Ant+ simulators & speed tests

Maybe you want to run some experiments on Zwift. That’s when you’ll need a way to automagically submit power and cadence to the system. Such a device is often known as ANT+ simulator. I’ll not enter in details of the actual set-up unless requested.

You also would hopefully clear the field of drafting effect. Ride a TT bike or enter workout mode or both. Anyway, else in workout mode on a given route, there is no guarantee your avatar will start precisely on the same point. After a lot of mumbling, IMHO, the easiest way is to take screenshots on know points, like a finish line, and manually calculate times. A Strava segment may be an alternative, but then you’re going to skew up your profile adding rides generated by the simulator and NOT by you.

Sure I’m aware of Zwift Insider amazing work on the field. But when I need something directly related to my figure or my equipment or a preferred route? I hope this thread will be here to register data from your tests.

To start the thread, I’m going to paste some messages from an older topic.

Cu. Paolo.





I did this even t at 12:30 on 4th April. I loved the TT format but I’m totally confused by the Zwift physics/algorithm!

For my event I rode a steady threshold pace and average 325W, I came 5th overall in a time of 1:13:05 I weigh 88kg so my average power/weight was 3.7W/kg. First place averaged 301W in a time of 1:09:48 over 3 minutes quicker with an average power to weight of 4.3W/kg. For a pan-flat course brute power should outweigh power/weight like it does IRL. A friend of mine observed the same thing, he averaged 333W and finished in a time of 1:08:34 yet another rider in his event finished with a similar time at an average of 230W - There’s something wrong here has anyone else noticed this or is it just me?

1 Like



  • It’s not pancake flat.
  • It’s not necessarily the greatest W/kg that wins.
  • It’s not necessarily the greatest raw wattage that wins.


Gerrie Delport TeamODZGerrie_Delport_TeamZ


Height appears to be a factor?

Height should be the main factor on a TT flat race, height is used to calculate your Cda (frontal drag area)



For a pan-flat course brute power should outweigh power/weight like it does IRL.

It’s pretty much w/kg on Zwift.

Well, that does explain quite a lot…

As a heavier rider, I fully expect to struggle on the climbs. But I also expected to have some advantage on the flats, and that never did seem to be the case.

So, how much do I need to weight dope to level the playing field against a 65 kg rider on Tempus Fugit…?



Control: 170 cm, 70 kg, 250 watts
Test 1: 170 cm, 85 kg, 250 watts (What’s the time difference?)
Test 2: 170 cm, 85 kg, 263 watts (Does that equalize the times?)
Test 3: 175 cm, 85 kg, 266 watts (Add 5 cm, is the time the same?)

Lead in Finish Line Lap Time
Control: 170 cm, 70 kg, 250 watts 03:44,0 29:18,0 25:34,0
Control: 170 cm, 70 kg, 250 watts 03:44,0 29:19,0 25:35,0
Control: 170 cm, 70 kg, 250 watts 03:41,0 29:14,0 25:33,0
Control: 170 cm, 70 kg, 250 watts 29:14,0 54:49,0 25:35,0
Test 1: 170 cm, 85 kg, 250 watts 03:58,0 30:40,0 26:42,0
Test 1: 170 cm, 85 kg, 250 watts 03:54,0 30:34,0 26:40,0
Test 1: 170 cm, 85 kg, 250 watts 30:34,0 57:13,0 26:39,0
Test 2: 170 cm, 85 kg, 263 watts 03:59,0 30:07,0 26:08,0
Test 2: 170 cm, 85 kg, 263 watts 03:59,0 30:07,0 26:08,0
Test 2: 170 cm, 85 kg, 263 watts 30:07,0 56:15,0 26:08,0
Test 3: 175 cm, 85 kg, 266 watts 03:51,0 30:03,0 26:12,0
Test 3: 175 cm, 85 kg, 266 watts 03:45,0 29:57,0 26:12,0
Test 3: 175 cm, 85 kg, 266 watts 29:57,0 56:09,0 26:12,0

Here we’re. Ride on! Paolo

1 Like




Clearly Zwift is not using the Cycling Power Lab formula. We should probably also consider rolling resistance, but going from 70 to 85 kg, rolling resistance should only increase by about 2%, so that won’t account for the difference.

