Watchface compiled with 4.2.1 cannot read bodybattery on a real device

As the title, I just compiled my watchface with SDK4.2.1. However some users told me the body battery became in-readable on their real devices. Everything works fine in my simulator. Do you have similar issues?

Top Replies

MobileDriveway over 1 year ago in reply to li2niu +1

A user just reported the same issue on his FR55 with my app.
flocsy🤠 over 1 year ago in reply to SylvainGa +1

find a workaround that you can use in the next 2 years, until it will be fixed
nnkn over 1 year ago in reply to SylvainGa +1

nth means nothing . I change nothing

All Replies

SylvainGa over 1 year ago in reply to MobileDriveway

For me it's the reverse. I needed to compile with 4.2.1 for BodyBattery to work on a Forerunner 55. User is at firmware version 7.08

But a user with the Forerunner 245M at firmware version 11.60 is experiencing that problem.

jim_m_58 over 1 year ago in reply to _psx_

By the way the app I tested uses Properties, Settings, has a glance view, and a background service that does makeWebRequests. What other features are you using where you see the problem?

flocsy🤠 over 1 year ago in reply to jim_m_58

the same old story (with a twist): what doesn't matter to you doesn't exsist, therefore shouldn't be fixed by Garmin... Just that this time even you do see the 0.1k increase in memory usage, just you don't care.

When you compare SDK v(N) and v(N+1) with the same parameters, then you don't suppose to see deteroriated performance or memory usage. WE (I know we don't count' but we are 2 and you're just 1 ;) consider it a bug.

_psx_ over 1 year ago in reply to jim_m_58

I can't switch optimization off because I don't want to return to the hopeless/bummer code without const and enums and waste time on fighting for every byte instead of writing something creative.

I've made a few tests but without basic feature supporting the dev (uploading app building with beta to store, giving app to test to somebody, fixing error, doc, examples etc.) it's a waste of time. So, no I gave up... Don't you think, it's strange that in 99.99% 3rd part devs help others instead of people who have the biggest knowledge owners of ciq?

jim_m_58 over 1 year ago in reply to flocsy🤠

There have been slight increases with different SDKS over the last 8 years, so no, .1k (probably less than 100 bytes due to rounding), does not bother or excite me.

This is a change from a x.1.x SDK to a x.2.x SDK, so more significant that a x.x.1 to an x.x.2.

If it was as big and common a bug as you think, others would see it.