How Apple, Huawei, Fitbit, ouraRing crushed Garmin 965 in HR and sleep tracking

Apple already crushing it on the algorithmic front, winning by a huge margin on the above tests of sleep tracking and wrist heart-rate. Also usability wise with regards to gesture detection. Will be interesting to see if Garmin will simply be overtaken as a sports watch company the next 1-2 years - what Apple needs to do is add some more training and sports related features and it will be very hard to argue any benefits of a Garmin 965.

Sad thing is, I would love for Garmin to be successful here, but we have not seen significant improvements on basic features such as sleep-tracking and wrist heart-rate in years, where as Apple is massively closing the gap with new hardware and new software features tailored towards sports use (e.g. https://www.youtube.com/watch?v=KixZMmsRKJs )

Based on the review, I compared the heart rate measurement with a chest strap when cycling yesterday. I was only 8.5 km on the road, but the pulse measurement was almost perfect here compared to the chest strap. Part of the route went through a park with off-road paths and a lot of shaking movements on the wrist. Again, there were almost no deviations from the chest strap. I now have the feeling that the watch (may sound stupid now) only has to get to know you over a certain period of time...

I also had this experience with the 955, only after various activities has the pulse measurement become more accurate.

Also, the measurement from person to person can be different due to skin type, tattoos, hair more adipose tissue, etc.

But there are also enough other reviews such as DC Rainmaker and others that have found a very accurate pulse measurement.

But I will also compare other activities with the chest strap!

As far as I understand there is no "learning effect" built into the way wrist heart-rate detection works: If you switch your watch with somebody else, the results are comparable.

You are of course correct to point out that the quality of wrist heart-rate depends on many things, and to get really good answers about the quality of the measurements would require a representative sample of people.

For activities the professional will always go for a heart-rate strap, these are simply more reliable measurement devices, and hardly bother you during most activities, while keeping their charge for multiple months. No-brainer to use it if you care about measurement quality.

Yes, that with the learning effect is a subjective feeling of me. It's because the pulse measurement gradually became more and more accurate for me. I also had it at the 955. I can't say why it really is?!?

Coming from an Apple Watch 6 (same HR sensor as the 7 & 8), I concur with his findings of the elevate 4 sensor, which is disappointing for a “sports watch” that is >2 years newer and almost x2 more expensive.

Not just that, using an external HR monitor (Polar H10) is problematic as well.

For instance:

>ends and saves activity (last reading from external HRM is ~145bpm)

>back to watch face and external HRM disconnects

>watch HRM kicks in and initially reports ~79bpm

>HR slowly increases to actual HR and eventually settles at ~110bpm

>I manually reconnect external HRM which reports ~135bpm

I’m not sure if this is only a problem for Polar/non-Garmin external HRMs but it is quite frustrating and they have yet to correct this in the latest firmware.

We can also see how inadequately calibrated the algorithm is. If the last reported HR is ~145bpm, the algorithm should not be accepting a reading of 79bpm. If the sensitivity of the background HR measurement is too low/slow to get an accurate reading quickly, it should automatically go into “full power/active” mode to obtain a reading then only report it if it is inline with the last HR reading (from the external HRM) seconds ago. 

As for Garmin’s sleep tracking vs Apple’s sleep tracking, I didn’t find Garmin’s to be too different (for someone who is a light sleeper and tends to wake up a few times every night). They both seem to have their hits and misses. 

I must be lucky. Aside from strength I don’t really see any issues with hr. 

That said, I do use a chest strap which is always best on your runs. No point on weight lifting as it’s not measuring anything useful.  

I don’t believe that apple, Fitbit are more accurate. I think that certain watches are better for certain people.  

As a tech ohr is not perfect no matter the watch you use.  In terms of sleep tracking, it’s generally a guess at best.   Even the most accurate sleep monitors are inaccurate.  Get your 7 to 8 hours and ignore the rest.  You can’t do anything about it too little deep sleep or rem.  it’s the hours that are important.  

If you are serious about training and have spent £600 on a forerunner. I would suggest spend an extra £40 on a chest strap   

Can I ask which HR strap are you using and do you have the same issue as what I described (my comment is the one right before yours)?

Im using a Garmin dual and I’m not seeing that issues.  Like all devices there is the odd blip here and there but never had your issue that you noted. 

 That said, what you note is likely due to the ohr finding your hr.  It’s not instant.  This is an error a lot of people make when running, you need to wait for the hr to lock.  Most, including me, wait for gps and go.  That why on the ohr on some runs you see very slow increases for the first few mins and then it sorts itself out.  

For me it’s generally accurate

I'm not sure you understood what I described. Take a look at the abrupt dip in HR in the screenshot. The dip is due to the false HR (~79bpm) reported by the watch's internal HR sensor after the external HR strap was disconnected at the end of the activity.

That’s what I am saying, it take a while to lock on to your hr.  I see what you mean though that it should be smoothed.  

Just checked and I see the same after a run. It looks like it’s 79 as well.  It’s likely the default hr when it’s searching for your heart rate.