Great question! I have the same set up as you i think and have the same confusion. Here's what gear i have:

- Kurt Kinetic smart machine
- Garmin speed and cadence sensors

I love the racing. I'm a middle of the pack cyclist. I try to set my resistance the exact same each time (ie. 5 full turns) and ensure my tires are pumped up to the same PSI. In doing this, i find the numbers i generate are quite consistent - ie. i never feel like i'm extra slow or fast on a flat course with no drafting.

When i do races (I've only done three) it doesn't feel right to stay in the same gear when going up a 10% incline - feels like i'm cheating. So....i change gears. I do my best to try and mimic what a steep incline will feel like. It's not perfect but at least it brings me closer to what others with smart resistance trainers are experiencing.

I only enter into race categories that allow this type of virtual power and so far i'm not close to winning anything so happy to continue with the current set-up.

Sorry, that may not answer your questions. Hopefully a virtual power genius can help answer better than i can.

Cam

Zwift should give you a virtual power based on the known curve. Your inride will also give you a power number as you ride. You get more power by riding faster. You get faster/slower or harder/easier by changing the gears on your bike and/or your cadence.

Disclaimer: I have never used Zwift but figure it is like TrainerRoad.

You'll want to use the inRide instead of Zwift's virtual power. The inRide is more accurate. I used inRide for about 6 month before I got my P1's and I found it to be quite accurate, my power curve didn't really shift much at all once I started using the pedals. You do need to make sure you do the spin down 10-minutes into ride, or you will get inaccurate results. I just used that 10-minutes as a warmup for whatever I was doing.

The trainer won't simulate hills, but you can basically do that through gearing, ie getting in a big gear for climbs so that you have a lower cadence, etc. But even that is not really necessary to enjoy zwift, as your in-game speed will adjust correctly based on power output.

Ok silly question, but is the Inride sensor in addition to OR in replacement of the Garmin speed and cadence sensor? Given that i already have the speed and cadence sensors, will i use all three? Or will the Inride take the place of the Garmin sensors?

Is the Inride sensor more accurate than speed+cadence?

in replacement, the inride sensor tracks the speed of the flywheel and transmits that to zwift.

In my experience (and other's I've read, including DC rainmaker) inRide is significantly more accurate that speed&cadence.

Possibly (probably!) dumb question... I have the same setup as the OP, although I still need to install my Inride 3 on the Road Machine. To date, I've been using a GSC10 for speed/cadence. Given that the Inride is doing some kind of cadence estimation, is it worthwhile/advisable to keep the GSC10 only for cadence?

Keep the garmin sensor for cadence. The inride cadence sensor isn't really accurate and won't ever go above 90 rpm, at least not for me.

I had the same experience as stevendex. The inRide was not great at cadence. I'd recommend keeping the Garmin for cadence as well

I used a standard KK road machine for a couple of years on Zwift. Always with a 23mm Felt TTR tyre at 100psi and 3 full turns from first contact.
This was an incredibly consistent setup and I was surprised to find after getting a PM that it matched within 1-2% except during sudden accelerations.

I too have a cadence sensor (wahoo). When I add the sensor to Zwift it will automatically override the inRide cadence?
Thanks and thanks for everyone's replies.

you would select the cadence sensor you want to use on the zwift startup screen where you pair devices. You pair the inRide for power and the wahoo for cadence, then whatever you use for Heartrate.

My Garmin speed & cadence sensors have been stolen - I put them on my son's bike to let him try Zwift and forgot to take them off. While his bike was locked up on the street a few days later someone swiped them. Arghghg.

So now i have a choice: Garmin Speed & Cadence sensors OR Kinetic Inride. I didn't have any real issues with the Garmin sensors so i'm thinking about repurchasing, especially since I've read some horrible reviews on the Inride.

Can anyone who has tried both make a recommendation? It's strictly for ZWIFT.

