Powerdifference from 2 riders at same rpm

Hello, a colleague of mine and I drive the TDZ and I noticed something strange. Maybe you can explain that to us. We both ride a 12% incline, both with 65 cadence. I produce 170w and my colleague 260w according to zwift.
My shortest gear is 36/28 of his 36/32.
I ride on a Zwift HUB he on a Wahoo Kickr v5 (im using an Wahoo rpm sensor for RPM).
He stated 85kg me 91kg. Height me 184cm he 183cm.

Where does the 100w difference come from, although his translation is even smaller then mine?

Are your Trainer Difficulties set the same? And both trainers recently calibrated?

yes both on 100% and both calibrated

Presuming power readings were accurate, your colleague was producing more torque by spinning a harder gear.

he was on 36/32 i at 36/28 so his gear is lighter then mine.

Are those wattages from a single snapshot or an average over the length of the climb?

What bikes and wheels are each of you riding in-game?

the built in cadence might be off and not as perfect as your external sensor… also the Kickr is a higher end trainer with better accuracy. His could be more accurate and your Hub reading too low?

too many possibilities and variables to ever really know…

1 Like

Different trainers apply different resistance at various RPM.

Here an snap over One km on the three sisters radio tower finish:



i was using the TT bike with the first wheel (i think the second one u can get on zwift, i’m only level 11) he was driving the tarmac SL7 with DT Swiss wheel.

@Mike_Rowe1 I also suspected that my sensor might be more accurate because it just doesn’t make sense, the bike can’t make that much of a difference.

Bike choice doesn’t affect wattage at all. It just makes your avatar go faster/slower at any given wattage. If I was just looking at those numbers alone, my assumption would be that he was in a harder gear than you. Is it possible he was in his big ring in front and thought he was in his little ring?

He says he’s very sure. Everything over 8% he drives on the lowest gear.

Some (very) rough estimates (using this website: Derailleur gears analysis - Cadence Speed Watts) for a 12% slope, a 36/32 combo at 65rpms and his weight (hard to know exactly due to Zwift bike weights being unknown) would put him around 9kph, and needing to be doing about 290w. The timing in that graph you posted was 8.7kph, so not too far off. And his 260w is somewhat close to the 290 figure from the estimator.

For you, same slope and a 36/32 combo at 65rpm would be around 10, 10.5 kph, and you’d have been needing to do over 300w. You did that climb in about 5kph. And the 170w is nowhere near what you should have been needing to produce with that gearing, at that RPM, on that slope.

So (providing I’m using the estimator right, lol, it’s been a long day already) it looks like the issue is with your setup more than his.

It’s all very weird to me, my stats feel realistic. I drove the route there very relaxed. A pulse of 160 is quite normal for me, I can still have a conversation (max goes up to 190). My FTP is around 230w so I rode 70% of my FTP there with 160w which felt like it too.

If I just look at the distance, incline, my weight and calculate the watts, I get 185w (he 281w for his numbers). So the raw numbers fit, just the gears somehow not. Can it just be the trainers?

Remember that the resistance value is not calibrated only the power measurement. So trainers will be different.

There’s no obvious reason why two different manufacturers should design their trainers to apply the same resistance (and thus demand the same power for a given cadence and gear) at the same speed. The actual resistance when riding at a fixed speed on a particular bit of real road outdoors will be rider-specific after all, there’s no “true” value to use.

I mean, I’m a bit surprised by the magnitude of difference, but not that there’s a difference.

I disagree, I think there is a reason why manufacturers would want to be similar: e-racing, and things like Zwift in general. Manufacturers have a definite motivation to make their trainers apply (roughly) the same resistance given the same Zwift parameters as other trainers.

(EDIT: this is a messy and bad explanation of what I’m trying to say, I got distracted by watts.]
Take a case of two riders that are the same weight, with the same bikes, same gearing, on the same bit of road on Zwift (+12% grade), but different trainers. If one manufacturer is designing their trainer so that one of those riders has to put out 100w more than the other, in the same gear, same trainer difficulty, and same cadence, to get up that climb, that’s a significant strike against that manufacturer.

And if either of them are markedly different from the sort of watts that it would take IRL to make that climb, that’s another knock against them.

People want riding on Zwift to be reasonably close to IRL riding–for training purposes but also for just enjoyment. And people also want as much parity as possible for competition on Zwift. If it was known that the Hub required 100w more than other trainers to get you up a climb, that would absolutely factor into people’s purchasing decisions, right?

This is not the case Your game speed is based on the power you generate so it does not matter how much resistance the trainer have as long as it is enough to get to the required power.

Some trainers have less magnetic brake force and therefore require more wheel speed to apply the same amount of brake.

Two ways to achieve brake force:

  1. bigger and more magnets closer to the non ferritic disk
  2. nonferitic disk rotating faster

Trainer specifications usually include wording like max power 2000w at 24km/h (or atlast they should)

Saris H3:


Yeah, I didn’t explain myself super well there. Wattage isn’t the only factor that affects racing parity–resistance is important, cadence is important, because how we use different muscle groups is important.

What’s happening for OP is that his Zwift experience is very far from what an IRL experience would be. He did 170w on a 12% grade at a set cadence, and did not experience anywhere near the speed that his friend would have had the friend shifted up to the same gear and ridden the same cadence. So in order for him to achieve closer to IRL speeds up the climb (IRL for that gearing at his weight and power), speeds more like his friend achieved, he’d have had to change cadence or change gearing. And all of that affects muscular endurance. Which affects competition.

Edit: To put it another way, if you and I have the same drivetrains, the same weight, same trainer difficulty, but our trainers interpret the parameters from Zwift differently such that we effectively have different drivetrains, one of us will be at a disadvantage. I’d think that alone would motivate manufacturers.

That is true it is not apples and apples when it come to feeling. But that is the case for a 50kg rider and a 100kg rider using the exact same trainer.

The braking torque to simulate a 12% climb for a 100kg rider will be much more than for a 50kg rider. The trainer does not know your weight so it will simulate something for a set rider probably 70kg. So the 100kg rider will need more wheel speed to get the required resistance.

It is one of those things that does not have a standard so the manufactures can define it how they want. They can specify that the trainer can simulate 20% incline but they don’t mention that you will need 40km/h wheel speed and weigh 50kg.

See this website it show the power curve of different trainers at various wheel speed. This is the same for smart trainers the only difference is the smart trainer can move the magnets to change the resistance, but that all have a maximum magnet position and that will give a set curve like these showen.

Two trainers both at 40km/h wheel speed.