Zwift speed differences to Garmin
ForumNewbie
Posts: 1,664
My ANT+ stick arrived today, so I was able to try out Zwift for the first time. I enjoyed it as definitely more motivational than just watching a training video.
The only thing that I found strange is that because I'm not using a Smart Trainer, I have found that the speed and distance is calculated by Zwift based on my weight, estimated power output, gradients, and even drafting other virtual riders. So my speeds varied greatly on the Zwift screen compared to my Garmin. Overall my average speed downloaded directly to Strava from Zwift was over 2mph greater than that on my Garmin over a 30 minute ride. Even the distance calculated by Zwift is different.
I'm thinking that I should maybe delete the ride downloaded from Zwift and replace it with the ride data downloaded from my Garmin.
Which ride data does other Zwift users download to Strava - direct from Zwift or from the Garmin?
The only thing that I found strange is that because I'm not using a Smart Trainer, I have found that the speed and distance is calculated by Zwift based on my weight, estimated power output, gradients, and even drafting other virtual riders. So my speeds varied greatly on the Zwift screen compared to my Garmin. Overall my average speed downloaded directly to Strava from Zwift was over 2mph greater than that on my Garmin over a 30 minute ride. Even the distance calculated by Zwift is different.
I'm thinking that I should maybe delete the ride downloaded from Zwift and replace it with the ride data downloaded from my Garmin.
Which ride data does other Zwift users download to Strava - direct from Zwift or from the Garmin?
0
Comments
-
You've already explained why the distance and speed is different, because it's simulating gradients, your weight, drafting etc etc, whereas all that your Garmin is measuring is the speed of your rear wheel - which for most purposes is meaningless, on a dumb trainer all that does it provide an input to the game.
If you're wanting something to track your progress, ignore your Garmin and put the Zwift data on Strava.0 -
My thinking was that my speed and distance was accurate on the Garmin, because if I was spinning in say 34/15 at 100rpm my speed replicated almost exactly what I should be getting in that gear at that cadence. If in another gear and at another cadence the speed was also what it should be as per this chart:You've already explained why the distance and speed is different, because it's simulating gradients, your weight, drafting etc etc, whereas all that your Garmin is measuring is the speed of your rear wheel - which for most purposes is meaningless, on a dumb trainer all that does it provide an input to the game.
If you're wanting something to track your progress, ignore your Garmin and put the Zwift data on Strava.
http://www.bikecalc.com/speed_at_cadence
Accordingly I had been satisfied that the Garmin speeds on my Elite Crono Fluid turbo (with no resistance settings) was accurate. I thought Zwift speeds seemed too high as I was reaching 39mph in the small front ring, whereas I have hardly ever reached that speed on the road, even downhill. I'll see what happens to my overall averages on Zwift when I use it again.
I just wondered if others thought their Zwift speeds were higher than they expected.0 -
My average speed is way down on what I'm normally able to sustain. Everything is setup correctly. All the values I've input to Zwift are honest.
I only use the values on Zwift as the real time data, the Edge isn't involved as it's fairly pointless in this scenario.
BTW, I am only using rollers (with the Garmin ANT stick and wheel speed sensor) and communication doesn't ever seem to be a problem, but the effort to climb any of the hills is way more than I would normally expend pedaling uphill.
The average speed is more concerning though, a drop of 7-10 mph average is what I'm seeing after an hour.
It kinda puts me off, and just plod away watching a film on the rollers instead.