If I have some time I’ll try to fit your data to some equations and calculate the power increase (according to Zwift’s physics) that would be required for a 175 cm 85 kg rider to keep up with a 170 cm 70 kg rider.

And the answer is …

No, not 42. The answer looks like 284w

Test 5: 175 cm, 85 kg, 284 watts 03:53,0 29:27,0 25:34,0
Test 5: 175 cm, 85 kg, 284 watts 29:27,0 55:00,0 25:33,0

I guess you’ll going to land on nowhere with Cycling Power Lab formula. That is for a static answer while Zwift needs to calculate a dynamic one taking in the count a not perfectly flat route and your momentum.


1 Like

You’re right about that.

I used equations to include aerodynamic drag and rolling resistance in a steady state. I neglected the 16 m elevation gain at 1% since I figure that makes a 0.01 watt difference average over the 16 km course. I added 10 kg for the mass of the bike. This is a simple analysis - nothing more than a reality check.

Using the Cycling Power Labs formula and classical physics equations produces significantly different numbers (higher across the board) than found from your simulations. It’s not even possible to fit all of the data points by tweaking Cd, so Zwift is using a different formula to compute frontal area.

Starting with the 170 cm, 70 kg rider, 25:34, 250 watts, and adjusting the Cd (to a rather silly number) to fit the equations to that point, the equations give 276 watts for a 175 cm, 85 kg rider holding the same speed. So, 26 watts more, or 10.4%.

Your simulation gives 284 watts, so 34 watts more, or 13.6%.

Does that seem right? 13.6% more power for an 19% increase in mass and 3% increase in height? Seems like a lot to me.

Does anybody have access to a wind tunnel, preferably one with a rolling road…?

1 Like

Hi I don’t know if this thread is still active but I have recently downloaded an ANT+ simulator from the thisisant web site and I’m struggling to get it to communicate to ZWIFT. I want to simulate power however I have absolutely no idea how to use it


what do you want to do with a simulator?

I would steer clear of using simulators.

1 Like

I noticed this as well over the years no change!
Instead people suck it. I’m always wondering why w/kg in the flat???
Its just total nonsense!
Did you ever saw a w/kg monster winning Paris-Roubaix?
No in the flat (indoor-outdoor) the heavy guys are outclassing the light-weights.
You can even call this a bug.

Hi Charlie.

You need two ANT+ dongles, one to transmit and one to receive. A single computer will be enough, but with two would be more comfortable, one for Zwift and one for the ANT software.

Connect the dongle(s) and fire up Simulant. Bind just ONE dongle to Simulant. Of course, the second must be free to bind to Zwift.
This is, if you’re using just ONE computer, you MUST RELEASE one of the ANT devices from SimulANT+ to allow Zwift to see that second one.

Now go to Add Simulator and select Bike Trainer (FE-C). This is not the only chance, but for the sake of simplicity, select this.
Go to Device Simulation Configuration and set cadence and instantaneous power. And click on Actions:TurnOn.

Zwift should now be able to pair. Enjoy.


I know (@Gerrie, I’m aware, OK ??) that this may be used to cheat. There are obviously a ton of ways to cheat within Zwift, and this is not the easier or cheapest, IMHO. I’ve been a top gun as a runner in my day. To be brutal, an actual athlete will never mix up real-life results with a game like Zwift. Add, if Zwift would care of, there are obviously strategies to mitigate the cheating, but this is up to Zwift, not me.

Cu. Paolo


Just before we all go any further with this topic, I’ve been asked to make sure y’all know a little more about our End User License Agreement.

Prohibited Conduct. (i) cheat or use, develop or distribute automation software programs (“bots”), “macro” software programs or other “cheat utility” software programs or applications that are designed to modify the Product experience to the detriment of fair play;

Of course, we want the forums to be a place for open discussions, but it may not be the best idea to inform others how to bot (simulate) the game. Please keep that in mind and thank you for your understanding!


Mike M.
Zwift Forums Lead