Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Analysis of Apple Watch running data

applewatchrunner.substack.com

Post reply on HN
Print view View original

91–100 of 171 posts

Re: Analysis of Apple Watch running data

#91

Post by jalla »

ciex wrote: post #66
jalla wrote: post #42

The watch needs updated GPS ephemeris data to accurately calculate the position, received by an aGPS (Assisted GPS) server (over IP) or by satellite. This takes at least 30 seconds by satellite. Solution: Tether with IPhone before leaving until fix is achieved, otherwise keep watch outside or near a window for 10-15 minutes before running so that it can update its almanac from overhead satellites and get a fix. There…

Do you have an explanation for this being a new issue?

OP could have changed exercise habits, leading to more frequently outdated GPS almanac data.

Otherwise, newer OS versions could have degraded the fix algorithm for older Watch-models.

If newer hardware comes with cellular connectivity, newer OS version could assume that by default and not prioritise (or test) backward compatibility with older Watch-models.

Re: Analysis of Apple Watch running data

#92

Post by ip26 »

m1gu3l wrote: post #75
kodisha wrote: post #58

Semi related iWatch Rant - if you think running data is bad, try tracking your sleep. It just doesn't work. - Sleep for 3h - wake up for 1, sleep for 5 more: records 3h only. - (being a parent of a small baby) do not sleep at all during the night, sleep 4+3 hours during the day: 0 hours recorded - sleep trough the night but wake up every 90-100 minutes (baby again): 0 hours recorded Damn, this is a $500+ device, and…

Sleep tracking always seemed like nonsense to me. You either know you are getting restful sleep or know you are not, and if you are not time is probably better spent taking a proactive approach to establishing healthier lifestyle habits that will lead to more restful sleep. How much is it worth really, to wake up and know exactly how much time you were tossing around not getting rest? I’d imagine you would feel it, s…

Datapoints for each night can help you isolate the causes of poor sleep in your life. An objective measure, even a flawed one, can help a lot in that process.

Re: Analysis of Apple Watch running data

#93

Post by jupp0r »

kodisha wrote: post #58

Semi related iWatch Rant - if you think running data is bad, try tracking your sleep. It just doesn't work. - Sleep for 3h - wake up for 1, sleep for 5 more: records 3h only. - (being a parent of a small baby) do not sleep at all during the night, sleep 4+3 hours during the day: 0 hours recorded - sleep trough the night but wake up every 90-100 minutes (baby again): 0 hours recorded Damn, this is a $500+ device, and…

Regarding sleep: the most important reason why these devices suck for tracking sleep is that you have to recharge them every 24h. That's either not being able to use them during the day for some period or do overnight charging.

Re: Analysis of Apple Watch running data

#95

Post by rubicon33 »

icebergonfire wrote: post #68
kodisha wrote: post #58

Semi related iWatch Rant - if you think running data is bad, try tracking your sleep. It just doesn't work. - Sleep for 3h - wake up for 1, sleep for 5 more: records 3h only. - (being a parent of a small baby) do not sleep at all during the night, sleep 4+3 hours during the day: 0 hours recorded - sleep trough the night but wake up every 90-100 minutes (baby again): 0 hours recorded Damn, this is a $500+ device, and…

As someone who went through the same process, including using a Huawei watch previously, I encourage you to try Autosleep. It has much better sleep detection magic and exposes a few knobs that enable you to fine-tune the autodetection and also manually correct the sleep records in sensible ways. It integrates with Health, so it has the bonus of tidying up the sleep data all across the board. Not related to the develo…

Autosleep is great. I've had none of the issues the OP mentions and I've used Autosleep since I first got my watch.

Re: Analysis of Apple Watch running data

#96

Post by Maursault »

macintux wrote: post #83
Maursault wrote: post #81

Earlier quoted context omitted.

Whatever it takes to get you moving, but it is a little self-deceptive to ignore or replace natural incentives for fitness with artificial incentives of mere shadows on a cave wall. Perhaps the real world is a little boring, but it has the distinct advantage of being authentic.

I make absolutely zero apologies for the fact that the rings on my Watch helped me lose 40 pounds last year. Sure, I’m gaming myself, but good grief.

Good for you and congratulations, but it was most likely due to your own hard work along with dietary adjustment, so you need not diminish your accomplishment by giving a device as much credit as you give yourself. The Watch really didn't do anything but keep time; you yourself did everything and filled the time.

Re: Analysis of Apple Watch running data

#97

Post by jerlam »

Oras wrote: post #85
kodisha wrote: post #58

Semi related iWatch Rant - if you think running data is bad, try tracking your sleep. It just doesn't work. - Sleep for 3h - wake up for 1, sleep for 5 more: records 3h only. - (being a parent of a small baby) do not sleep at all during the night, sleep 4+3 hours during the day: 0 hours recorded - sleep trough the night but wake up every 90-100 minutes (baby again): 0 hours recorded Damn, this is a $500+ device, and…

I used Garmin Fenix 5 for sleep tracking and it has similar issues. In fact, if I sleep during the day, it doesn’t recognise it as a sleep! I checked the garmin support forums and some other users reported the same thing but garmin didn’t care to fix.

Most sleep trackers have similar issues. For example, most won't detect biphasic or polyphasic sleep correctly - it's assumed that sleep happens in one unbroken period once per day. If you get up in the middle of the night for a snack and then go back to sleep, it's a coin flip whether it will be detected as the "end" of sleep, or as a long "awake" period in the middle.

When I hit the snooze button on my phone and go back to sleep for another hour, both my Garmin and Oura are inconsistent whether that extra hour counts as sleep or not.

Garmin's even weirder than most as it asks you for your normal sleep hours when setting up the watch, which suggests that it's not as smart as it should be.

Garmin, to its credit, is slowly moving away from pure sleep tracking and using other metrics like HRV, stress, and yesterday's activity levels to calculate readiness for today's workout.

Re: Analysis of Apple Watch running data

#98

Post by jupp0r »

I used the Strava App to track my runs on my Apple Watch 2. At some point it started crashing and I tried out the built in app. It's really not for me, the data is somewhat locked in, graphs are non zoomable, etc. Most notably I couldn't send friends a url to the run.

I ended up switching to a Garmin Fenix 7 recently and I'm genuinely happy with it so far. Battery lasts more than 2 weeks and it does 90% of what I was using my Apple Watch for.

Re: Analysis of Apple Watch running data

#99

Post by ldrndll »

8ytecoder wrote: post #45
stefan_ wrote: post #28

Earlier quoted context omitted.

I thought this was a well known issue with the Apple watches? They only start looking for a GPS fix once you start the workout, which is the worst time since, well, you already started the workout and are running, making getting that fix that much harder. I think the Ultra watches now have an option to wait for a fix before starting (which is what always happens on the Garmin's).

Apple Watch uses the iPhone’s gps when the phone is close by.

This is no longer true for Series 8 and Ultra watches. What I’ve not found out is whether this true of any Apple Watch running watchOS 9 or just the latest models.

Re: Analysis of Apple Watch running data

#100

Post by jerlam »

lifeinthevoid wrote: post #19

aside: My girlfriend tracks her runs with Strava through her Iphone 12 mini. Upon zooming in, the "straight" parts never look straight and look a lot more like a triangle wave. She consistently tracks around 10% larger distance than my Garmin watch. I've checked on the map, and the Garmin distance is the accurate one of the 2.

This would also depend on how she's carrying the iPhone. While putting the GPS receiver on your wrist isn't optimal, putting the iPhone in a pocket or somewhere closer and lower on the body where it sees even less of the sky is worse.
Post reply on HN