Any suggestions?0 -
No as Mark says Zwift converts your w/kg to an approximation of speed given the virtual terrain you're riding on its the only way that you can do it on a dumb trainer otherwise riding up a hill would be no more difficult than riding on the flat/downhill. Re going fast downhill if you're anything like me you probably don't normally put any watts in downhill, especially in the little ring you'd be going too fast for the gear, that's where the simulation part comes in.Saracen Tenet 3 - 2015 - Dead - Replaced with a Hack Frame
Voodoo Bizango - 2014 - Dead - Hit by a car
Vitus Sentier VRS - 20170 -
Qube - it sounds like your rollers aren't supported, my averages are down on the real world in Zwift but nowhere near that much. I'm hoping that Zwift is under estimating my power because I'm way down on the people I normally ride with who use power meters, my way of thinking though is if you're working hard it's doing you goodSaracen Tenet 3 - 2015 - Dead - Replaced with a Hack Frame
Voodoo Bizango - 2014 - Dead - Hit by a car
Vitus Sentier VRS - 20170 -
prawny, I know what you're saying about the hills, but I've just finished another short Zwift ride. On the flat part of the course (gradient showing at about 0%) when I was in the 34 front ring and not drafting anyone, my Garmin was consistently showing around 15mph, but on Zwift I was consistently showing around 20mph. I was not even pushing myself hard and was easily staying at 20mph on the flat according to Zwift, which would be nice, but I know I wasn't actually doing that speed!No as Mark says Zwift converts your w/kg to an approximation of speed given the virtual terrain you're riding on its the only way that you can do it on a dumb trainer otherwise riding up a hill would be no more difficult than riding on the flat/downhill. Re going fast downhill if you're anything like me you probably don't normally put any watts in downhill, especially in the little ring you'd be going too fast for the gear, that's where the simulation part comes in.0 -
The speed of your rear wheel on a turbo trainer means nothing!0
-
The speed of your rear wheel on a turbo trainer means nothing!
This, turn the Garmin off and give it deathSaracen Tenet 3 - 2015 - Dead - Replaced with a Hack Frame
Voodoo Bizango - 2014 - Dead - Hit by a car
Vitus Sentier VRS - 20170 -
speed means nothing on a turbo. Wouldnt bother me if it told me I did 1 mile in an hour as long as I was riding to the planned power0
-
If I am out on the road in say, gear 34/15 doing 90rpm, I would be travelling at 16mph. On the turbo in the same gear and at the same cadence I am also doing 16mph and that is what shows on my Garmin, so it is accurate.The speed of your rear wheel on a turbo trainer means nothing!0 -
If I am out on the road in say, gear 34/15 doing 90rpm, I would be travelling at 16mph. On the turbo in the same gear and at the same cadence I am also doing 16mph and that is what shows on my Garmin, so it is accurate.The speed of your rear wheel on a turbo trainer means nothing!
It might be the case but Zwift doesn't care about that, all that matter is your power input. After all you may be doing 16mph in 34/15 but your power output would be vastly different doing that on a 5% slope compared to on the flat.0 -
If I am out on the road in say, gear 34/15 doing 90rpm, I would be travelling at 16mph. On the turbo in the same gear and at the same cadence I am also doing 16mph and that is what shows on my Garmin, so it is accurate.The speed of your rear wheel on a turbo trainer means nothing!
Its meaningless in that the relative power required in both instances is not comparable and so speed on a turbo is, as already mentioned is meaningless.0 -
The speed of your rear wheel on a turbo trainer means nothing!
This, turn the Garmin off and give it death
And again, this. Switch your Garmin off when using Zwift. It becomes redundant.0 -
I assume in your case that you just provide Zwift with power output. Zwift has no idea what gear you're in. Since Zwift cannot alter the braking on your turbo, it just alters your virtual speed. ALL the turbo does is provide some random resistant against which you can deliver power. You could ride around in one gear at a steady cadence and steady power (steady speed in the real world) and your Zwift speed will vary according to the virtual world.
As above - forget the GarminROAD < Scott Foil HMX Di2, Volagi Liscio Di2, Jamis Renegade Elite Di2, Cube Reaction Race > ROUGH0 -
I assume in your case that you just provide Zwift with power output. Zwift has no idea what gear you're in. Since Zwift cannot alter the braking on your turbo, it just alters your virtual speed. ALL the turbo does is provide some random resistant against which you can deliver power. You could ride around in one gear at a steady cadence and steady power (steady speed in the real world) and your Zwift speed will vary according to the virtual world.
As above - forget the Garmin
Exactly - I don't think the OP has a power meter - so is just using cadence and rear wheel revolution to feed into zwift - even if you put a power meter on then the Garmin still wouldn't nescersarily read the same speed as Zwift.