Cam

In terms of power, as long as you do the spin down, the inRide will give you more consistent and accurate power numbers, especially compared to a speed sensor and estimated power. I used inRide before switching to PowerTap pedals, and I found the power to be pretty much the same after the switch. You do need to do the spin down after 10 minutes, every ride, to get accurate power, but as long as you do that, I found it to be very accurate.

In terms of cadence, the inRide is not that good and measures a consistently lower cadence, or at least for me it did. Personally I would recommend using the inRide with a cadence sensor, if you are looking for the best accuracy out of the methods you are considering.

I haven't used the Inride but I see no point in it.
A speed sensor and a KK Road Machine or Rock'n'Roll gave me sufficiently accurate "virtual power" data. As I understand it the Inride does the same thing as Zwift, using the power curve to determine power from the wheel speed. I'm guessing the Inride roll-down is used to normalise for any inconsistencies between your setup and those used to generate the curves. I think that's largely unnecessary. You need consistency more than accuracy. Set it up the same every time and it doesn't matter too much if your figures are spot on, they'll be consistent.
As I mentioned previously in this thread, after using my KK RM with the speed sensor on Zwift for a couple of years, I got a PM (total power spider based, not one sided crank) and compared data for a few rides. I did workouts consisting of various power levels for various duration intervals. My "virtual power" as recorded by Zwift, was within 1-2% of the figures from my PM with the exception of the first few minutes of warm-up (thus the warm-up before Inride's roll-down test!) and some of the short (<1min) intervals that started with rapid accelerations. Even then the error was not huge (<10%) and occurred during the accelerations. I was impressed. Obviously a different trainer setup and different drive train losses would have shifted the power up or down a little, I got a little lucky, but it still shouldn't change too much.

My advice:
Don't bother with the Inride. Get a speed sensor (and cadence while you're at it) and set up your bike/trainer as follows....
100psi and a reasonable but not super low RR tyre and tube (I used the 23mm Felt TTR tyres and basic butyl tube that came on my tri bike as stock, since I wasn't using them for outdoor rides). I inflated to 100psi and turned the roller compression knob 3 full turns after first contact with the tyre every time. My drive train was fairly unexceptional. Ultegra 6800 with a Wipperman Connex 11SX chain and Rock'n'Roll Gold chain lube. That combo of hardware and setup resulted in accurate power data.

How do you know it’s accurate?

---

Human Person

How do I, or how will he?

How do you? Virtual and Z power are both notoriously inaccurate. Just wondering if you're also measuring with a PM.

---

Human Person

Yes, as per my previous post, I used z power for 2 years with my KK Road machine and the setup listed before I got a PM. I was pretty sure the values I was using were consistent as all the other data at my disposal was very repeatable (max effort, 20min intervals, average HR for a given intensity, etc...) but I was well aware they could be very inaccurate. When I got a PM I continued using z power on Zwift and recorded the PM values on my watch. Comparison showed surprisingly good agreement as described in last post.

I have only seen good agreement with the KK Road Machine. The Cycleops fluid 2 and others don;'t seem anywhere close.

Missed that. Sorry.

---

Human Person

You are not necessarily wrong in regards to just using the Speed Sensor, it does have a known power curve, so if set up correctly you can get decently accurate power number. The problem is that there are so many variables it is hard to get a consistent setup each time. You have to ensure the same tire pressure each ride, the same tightness to the flywheel, air temperature, etc. Some of those are hard to control. By doing the spin down, you remove the need for those variable to be exactly the same each time, allowing you to get accurate power regardless.

It can definitely be done with just a speed sensor, but it's more work, and even then there are going to be times when you will lose accuracy due to things that are hard to control. I suppose it all depends on how accurate/consistent you want/need it to be. I know I would much rather use the inRide then speed sensor, but YMMV.

In my experience temperature is not a big factor for the KK resistance unit. A small variation during first 5 minutes or so as unit warms, then very stable even with fairly big power variation (sustained periods of 100W-300W). There will be small impact but IMO it's fairly negligible. Unfortunately not so for most or all other units.

Tyre pressure and clamping amount are trivial. No harder than a roll down, and less intrusive.