Virtual shifting and Neo 2T only partially working

I have a theory about where the bug may be. I switched to SRAM and back when I first tried out virtual shifting.

I am willing to bet that there is a bug in the Zwift code that causes it to mess up the 1-24 range if you switch to SRAM (which is going to half the range, effectively) and then back. It looks like I am going to have to uninstall Zwift and reinstall to try and fix this anyway, so if I do that and it works for 1-24, I’m then going to test my theory.

I’m willing to bet this is a bug in the software and nothing to do with my trainer or what gear I’m in. Garmin/Tacx has not had reports of this and doesn’t think it has to do with the trainer.

mines been on sequential for about a month or more of use, pretty sure i haven’t tried sram style gearing, i’ll have to try it next time on on to see if i can replicate the issues

That is what I’ve done about 2 months ago (mid July) - had no issues (but my last Zwift indoor ride was July 29th, there have been Zwift updates!).

I’ve felt this a couple of times, just brief slight changes in resistance. I think it’s down to Bluetooth interference, not enough to cause an actual dropout but just enough to perhaps cause a split second ‘confusion’ between your rig and your trainer. I think all you can do is reduce possible causes of interference as normal plus try and optimise your ‘line of sight’ from your rig/bluetooth dongle and your trainer.

Tried quite few things last night… clearing the log & re-linking, switching off all unused BT & Ant+ devices, moving the antenna. Took the play controllers off so they were on the shelf next to the antenna. Nothing really made a difference.

Changing trainer difficulty made a notable difference, 100% and I could hear and feel the changes quite a lot - progressive reduction lessened the effect, zero (erg) made the glitches very subtle and infrequent.

Hey all - have you updated your Neo 2T to firmware v 2.6.0 by any chance?

Garmin’s forum is also seeing some reports of weird behavior that seem to resolve when firmware is rolled back to an older version. Would you head over to the Garmin forum and reach out to Gamin’s support folks to ask for a rollback?

Please let us know if this solves the resistance / pairing / virtual shifting issues or no?

Isn’t firmware 2.6.0. the only one that supports Virtual Shifting with Neo2T? I have updated firmware back in June when i received my Cog but i wasn’t using Zwift much during summer. Last week when i wanted to get smaller jumps between the gears i tested different Gearing options and noticed issue. What i think is happening is that gears 1-12 have normal gearing range and 13-24 have really small gear jumps so you almost can’t notice any resistance change.

@DejanPresen It sounds like you have had the same experience as me. When up near 24 on a flat surface the resistance should be so high it should be nearly impossible to pedal. Mine stops going up at 12.

I have not had time to try totally uninstalling and reinstalling Zwift to see if that will fix this. My theory is still that switching the gear configurations in Zwift messes things up and the 1-24 never returns back to normal. But to confirm that I need to start from scratch and see if 1-24 then works.

Today I managed to get the gears working again.

Yesterday I deleted knowndevices.xml but It didn’t help. Today I Paired my Assioma Pedals as Power and Cadence and everything else was the same and Gears 13-24 started to work again.

At the end of the ride I tried to replicate the Issue and I couldn’t…my gears were working no matter what combination I picked.

So I’m happy if this happens again I have a quick fix.

I’m almost sure it’s something with Zwift but i didnt figure out why this happened.

I’m still searching for answers about GEAR issue and I remembered that I picked a workout last week and used ERG On/Off and all other stuff that adjust the resistance because when I turned ERG Off I couldn’t feel any resistance/gradient changes.

Did you do workout or something similar?

Strangely, did the same thing this evening I did before and the shifting seemed to be working, and playing around I was unable to replicate the issue, so I’m not sure what is/was going on. I guess I’ll continue to monitor.

I give up, I’m going back to mechanical.

It cost me a race last night, changing resistance irratically on a hill climb (without showing a change of gear). One minute I was spinning at 100RPM, next I was grinding at 50!

Looked at my power data and it also dropped by 100W at least each time.

The difference that I think is relevant… Front chainring size. On the problem setup the bike has a 48T, the one I switched back to has a 52T. I did a race last night and although I noticed a very slight change once, the bike performed flawlessly. I think the big front ring spins up the flywheel beyond the problem zone, is the turbo faulty? Is it a VShift issue? I dont know. All I do know is that I’m leaving this bike on for now!

I have a TACX NEO 2T paired with the Zwift Smart Frame so not sure if it is the same issue but when the Zwift Ride controls Bluetooth disconnects, you lose virtual shifting (of course) and immediately after reconnecting, the shifting mode goes default to Sequential even that mine is set to Shimano Style B. The fix for me is going to Controls, uncheck the Zwift Ride (or Play controllers in case you have them) and click on them again then the shifting mode goes back to Shimano Style B. When this happens, there is a weird rubber band camera effect, which BTW is very distracting. I have 2 x Zwift Ride handlebars and it happens with both. I also have 4 Bluetooth external adapters and 2 of them are more reliable than others. In fact, I found the specific Bluetooth adapter that works better for each Zwift Ride handlebar to the point that I haven’t had any disconnection for the last 3 days.

Which one is that?

Most reliable is the TP-Link UB400:

The second best: TP-Link UB500 Plus:

I also had good luck with the TP-Link UB400 (not using a Neo though). I got a different one after I stepped on it and worried about it failing (Edimax BT-8500). That also works well.

Mike’s test results are spot on. UB400 is definitely one of the most reliable:

Hi Lee, I have exactly the same problem as you and it’s really frustrating.

I have a theory about what might be happening. It seems to me that every few minutes the trainer performs an auto-detection of the gear configuration, which is not working properly. Let’s say that sometimes it detects a 10T and other times a 36T, for example, which makes the resistance suddenly change drastically to adjust to that new configuration.

It’s possible that this auto-detection relies on the trainer’s cadence sensor to work correctly, and coincidentally in this firmware version the calculated cadence is not stable (there are other threads talking about this issue), which would explain the “random” resistance variations during the ride.

My question is: have you noticed if your trainer’s cadence is stable, or does it sometimes show spikes or strong variations? In my case, it does, because my bike is an MTB and the crank arm doesn’t even pass properly by the sensor, which forces me to use an alternative cadence sensor. A few days ago, I added a piece of metal to the crank arm to extend it so that the trainer’s internal cadence sensor could work better and, coincidentally, the “random” resistance changes during the ride, although they haven’t stopped, have improved noticeably and are no longer as drastic.

Now that makes some sense!
My fixed speed bike has slightly shorter cranks and longer chain stays!
Looking at the data, it does jump around at the same time as the extra resistance.
Thanks for the idea, I’ll pop an extension on the NDS crank to see if it solves the issue and get back to the thread.
Will be next week now though as I’m away