The only way you're going to get it to read anywhere near the same is to use a smart trainer that Zwift can control.0 -
Correct, I don't have a power meter or a smart trainer, so Zwift can only go on my cadence, but doesn't know what gear I'm in. So if I am doing a cadence of say 100rpm on the flat in an easy gear while warming up, Zwift shows my speed as over 20mph which I know is unrealistic, so when I have download Zwift rides it is definitely not true average speeds. I'd be kidding myself if I thought I was actually achieving these average speeds.I assume in your case that you just provide Zwift with power output. Zwift has no idea what gear you're in. Since Zwift cannot alter the braking on your turbo, it just alters your virtual speed. ALL the turbo does is provide some random resistant against which you can deliver power. You could ride around in one gear at a steady cadence and steady power (steady speed in the real world) and your Zwift speed will vary according to the virtual world.
As above - forget the Garmin
Exactly - I don't think the OP has a power meter - so is just using cadence and rear wheel revolution to feed into zwift - even if you put a power meter on then the Garmin still wouldn't nescersarily read the same speed as Zwift.
The only way you're going to get it to read anywhere near the same is to use a smart trainer that Zwift can control.0 -
Correct, I don't have a power meter or a smart trainer, so Zwift can only go on my cadence, but doesn't know what gear I'm in. So if I am doing a cadence of say 100rpm on the flat in an easy gear while warming up, Zwift shows my speed as over 20mph which I know is unrealistic, so when I have download Zwift rides it is definitely not true average speeds. I'd be kidding myself if I thought I was actually achieving these average speeds.
It doesn't use your cadence (except to animate your avatar)
It uses the speed of your rear wheel mapped onto a power curve for your specific trainer which then generates a power number in watts, which is then used as the input into the game, your rider in the game then goes at a speed determined by the power input. So there isn't a direct relationship between your speed and the riders speed.
It does rely on your trainer being calibrated correctly, a bit off and your power numbers can be entirely wrong.
I have a smart trainer which does a reasonable job of power numbers, but my average in Zwift is always a couple of mph above real life, I think mostly because there's no wind, no junctions, traffic, and importantly never any braking!0 -
Apologies for the necropostThe speed of your rear wheel on a turbo trainer means nothing!
This, turn the Garmin off and give it death
And again, this. Switch your Garmin off when using Zwift. It becomes redundant.
I keep my Garmin on and use it to record my power data. I've got a problem with my left knee and it's good to see whether I'm managing to maintain something like equal balance between Left and Right (Vector user)
Agree that the speed calculation on a turbo is next to worthless though. Zwift is also much, much worse at 'guessing' power (obviously) than it is with 'real' power data. I've ridden with people on Zwift using 'smart' trainers who I would absolutely crush in the real world, only to see them put out 5w/kg for 20 minutes at a time.
I treat Zwift as a community-enabled Sufferfest - and am starting to use the workout mode to replace more structured training efforts. Anyone who 'races' on it is kidding themselvesFat chopper. Some racing. Some testing. Some crashing.
Specialising in Git Daaahns and Cafs. Norvern Munkey/Transplanted Laaandoner.0 -
In the end I didn't go beyond the trial period in Zwift as the speeds estimated by Zwift for my turbo rides were much higher than I was actually capable of, and it just didn't feel right downloading that data to Strava.
Apologies for the necropostThe speed of your rear wheel on a turbo trainer means nothing!
This, turn the Garmin off and give it death
And again, this. Switch your Garmin off when using Zwift. It becomes redundant.
I keep my Garmin on and use it to record my power data. I've got a problem with my left knee and it's good to see whether I'm managing to maintain something like equal balance between Left and Right (Vector user)
Agree that the speed calculation on a turbo is next to worthless though. Zwift is also much, much worse at 'guessing' power (obviously) than it is with 'real' power data. I've ridden with people on Zwift using 'smart' trainers who I would absolutely crush in the real world, only to see them put out 5w/kg for 20 minutes at a time.
I treat Zwift as a community-enabled Sufferfest - and am starting to use the workout mode to replace more structured training efforts. Anyone who 'races' on it is kidding themselves
Zwift speeds are maybe more realistic with a power meter and/or using a smart trainer. I know it is not the same as on the road, but the speeds recorded on the Garmin on my turbo are more accurate than the Zwift estimated speeds.0 -
If I am out on the road in say, gear 34/15 doing 90rpm, I would be travelling at 16mph. On the turbo in the same gear and at the same cadence I am also doing 16mph and that is what shows on my Garmin, so it is accurate.The speed of your rear wheel on a turbo trainer means nothing!
