0.1.48 - Multiple Issues

[Garyrk001]

Thanks for all the work on this app. Been using for several months with no issues. Today with v0.1.48, I had multiple issues.

Samsung Galaxy S9
Andriod v10, Chrome browser
Wahoo KICKR Snap

Everything seems to work ok with Zwift on phone.
Issues:

1. Trainer in ERG mode, going through a workout, control is not working. At start, I can pedal at any power output without any response from the trainer to match the workout (does not seem to be in ERG mode). I also get some very quick changes to resistance levels from the trainer not in the workout for a short period - like it jumps to a steep slope. Further, changing the power via the +/- buttons has no effect on trainer. Trainer is paired with app and shows green. And it is showing power output correctly. Just not in ERG mode correctly.

2. Cadence sensor not working. I am able to pair a generic BLE cadence sensor to the app, it shows green, but the cadence is never registered.

Are old versions available? I can't use the app now.

[dvmarinoff]

@Garyrk001 Thanks for the report!

Just added version 0.1.43 here: https://flux-stable.vercel.app/.
This one is before the big Bluetooth changes. Meanwhile will try to fix Wahoo Kickr Snap for the latest version, and will let you know when there is some progress.

[Garyrk001]

Thank you!
With 0.1.43, the trainer seems to work correctly.
Strangely, I am still having problems with the cadence sensor on 0.1.43. I checked it on other apps, and it works correctly in those apps. This is not a big deal, as I can run through workouts without cadence readings.

I went back to confirm 0.1.48 trainer issues, and ERG mode seems to have the issues. The slope and resistance modes seemed ok when I manually change the level with the+/- buttons. Just ERG mode doesn't do anything.

If you need any data or debugging info, let me know details and how to collect and I'll try to provide.

Thank you for the quick response.

[dvmarinoff]

So I think I found the issue with the code for the Wahoo Kickr Snap. Power was not right, because I set the value to index 0 instead of index 1. This is now covered by the test cases, and I additionally tested with a Snap simulator. The simulator logs look good, but a test with the real device would be better.

When you have the time could you follow the instructions from this post and link the logs from the dev console here:
how to get logs
Just make sure to hard reload the page in order to get the latest version.

The cadence issue is probably related to localstorage, but the logs will show that if you can switch on and off cadence from different devices in settings during the test ride. If that's the case I'll address it in the next update.

[Garyrk001]

Thank you again for looking into it.
On the logs, if I read the post you reference correctly, it is only possible on a PC/MAC? Is there a way to get the logs on Android? If not, it might take me a little longer to find a friend's laptop I can borrow to collect the logs. If my friends are kind, I might be able to get to it in a few days.

[dvmarinoff]

No worries, a simple test with the latest version on Android would do. Just to confirm it works.

[Garyrk001]

I just did a workout with 0.1.48.1.
The trainer worked and was controlled as expected through the workout. This is great!

I then put on my QA hat and played with the+/- buttons. They worked, but we're off a little. For example, if I increased the target power to 220 in erg mode, the trainer seemed to keep me at 210. Note that I never used this before so I do not know if it is new. And honestly, I don't have a need for it. I just follow the workouts

The cadence sensor still didn't work. It pairs, but I just get -- for my cadence. I'll play with it a bit more.

Thank you again for the help and quick turnaround on the trainer issue.

[dvmarinoff]

Nice! Thanks for the test!

About the +/- buttons, the issue comes from the fact that the User can click the buttons faster than the trainer can do the power target updates.

In the old version I handled this with by denouncing the click function, but if the User clicks in a certain pattern they could still end up with the power target minus one increment, -10 watts in this case.

I will be solving this issue, but I am still looking for an elegant way to approach it.