Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

NME42

macrumors 65816
Sep 15, 2019
1,188
659
Had a support call scheduled for 1pm today. Well, did go different than what I thought. Phone call was on time and then I was listening to music for about 15 minutes until I hung up. Thought somebody from 3rd level support or even development is calling me but that was not the case.
 

catean

Suspended
Jun 16, 2013
531
555
London, UK
It could be the case, that on the cellular model A-GPS is used more instead of full GPS. A-GPS (assisted GPS) is GPS data from cell towers, not from GPS satellites. So, when you run with LTE on and good reception, the track could be 100% reliant on A-GPS, and with LTE off it could trigger the full GPS. I can imagine they would do this for power management purposes.
 

rbart

macrumors 65816
Original poster
Nov 3, 2013
1,184
876
France
Your are probably right, from the beginning I think Apple uses LTE position instead of GPS in some situations.
But it's hard to prove it, only Apple knows why ...
 

NME42

macrumors 65816
Sep 15, 2019
1,188
659
Would be interesting to know if and why this is only the AW5 that is concerned.
 

rbart

macrumors 65816
Original poster
Nov 3, 2013
1,184
876
France
I am not sure it's limited to AW5, I am not sur all AW5 LTE has the same issue.
It can also depend on mobile operator settings ...
Only apple can troubleshot
My guess is that very few people use AW to track runs precisely, most people are casual runners that don't care about accuracy and Apple has not put an high priority to investigate on this.
 

NME42

macrumors 65816
Sep 15, 2019
1,188
659
Yes, understand. Guess a lot of folks just take it as it is, you are right. Would be great to search Strava or so to see if others have the problem too. Maybe we are just the only two in the world ;-)

Todays run was under four bridges where the AW struggled sometimes. Today is excellent again. It is really random or we do not see the factor that determines struggling.
 

MJ22

macrumors 6502a
Oct 3, 2017
591
762
This thread is very interesting. I have an AW4 with LTE and I am constantly measuring long compared to everyone else I run with, whose watches range from Garmins of various versions to other Apple Watches. Usually I am at around 5.1 miles to their 5 miles. I’m not so concerned on a day to day training run but last week I did a 9 mile race and thought I was on target to hit my goal time, as it turned out my watch measured 9.27 miles and I was a few seconds per mile too slow. I was really disheartened. That type of overestimation is not acceptable to me! I know the course is accurate because it’s wheel measured and has been going on for 110 years. When I hit the 9 mile mark on my watch I was exactly on pace, but still had a ways to the finish. I missed an award by :30, too. I’m going to try turning off LTE for my next group runs to see if that improves how my distance compares to others.

edit: I’ve done this same race with an AW two other times. Last year was so bitterly cold that my watch actually shut down and restarted so I can’t use that for comparison. The first year I used an Apple Watch, it was a Series 3 with LTE and it recorded 9.07 miles - I can live with that deviation. How can GPS actually go downhill in a “better” version of the watch?!
 
Last edited:

rbart

macrumors 65816
Original poster
Nov 3, 2013
1,184
876
France
I also have this type of distance error, and it's not linked to LTE.
I have made a race last sunday night.
For most people it was 24km, for my AW : 24.5km
I have also ran on official 10km, my AW gave me 10.2km
It's 2% optimistic, and I don't know how to fix that
 

NME42

macrumors 65816
Sep 15, 2019
1,188
659
I have also ran on official 10km, my AW gave me 10.2km
It's 2% optimistic, and I don't know how to fix that

I have these 2% too compared to Garmin and GPSies. But it is consistent from run to run.
In trainings I do not care too much, in races I would only run with my Stryd (as I also did with my Garmins).
 

NME42

macrumors 65816
Sep 15, 2019
1,188
659
Had a call today with Apple Support for almost an hour. Will also read here, hi ;-)

I will send additional data like tracks, etc. Together we will get this resolved, the hardest errors are the funniest ;-).
 

rbart

macrumors 65816
Original poster
Nov 3, 2013
1,184
876
France
Nice.
They was supposed to call me back last Sunday but they didn't.
I will ask them again.
 

NME42

macrumors 65816
Sep 15, 2019
1,188
659
Strangely they called me Sunday on mobile phone and today on landline. No idea why.
 

rbart

macrumors 65816
Original poster
Nov 3, 2013
1,184
876
France
I am a bit afraid they will never fix these accuracy problems.
So small problem, for too few users ...
Are you optimistic after discussing with them ?
 

NME42

macrumors 65816
Sep 15, 2019
1,188
659
Too early to conclude. We‘ll see what happens.
From my call, I would say yes, they are interested. But right, it could be a minor problem as only a small amount of people seem to be affected. It all depends on the support and developer who get the tickets. If they are nerds who want to fix that, we‘ll fix it together.
 
  • Like
Reactions: MJ22

MJ22

macrumors 6502a
Oct 3, 2017
591
762
I am chatting with support and all the person told me to do is recalibrate my watch. I had calibrated it when I got it. I don’t see how that would help.

Is it possible that when I upgraded my phone, it somehow affects my GPS? It does seem to have gotten worse since I upgraded to 11 Pro. For what it’s worth, I do not run with my phone.
 
Last edited:

diba

macrumors member
Feb 15, 2019
82
63
Might be an algorithm issue. I have FITIV app installed as well. All my runs are "shorter" on FITIV than with any other app. Not only while using FITIV or workout app but with imported activities from health as well. I.e. yesterday 5.43 km in activity app, 5.28 km in FITIV (imported). Some days ago 7.53 in activity, 7.33 in FITIV. I am talking not about running same distance twice using different apps. Its the data, which is taken from activity/health which leads to different results.
 

rbart

macrumors 65816
Original poster
Nov 3, 2013
1,184
876
France
You are right.
All the apps use the same GPS data from Apple API.
But each app can calculate metrics a bit differently, by smoothing or not results.
 

rbart

macrumors 65816
Original poster
Nov 3, 2013
1,184
876
France
Finally, I have called Apple again to follow up my support case.
They have asked me (again) some basic information ...
I don't have high expectations on what they will do ...
 

TTTedP

macrumors 6502
Nov 27, 2017
320
322
Just wanted to chime in that I had bought an AW5 LTE and noticed the same issues with poor track rendering. Did a bunch of tests with the AW5, Garmin 935, and Edge 530 / Wahoo Bolt when biking. In the end the Garmin 935 I have been wearing on my wrist the last two years had the best tracks (when overlaid to maps).

People say its only rendering and that actual distances should be close (which they were - if you consider any deviations <5% "close enough"). In the end, the tracks drove me nuts. I returned the AW and keep wearing the Garmin.
 

MJ22

macrumors 6502a
Oct 3, 2017
591
762
I ran with a friend today to see if the recalibrating helped. It did not. After 4 miles. I was .1 of a mile ahead. That’s a lot for a shorter run like that.

Another friend ran with us who also has a series 4 like me and she was closer to the same distance as me. What is Apple doing that makes it so off from other watches? Could it really be the lack of locking a signal? Is having LTE turned on a problem? I can’t race with this much of a deviation.
 

NME42

macrumors 65816
Sep 15, 2019
1,188
659
Distance comes from pedometer. Don’t think switching off LTE does help.
 

rbart

macrumors 65816
Original poster
Nov 3, 2013
1,184
876
France
Not only, they combine podometer and GPS. If the GPS is not accurate, the distance will be false.
Podometer is only used when GPS reception is bad
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.