I’m not sure why some people have so much trouble with Bluetooth. I don’t recall ever having so much as a single dropout, ever. A few years ago Apple TV and Zwift each had some problems that made setting up the various connections annoying but I’ve never had any trouble at all during actual operation.
My best guess is that it’s because all my internet connections (other than my phone) are wired. I think BT and some wifi are close in the spectrum so maybe noise becomes a real problem. My wifi is definitely in use while I do my workouts though because I’m always streaming a podcast, playing a word game, or whatever on my phone during a workout.
Maybe non Apple products Bluetooth implementation is poor. I could see that with using any Windows based gear. The few times I did use a Windows laptop or PC with workouts, the experience was just so awful I never really used it enough to see if I had any Bluetooth dropouts.
I don’t even know what “shutting it down properly” means.
I have no idea what people might be referring to regarding accuracy vs RPE or power feeling wrong. I don’t see why anyone would even bother comparing the two in that way. You shift to simply increase or decrease the resistance at a given cadence. In that respect Zwift Click works just fine, just like a cassette only without all the noise and potential for chain drop.
I think you are correct about noise particularly as we live in a townhouse with people on either side of us. When I scan for BLE devices or WiFi networks on my phone I get a LOT of potential connections.
If I connect my trainer to, say my old ipad, by bluetooth and when I’ve finished don’t close the app on it (ie swipe it from the screen) the app can keep the connection stopping me connecting for the next workout on another device.
I’m pretty sure lots of other Bluetooth “traffic” in the area will indeed cause trouble. Such a common issue with wireless stuff in general, especially in urban areas. We have just enough space around us that we don’t see anything Bluetooth that isn’t ours and the few other Wi-Fi signals are too weak to matter much.
The captive Bluetooth connection “problem” is mostly a design choice. I don’t think there are any rules or even best practices regarding when an application lets go of a paired Bluetooth device. So what you’re describing is really just normal behavior. Of course the app could be written to unpair the device when you stop using it, but it’s sometimes not so easy to define “stop using it” and it might not be the behavior most people want anyway.
Here’s another graph comparing the same power meter that started this thread with my new trainer. At first it looks mostly like a zero calibration issue but that’s not it. The Stages power meter (left arm only) was zeroed before this test and applying a simple offset to the data doesn’t quite get them to where they should be.
My left leg is considerably stronger than my right so that might be causing some problems with this comparison (Stages PM being left side crank-based). There are also the drivetrain losses but those are small compared to the leg imbalance. The difference here is HUGE!
Unfortunately, there is no way for me to apply a correction to the zero calibration or change the slope with the Stages power meter. A simple offset would help a lot but the remaining slope difference is still too much IMO. Again, definitely not the 1.5% accuracy you may think you’re paying for and obviously a BIG problem if you’re mixing power data from both meters into a training plan.
