Right, they constantly swerve in small amounts left-right to control balance and lane position.
I do think movement is better than rigid (mainly related to comfort), but not all movement is the same in ride feel and feedback.
The ROLLR (or any of this type that lock the front and leave the rear to wiggle) is not a good proxy for outside riding. As mentioned above, we swerve outside. That action starts with a turn and lateral movement of the FRONT tire, and the REAR tire follows that action to a reduced amount due to the frame geometry. So outside, we get more lateral displacement at the front than the rear.
The ROLLR is the exact opposite of that, with the rear moving MORE than the front tire (that is effectively locked from displacement and turning). As such, it makes perfect sense to me that this motion wouldn’t be a feeling similar to outside riding.
If comfort is the goal (as opposed to “realism”), I expect this movement is better than a typical rigid trainer, but lacking when compared to something like a rocker plate.
Started thinking, my initial fears about on/offs and over/unders are unnecessary: during indoors season (base period), I only do LSD + long Z4 intervals. When it is time for more complicated stuff in spring (build phase), I am already outside. Your list of ROLLR strengths/weaknesses is good general information for next interested persons, though.
Rocker plate was my first thought indeed. Started considering ROLLR only because it is currently sold at discount here. But if you are saying rocker is better imitating realism, I’ll go rather with it then. Thanks
I won’t claim a rocker plate is a simulation or replacement for riding outside, but specifically comparing a rocker to the ROLLR, a rocker is far better at feeling closer to outside.
Key in this is the fact that we lean the bike in small and large amounts through a large part of riding outside. Simple positioning on the road even while riding straight is done with leaning and steering action. Pure riding in line also ends up with some amount of lean with the pedal stroke for many people. I’ve studied many people on group rides and most do NOT ride with the bike locked in a vertical position (like some assume or insist is the case).
All that to say that leaning is a necessary component of any trainer motion device IMO. The ROLLER is still fixed vertically based on the front wheel restraint. Even a simple leverage input rocker does a better job giving a degree and direction of motion that is closer to what we experience outside.
I am not chasing so much for exact outside feeling but some side-to-side motion to alleviate slight backside rubbing with rigid smart trainer. Mind is strong but ass lets me down
Seems Wahoo are trying to give new life to the Rollr, with power reporting just added. This removes the old requirement of needing to use a separate power meter on the bike for power data.
It had it and they just hid it, didn’t offer it, because they wanted to focus on the “serious” riders who likely had power meters on bikes already.
It lacked it either due to #1 or them just not knowing how or bothering to figure it out in the first place. But current market & Co. conditions pushed them to figure out how to try and recoup some of the tooling and R&D investment costs by making it more widely capable and desired by users without power meters.
I thought dcrainmaker had stated that getting accurate readings from rollers was really difficult to do.
I liked the Rollr when I used it as the bike fitters. And I can see for many people just being able to quickly toss your bike on there without removing a wheel or anything is nice.
Correct. From the minimal amount of info shared in the articles, I would not necessarily guess Wahoo is going to claim a high level of “accuracy” for this change. The only article to cover any detail literally listed it as “virtual power” per Wahoo’s comments.
If true, this implies even Wahoo is acknowledging this new power data is subject to the many variables and inaccuracies already covered with trainer devices like this and even other smart rollers like those from InsideRide and Elite.
ETA: We don’t even know what level of precision they even plan to claim at this time. I just checked their product page and it still has the “old” info about requiring a power meter. It will be interesting to see more detailed info from Wahoo once they actually get it in writing.
That’s probably also why it wasn’t available at launch - may have taken this long to further refine things to make it “good enough” for a segment of people.
I’d love one for this reason. I have a 12 speed AXS bike (road with the flattop chain), so I can’t use the bike on my current 2016 Kickr (no XDR free hub available). So, my trainer setup is currently a singe-speed cross bike on the Kickr. Fine for short erg more workouts, but miserable for trying to do 2-3 hour rides on Zwift.
But, at the current price, it’s too much to bother, so I’l stick to the SS setup for now. If they ever discount them by 30% or so, I’ll be tempted. I’d also love if they added a few inches to the connection so I could toss my mountain bike on, but that’s def a nice to have, not a must have.
I like the idea of the rollr, but I still don’t know who their target market is with this thing…. Anyone seriously training is using a direct drive for the accuracy, anyone wanting a warmup trainer is probably on a feedback omnium at half the price and orders of magnitude more portable. This has been out for over a year and I’ve yet to see one in the flesh or meet someone that had one. I feel like they’re bluffing on a 7-2 off suit and keep raising their bet…
From what I can work out there is no change to the Rollr… well, there isn’t in the way of a firmware update or it reporting virtual power. It looks like RGT (maybe Zwift?) are applying a virtual power curve to the speed transmitted from the Rollr. There’s a lack of technical detail in these press releases.
After a few hours on this… looks to only work with RGT over BLE. In the absence of a paired/linked power meter on the Rollr, RGT will apply a set speed/power curve to give a ‘virtual power’ estimate. Gradient simulation is still sent to the Rollr. I assume ERG will be too. The speed/power curve doesn’t factor in SIM gradient. It’s pretty basic but works if you line the ducks up.
Testing this with Zwift and TrainerRoad presented very different experiences that I’m seeking clarification on…
Wahoo likely used some testing done on the Rollr in tandem with a real power meter, then made a “Virtual Power Curve” that correlates a given wheel speed with an approximate power level. That is how TR created their VP curves way back when.
The same process was done with the old Kinetic “Smart” trainers that used the simple magnet counter on the roller to estimate power from the flywheel rpm. It was slightly better than a wheel based power estimate since the sample rate was higher (several roller revolutions in each single wheel revolution) so it could update faster.
Strava sort of does this, but has far more factors when you consider wind resistance, road condition related to rolling resistance and such.
All that makes the Strava estimate potentially worse since one or more of those factors can be off and drastically skew the delta to a real power value.
