I’m on the OG Neo as well. Generally I have terrible perception when it comes to noticing changes like this, but the Zwift is a way worse experience. I’ve been having some fluctuations in steady state intervals that I’ve noticed in the legs. I’ve never felt this in TR, but working out pretty much every week in TR since 2016 I guess I’m pretty used to the feel!
How does doing TR ERG workouts on Zwift compare to doing Zwift ERG workouts on Zwift? I was doing Zwift a few years ago and back then their native workouts in ERG mode worked fine, as well as and maybe better than TR works for me now wrt response to power changes ramping up/down and maintaining.
I’ve only ever used Zwift for workouts and the new integration seems to work almost exactly the same. The only difference is that Zwift workouts always rounded to the nearest 5 watt level and the TR workouts will actually be at that level. 221 watts vs 220 watts.
Other than that there is no difference that I’ve noticed from a standard Zwift workout. Erg mode vs incline might be the only difference but I never tried that prior to the TR integration
I have been finding zwift erg mode fine, but I am experimenting a bit with incline too. I never really used resistance mode in TR, no idea why really. Just seems more intuitive in zwift companion app workout player.
I’m a power match user and have also noticed the slow ERG response in Zwift compared to TR. Stages Gen2 left crank PM and Kickr V4 trainer.
I’ve done a few workouts in Zwift now and have noticed it seems to be harder to keep my cadence up and workouts generally feel a lot harder than they should.
Did some searching around and think I may be experiencing this, as yet unproven, phenomenon.
Easier going uphill in ERG mode,harder downhill - Bugs and Support - Zwift Forums
Interestingly I’ve always noticed this when doing the workout in TR and just free riding in Zwift, and have posted up about it on here, although how Zwift can be controlling the trainer resistance when controllable trainer is not actually selected is beyond me. But definitely more noticeable when doing the workout in Zwift.
TR & Zwift side by side power weirdness - TrainerRoad Software - TrainerRoad
The key point is that we are all spoiled with how good and optimized the TrainerRoad app is.
Coming from Zwift to TR I immediately noticed how well TR app controlled the trainer, they seem to use some kind of feed forward control scheme that anticipates the steps and eliminates the latency.
This is true when using the power meter to measure power and also simply the smart trainer.
Zwift engineers don’t seem to have a concept of what a PID control loop is, they rely on a heavily filtered/average power in order to remain stable and that results in very laggy/slow control loop.
I’ve used SYSTM from Wahoo and that also seemed to work well with their trainers, but still not as good as TR
For anything under 15s I usually switch to slope mode, but it is still pretty good in TR.
Would be good if TR engineers gave Zwift folks a bit of a primer on how to do that stuff (or maybe it’s trade secrets)
Kudos to TR!
Here’s hoping Zwift takes notes from TR and improves this in a future update. They generally update their app every week or so, so I’m sure this issue is on their radar.
Agree on the pedaling through sand analogy.
Still not using TR zwift integration for anything because of simple stuff like the trainer resistance slider not being easily accessible on zwift
Yeah, I’m sticking with the OG 2X method for now.
I still wonder if this issue is not related to certain brands / models of trainers vs. a Zwift / TR issue.
On my H3, I notice zero difference between the two apps in ERG mode. Even the ramp-up for me on short, sharp intervals is quick on my H3…I am up to my goal wattage within 5 secs usually.
Ive got a zwift hub and notice zero difference in erg response between zwift and the TR app.
I’m also on an H3 and I find Zwift ERG unusable. I’ve switched back to the running both and letting TR handle ERG while Zwift shows me it’s cartoon goodness.
I’m on the latest Kickr and the ERG differences are noticeable for sure. Especially if you are long time TR user like I am and used to the smooth changes the TR app provides. I have not tried a 30-30 type workout yet, but for the longer intervals I don’t find it to be that big of a deal. I’ve become more aware of when interval changes are coming and am really honing in on my cadence to be ready for them. It’s almost like riding outside and shifting under load, just that slight adjustment in cadence/torque and you can get through it smoothly.
I did a workout with a lot of gradual ramps the other day and the changes there are definitely not handled smoothly the way they are in the TR app. The target only changes a few times during the ramp and not smoothly every few seconds like it does in TR. Not a huge deal with gradual ramps, but I have Vennacher coming up next week I would think the ramps in the overs of the over/unders won’t feel that great. Might be doing that workout the “old school” way.
I have the H2 and ERG mode is noticeably different between Zwift and TrainerRoad. TrainerRoad is noticeably faster and smoother when intervals start. Zwift is also not consistent where some intervals seem to hit the target power much faster then other intervals and additionally at also it tends to overshoot the target power then settle down to it.
I noticed this as well during testing and it prompted me to switch to resistance mode. It might sound crazy to loyal ERG mode folks AND my graphs look less pretty 
.
But omg folks, try it for 3-4 workouts. It’s the best. My legs have become erg mode. Even outdoors I’ve noticed finding power targets is much easier and I’ve gained a skill that I think is holding a lot of people back.
thanks, i wondered for a while if I should switch erg mode off more and try to actually ride power targets myself. Also to help once I transition to outdoor training when the weather gets nice. I think I will make the switch as well
I find the erg mode in Zwift does feel subtly different but as Power13 says, the ramp up on my Kickr Core is usually fairly quick - within 5s. Some of it looks to be to do with the averaging time Zwift uses as well as I can feel the resistance has kicked in before the number on screen has settled out.
The more annoying thing is how it deals with warm up/cool down and ramp ups, e.g., I did Thimble yesterday which is intervals of 50s at 127% preceded by a 10s ramp up from ~80-127%, whereas Zwift just gave me 10s at about 76% because it doesn’t handle the ramp properly. You can see this on the power trace - it should’ve (roughly) followed my blue line (although of course my blue line is terribly drawn):
Kind of adds to the ramp up time because if it had done the 10s ramp properly, the ramp time to 127% for the interval would have been essentially zero. Although it is only a couple of seconds there anyway.
Not the same issue, but maybe related. On my older AppleTV, the Zwift client will sometimes fail to update power targets for over a minute. It’s pretty crazy. I just canceled my Zwift sub and will go back to watching race vids. I’ll give it another shot if a new AppleTV release produces a trickle down device in my household.
I’ve notice this also.
So far I’ve doing zone 2/ recovery workouts on Zwift.
For the harder workouts I stick to the little blue bars from TR.