I found a bugs during a recent Erg ride:
The cadence will oddly jump around despite me keeping it very steady. For example, today it went from 95 to 148 to 124 to 107 all within seconds. This is the second ride I’ve noticed this on
I can guarantee I was not riding anywhere near 148 or 124 RPM.
Taxc Neo 1
Samsung Tab S 9+
Ok someone else mentioned this. We will review/fix.
I did some fixes in the next version that might fix this issue but we have an emulator that should be like your Neo 1, but it’s perfect cadence. Since we can’t replicate this issue, it’s hard to know if our fix works.
If you have chrome on mac, windows, android and wanted to connect to your trainer here and pedal for 20 seconds and send me the file that would help us track the issue. Or pedal long enough to see these jumps (assuming they happen in the web version which is theoretically the same logic). support@trainerday.com
I reported the same issue in another thread. I can check with my wife’s NEO in the afternoon (CEST) and send you the file.
Ok, I did a fix that I think fixes this in release coming in the next few days but since I can’t really replicate it I can’t be sure. But yes please do.
Ok, but then I would prefer waitinging for the version with the potential fix and making the analysis with the browser then.
Ok new version is there. Let me know. I hope it is resolved.
What is the version number of the version to test? Latest beta currently available for me is 8.1.5…
Yes. If it still jumps in 8.1.5 and you are willing to record on our web logger that should answer any questions and allow us to fix it quickly.
Just connect pedal for 10 seconds and send me the file to support@trainerday.com
Unfortunately, the issue persists in 8.1.5. Mail with diagnostic data is underway…
Hi Alex,
This morning I gave version 8.1.7 a try with my Tacx NEO 2T. The RPM reading dropped every few seconds from around 90 to a value of 42, and I aborted the session after about a minute. I tried again, and it dropped again, so I decided to do my workout with Zwift.
The funny thing is that, as I can see in the workout recording, every time the displayed RPM value dropped, it dropped to the exact value of 42, which is quite interesting. I didn’t have the time to test on my wife’s Neo I, but for the Neo 2T there definitely is still an issue with the RPM readings.
Oops, I just said that wrong. Old neo and 2T should be the same. Let me test on my 2T.
Good news, my 2T has the same issue so that makes fixing it easier. I was thinking wrong that old neo was different protocol then new one. Most of my testing is on wahoo or emulator. Will be fixed next week.
Finally. Neo is even crazier than I thought. :). It has 3 cadence sources. It also has what is called CyclingPowerVector which gives highly accurate stroke pressures like 10 times per second. We should only be taking one cadence source, but now the vector data was interfering. This is fixed on my 2T coming early next week. I added the “feature” of reading vector data was probably the source of this problem.
Thanks for the explanation, this is very interesting.
Just out of curiosity: I haven’t noticed these problems in previous years, even though I was already using TrainerDay on the Tacx NEO 2T back then. Has something changed in the basic algorithm, or why has this problem only started occurring recently?
EDIT: Sorry, I just saw that you already answered this in your previous posting.
This is still happening.
On one ride, I kept cadence between 87-94 but every 3-4 seconds it would show 42. exactly 42 every time.
On the second ride, I was sitting around 88-90 rpm but it kept jumping to 114. Again exactly 114 entry time, never anything else.
So not sure where the algorithm is wrong but it seems to always return one wildly different rpm in the ride over and over
New build has a confirmed fix, coming today or tomorrow. Waiting on google/apple
Ok new version with app is available with the fix.
My wife just finished today’s workout with the new app version on her Tacx NEO, and RPM readings were perfectly normal and showed no unrealistic peaks 