For sure , and on that basis if they cant work out what the cause is , roll back the feature change that triggered it until such time as it is diagnosed . You cant keep piling issues on top of issues every month ( e.g. like the jersey issue) it just gets too tangled and lost.
Have had this issue since update, U-turn has been selected on a few random rides/locations around the volcano and it doesn’t U-turn. Also can’t steer off the course once I’ve tried this U-turn. (My chosen route in the volcano so I am completing the route in 6-8 minutes, U-turn some 20-30 minutes in). Speed is then limited to the U-turn turning speed. Almost as if you are stuck at the start of the U-turn procedure where speed is reduced but as turn doesn’t even start or complete, you then stuck in the U-turn procedure, and as it doesn’t complete, steering etc doesn’t work anymore. Exit and restart I get the U-turn immediately on restart. I am using Windows 10 and looking at my Strava feed now, lots of people are doing small/multiple rides on Zwift in a day, they must be experiencing this issue too and they weren’t doing this before.
Please can we have an update on this, progress made, bug confirmed in house, possible expected fix date?
Zwift gave an update a couple of days ago; progress - really?; bug has been confirmed (they know what change they did to introduce it); possible fix date - really?
This is Zwift, not some large startup with loads of financing charging double-digits dollars per month to subscribers or something.
They clearly a) disabled U-turns under certain conditions, b) instead of just disabling U-turns, they made them hang up, causing the speed limit and no turn effects; and c) got that hang-up so deep that even if you stop and save, the U-turn is still there and waiting for your next ride to finally express itself.
I guess the concept of software QA is complicated. And the concept of rolling back failed changes is also complicated.
Actually it is on both counts , secured lots of financing last year and subscription is double digits.
Despite that however , a good feature release strategy is not something that comes with a price tag parameter for doing it , it comes with a price tag for not doing it .
You release a feature and it doesn’t work . You see if you can hot-fix it . if you cant you roll back until you can .
Not doing so not only gets users annoyed , bad enough , it actually puts more pressure on development teams than is necessary and more mistakes are made and for another set of changes to be pushed over the top of this feature that might make it impossible to fix . We see the forum littered with such outcomes all over the place with features released with issues that were not fixed or rolled back . There is only one outcome for not doing this and that is tech debt.
That was sarcasm on my part, in case it wasn’t clear.
Gang - we’re working on it. We’ll update here when there’s news to share.
+1 for me on no U-turns.
I finished the whole lotta lava course and wanted to head over to the desert but missed my turn. Tried keyboard, companion app, on screen. Nothing. My choices were 1.continue riding 2. quit and select tempus fugit 3. or quit ride. I chose quitting my ride after a quick Google search during the ride.
Some post talks about Strava and how people are exploiting the u-turns that impact local legends and stuff. That should be a problem for Strava devs to figure out, not cripple zwift users.
Otherwise zwift has been great for me.
I doubt this is the case because you can do u-turns IRL to become the local legend.
Pretty sure this strava thing is red herring . To register a segment you have to pass through the start and finish point and follow the route I think . you cant just roll across the finish line and get segment time .
Stragently enough rather common sense and the correct approach that should have been implemented to address the perceived exploit and not the workaround hack that was attempted.
I know Zwift are working to fix this bug they have introduced but I just wanted to add that this has caused massive disruption to the Zwift Customer Base and destroyed the Ride Dynamics of the Platform. The change was either poorly tested or the implications were known but implications on the User Base misunderstood which is very poor either way.
Lets puts some implications out there so Zwift can learn from these:
- U-Turns are key when Everesting or Basecamping. I and others have had to postpone these attempts until Zwift fix this problem.
- I often ride ahead on Climbs with friends, descend and then U-Turn to assist. This is no longer possible.
- I will be riding solo and see some friends riding the other way but can no longer U-Turn to join them destroying a big social feature of Zwift.
- As per 3 above but when trying to join Pace Partners.
- I may take on a climb and then decide that I want to abandon, now to be honest I rarely do this because it is plain wrong, however it now means Riders will be forced to abandon rides instead of U-Turning.
And if Riders have to abandon when U-Turning and getting stuck in slow mode they will abandon and then re-join. This puts increased pressure on Zwift Servers as this will take more resources than someone simply staying in the platform.
I do hope that lessons will be learnt from this particular bug as it has done a lot of damage. Lots of people I know are outraged with Zwift and are threatening to move to other Platforms.
Word. The problem addressed by the bugfix that seems to be to blame for this is awfully trivial (unearned route badges? at most you are cheating yourself and maybe scoring a few bonus XP…) compared to the damage it has caused.
“ * Fixed an issue which allowed Zwifters to do a quick u-turn on a route and get the achievement for the reverse version of their current route.”
This is what the release notes say they were attempting to do. I still fail to see how that was a problem - didn’t you still have to a) complete the initial route and b) ride the complete reverse route? In that case, didn’t you indeed ride both routes? How is that different from ending the ride, picking the reverse route, and then riding it?
No, I guess the bug was that you could complete the route in one direction and then right after finishing that do a u-turn and get the credit for the other direction as well. Which concerns like, five routes total? (Most importantly, of course, The Sucker in Innsbruck, I guess.)
I was joking…
Confirming workaround for multiple Alpes du Zwift: no bug if you manually turn off a route
To test the results and suggestions of @P_edro_Pete, @Calvin_Mitts_64y_VV , @James_Bridges and @Aoi_Niigaki above, I’ve just done two Alpe du Zwift runs and was able to u-turn whenever I wanted without any problems, thanks to starting the Jungle Circuit instead of Road to Sky, and then doing one manual turn off that route, to get to the foot of the Alpe.
On Monday, using the same equipment and otherwise the same approach (Zwift on iPad Pro / App always force closed after each ride / Freeride), I experienced the u-turn bug after my first descent, apparently because I had followed Road to Sky and therefore had already completed that route before trying to u-turn at the base.
This suggests that anyone wanting to vEverest in the coming days (until a patch is released) should be able to just replace Road to Sky with Jungle Circuit. The spawn point is the same and there is only one manual turn required, very early on, instead of the two that one has to do if using the also useful Serpentine 8.
[Edit: added credit to Pedro]
Oops. Yeah sarcasm on forums sometimes get missed. Add to it I was mad being stuck indoors after the east coast getting dumped with 3 inches of rain ( West coast please don’t hate I know your week is way worse).
I don’t know about worse: It’s 75, sunny with a light breeze today. We could use the 3 inches of rain though…
Yep - for the last two days (Oct 26 & 27) U-turn function is not working. Tried with keyboard arrow keys, and tried the U-turn icon on the companion app (Android).
For what it’s worth: 10/26 was RGV France, and 10/27 was Tempus Fugit Watopia.
Here’s an idea: Zwift can implement an “in-place rollback” feature, where the pervious installed version is always kept. The launcher will then select which version to actually invoke based on information provided by Zwift servers.
Unfortunately, this feature’s (rollback) implementation would be irrelevant for the current issue, but will allow better handling of future issues.