As I quoted above, is it a known thing that this methodology is inaccurate – ie. somehow tested and proven?
To determine left and right power balance, the Elite trainers use a method called “Pedalling Analysis”. The high rotational speed of the axle allows the optical sensor to gather hundreds of data points for each pedal stroke. The trainer analyzes the power output during the different phases of each pedal stroke, attributing power to either the left or right leg based on the detected torque changes. This data is then presented as a percentage split between the two legs, allowing riders to monitor any imbalance.
Beyond this, what I’d like to know, is there something a trainer manufacturer can do (ie. with firmware) to send L/R data that Zwift can handle and display - if they so desire? Is it a known standard that trainer makers can just opt to support?
Seems hardly conclusive testing on the matter and provides very if none evidence as to how much better non-trainer alternatives are. That said, it’s irrelevant, unless Zwift is saying they won’t provide L/R power readings unless there’s some proven accuracy rate. Which would be comical, since they don’t have this qualifier in general with regard to the accuracy of power meters and allowing their data into the game.
That’s right, it’s a quick demonstration, not in-depth comparison testing.
I understand the reason why Zwift hasn’t included things like the Neo products is because they do not transmit the the data in the same way as most power meters. If they did, it would be available regardless of its utility. I have no idea if it’s possible or if they have any desire to add it.
I guess what you saw on the Kappa Climb was a KAPPA(I haven’t seen it yet). Kappa is a fictional animal that appears in Japanese folk tales.I’m going to meet the monster with the red eyes that you told us about
Oh you’re right, that’s totally what it is!
Neat! Thanks for sharing!
Never considered looking up where the name Kappa came from, how’s that for some Zwift Folklore!
Here is a grab of the statue versions around downtown Neokyo:
What do you want to see? I have a few 100 (if not 1000) dual recordings I could likely quantify the differences… Which the Internet would then throw back at me with a with “Sure. For YOU. What about a larger sample size”
Fact remains that when a meter is measuring at a single point, it’s not going to give L/R balance as accurately as a true dual measurement meter.
What I want to see is Zwift still take the estimated values provided from the trainer and display them in the HUD.
I’m not sure how the conversation or spin, has now shifted to be that Zwift isn’t showing Trainer-based L/R values because they’re trying protect the masses from viewing potentially inaccurate estimates on the L/R split? The explanation Zwift provided was simply that it was easier to get this data from pedals’ BT broadcasts.
I suspect there’s no ‘standard’ within the BLE protocol for smart trainers (FTMS) to report power balance. Maybe there is… checking….
ok - There’s no mention of ‘balance’ in FTMS_v1.0.1.pdf - the technical spec trainers will likely/should support. There’s a ton of references to it in the CPP_v1.1 (Cycling Power Profile BLE spec sheet) - the technical spec power meters will likely/should support.
All speculation - The smart trainers that are reporting balance are likely doing it as part of a custom extension to the FTMS protocol. How they’ve implemented it could be with some kind of proprietary method/reporting/fields. Fine for their own developers to implement support for, impossible for third parties to do without being supplied the spec or reversing it.
After reviewing the BLE technical specifications, the explanation Zwift provided is likely 100% correct and true.
This sounds very likely. But then this is what we’d hope Zwift and the major trainer manufacturers might have collaborated on over the last 6-12 months in order to solve. Then again, maybe it’s because Zwift’s most major partner (Wahoo) doesn’t seem to offer L/R data at all, and they wouldn’t want them looking deficient.
I did come across this older video of yours – good stuff! Do you by chance have anything more updated on the My-Etraining app? I assume since then, their trainers and software have probably improved a bit.
I haven’t revisited MyETraining in a LONG time! I’d hope they’ve made some updates. If they send over their new Rivo trainer I’ll be sure to test their pedal analysis with it.
So perhaps if enough customers appealed to eg. Elite and Tacx to tweak their respective firmware in order to broadcast the calculated L/R power data in the same way that pedals do, then voila it would work? At least as far as Elite goes, they’d have to weight the pros/cons of determining if it’s better to give away this data but provide more mass-market appeal for the (likely larger) market that buys their trainers mostly for exclusive use with Non-Elite brand apps, or to protect their subscription fees to My E-Training where the L/R data is only available for a paid tier of their app.
It does seem a missed opportunity with the way it has been implemented and it now puts the community at loggerheads with those wanting this data and race rules..
Its made it a bit of niche implementation (BT, Power meter only, non smart trainer compatible) when it could have been something used to benefit all.
Add in Core support as another niche implementation and its probably fair to ask, was there better opportunities to throw resource at or lower hanging fruit?