Its meaningless in that the relative power required in both instances is not comparable and so speed on a turbo is, as already mentioned is meaningless.
This. Unless you're using a real power meter (not virtual power) then indoor training speed is only useful in relation to other indoor training sessions, assuming the wheel/tyre/pressure/ambient temperature etc all remain the same.0 -
Why does it matter what speed data is recorded for a trainer ride that is uploaded to Strava? Nobody's going to be fooled by it (if you're into Strava average speed willy waving?) and Strava doesn't advertise your average speed on your profile so far as I'm aware.
As mentioned all that really matters is whether your setup is the same between rides is the same and therefore your proxy power is consistent between workouts.0 -
Yes, turbo speeds on the Garmin can be compared between sessions. All I was saying is my Garmin speeds on the turbo are more in line with my speeds on the road than Zwift speeds. The estimated Zwift speeds based on virtual power do not relate to my cadence or gear that I'm in, so are unrealistically high.
If I am out on the road in say, gear 34/15 doing 90rpm, I would be travelling at 16mph. On the turbo in the same gear and at the same cadence I am also doing 16mph and that is what shows on my Garmin, so it is accurate.The speed of your rear wheel on a turbo trainer means nothing!
Its meaningless in that the relative power required in both instances is not comparable and so speed on a turbo is, as already mentioned is meaningless.
This. Unless you're using a real power meter (not virtual power) then indoor training speed is only useful in relation to other indoor training sessions, assuming the wheel/tyre/pressure/ambient temperature etc all remain the same.0 -
Doesn't really matter whether it is unrealistically high or low though?0
-
It does to me. I'd rather download realistic and consistent speeds as related to the effort I was putting in, and I get that with the Garmin speeds.Doesn't really matter whether it is unrealistically high or low though?0 -
Well fair enough, but I don't really get why it matters.
As mentioned by everyone else, speed on an indoor trainer is a completely meaningless number when compared to real life cycling. So long as it's consistent between workouts it's fine.0 -
The estimated Zwift speeds based on virtual power do not relate to my cadence or gear that I'm in, so are unrealistically high.
Your cadence and gear aren't at all relevant, they are just the means of spinning the back wheel against whatever resistance you have dailed in.
That said if you're using just rear wheel speed, i.e. zPower and not a smart trainer or power meter then it's never going to be *that* accurate.0 -
My turbo is an Elite Crono Fluid Elastogel and it has no resistance settings. The resistance is just created by my weight on the rear wheel as it spins on the turbo. I therefore assume the Garmin speeds on my turbo is the same as the speeds you would see if using rollers.The estimated Zwift speeds based on virtual power do not relate to my cadence or gear that I'm in, so are unrealistically high.
Your cadence and gear aren't at all relevant, they are just the means of spinning the back wheel against whatever resistance you have dailed in.
That said if you're using just rear wheel speed, i.e. zPower and not a smart trainer or power meter then it's never going to be *that* accurate.
The point I was trying to make is that if I was cycling in say, gear 34/15 at a cadence of 90rpm on the turbo my Garmin speed would be showing as 16mph. On a flat road in the same gear and cadence, my speed would also be 16mph - it just would require a bit more effort to achieve that cadence if I was going into a slight wind. But on the other hand, if I had a bit of wind behind me it would be less effort to achieve that cadence.0 -
My turbo is an Elite Crono Fluid Elastogel and it has no resistance settings. The resistance is just created by my weight on the rear wheel as it spins on the turbo. I therefore assume the Garmin speeds on my turbo is the same as the speeds you would see if using rollers.
Not really. Rollers really do have no/little resistance. Most fluid trainers increase resistance as your speed gets higher - to simulate some wind resistance. So rollers you can just keep accelerating without limit but your turbo there is a terminal velocity of sorts.The point I was trying to make is that if I was cycling in say, gear 34/15 at a cadence of 90rpm on the turbo my Garmin speed would be showing as 16mph. On a flat road in the same gear and cadence, my speed would also be 16mph - it just would require a bit more effort to achieve that cadence if I was going into a slight wind. But on the other hand, if I had a bit of wind behind me it would be less effort to achieve that cadence.
It may be, it may not be, it depends on the setup of your turbo, there's no reason that it *has* to be the same as you would be doing on the road.0
