some latest YouTube reviews for Garmin watches, and it is why optical sensor on Garmin watches is considered much worse that on competitor's watches. Garmin-Laurie, will there be any updates on this? Optical HR is basically unusable for almost a year already.

I did some simultaneous testing with 23.18 vs. 15.77 and found this errant behavior - while still there - vastly improved from 21.19. It is still not as stable as I found it in 15.77, but really a good step in the right direction. the period of this erratic flipflopping with up to 30 bpms around the correct HR does no longer take place for many minutes but for 1 minute max depending on how rapidly your actual HR is changing.

Also the 2 months before 23.18 I tested 15.77 vs. 17.28 and found no difference in this aspect between 15.77 & 17.28.

As for the specifics of the testing done see my older posts above.

Looks in a real workout similar to me:
 only two 2 minutes long real bad gaps, the rest looks okish.

For a walk, there was a 4 minute gap yesterday, which seems similar to everything bad I've known for walks for quite a while now:

It was just yesterday that in the middle of a walking activity there was a phase of pulse in the sixties, which I only have when sitting or laying - not during a walk:

This doesn't happen in every walking activity but quite often - so it is totally unreliable.

How long do we have this problem now? It can't be so hard to fix that...

私もアクティビティ中に写真の様に運動強度を変えていないのに心拍がおかしくなりました。その後は追い込んでも以前のように心拍は140前後以上にならなくなっています。今が正しいのかもやもやしてます。

5289348 said:

私もアクティビティ中に写真の様に運動強度を変えていないのに心拍がおかしくなりました。その後は追い込んでも以前のように心拍は140前後以上にならなくなっています。今が正しいのかもやもやしてます。

Obviously this is around for a while, at least since 21.19. A workaround is to use a chest strap, but ofc. not one everybody will like since OHR has been very reliable in earlier versions.

You won't see this behavior in versions up to 17.28 though - so this might be an option too but ofc. comes with ramifications as for limited CIQ options.

Since this is still there I'll skip even testing 23.48 this time.