Flight Model physics improvements


Much left to be done for realistic behavior:

  • wheel friction on crosswind - totally unrealistic at the moment, airplane weathervanes too much and requires way too much rudder when on the ground
  • turbulence behavior and upward/downward drafts and currents
  • adverse yaw effect
  • turboprop drag
  • p-factor and slipstream
  • Inertia. Responsive like an RC plane. Rotates too easily around CG.
  • Ground effect. Still a bit tweaking needed.
  • Wake turbulence !!! - allow us to feel the impact ( and danger!) of entering AI / Multi-player aircraft wake streams
  • Thermals and albedo effect
  • Wake turbulence !!! - allow us to feel the impact ( and danger!) of entering AI / Multi-player aircraft wake streams

You could also mention:

  • Inertia. Responsive like an RC plane. Rotates too easily around CG.
  • Ground effect. Still a bit tweaking needed.

yes! totally forgot about this. Could you add this to your original post?

1 Like

I don’t know if anybody noticed but the aircraft seems unresponsive to wind when on the ground, when placing the cub with 150 kts crosswind its steady as a rock, not even a slight bank or weather vane into the wind. Only when you turn the nose into the wind does it start to lift-off and bounce around. I did notice this with X-wind take-offs as well, there is no need to put the aileron into the wind during take-off or after landing to keep the wing down.


My experience on this topic has been completely opposite, meaning I’ve had to worry a lot about putting ailerons to the wind significantly during take-off and landings. And if I don’t, the aircraft will tilt to the other side, almost resulting in a wing strike unless and until I fix it. However, you mentioned the X-cub which I never tried because I rarely use any of the default aircraft in MSFS, nor do I ever try flights in extreme windy conditions (such as 150 kts crosswind like you tried). I’ve been heavily using only the M20R and PA44 from Carenado these days.

Another thing I noticed yesterday about adverse yaw: while all of the default prop GA aircraft has little to absolutely no adverse yaw, Carenado’s PA44 does have a lot of it.

1 Like

Interesting, I did check the same with other aircraft as well. 150 kts is extreme obviously but even with that nothing happend… I don’t think the MSFS flight model is advanced enough (yet) to calculate the force exerted on the side of the fuselage required to knock it over or at least make the aircraft bank.

Equally I’m not sure the flight model is able to calculate the part of the downwind wing which is being blocked by the fuselage with crosswind or effective chord changing in case of a swept wings. This is what in real life causes the upwind wing to lift with X-wind.You need a pretty advanced flight model (with fluid dynamics?) to be able to take those effects into account. X-plane is doing a pretty good job in this area.

1 Like

in that case I guess MSFS has this advanced flight model in the sim right now already, but just not modelled/added in the default aircraft. I’m installing GE-Force Experience right now to be able to make a video of how if there’s strong crosswind and I don’t make appropriate aileron input in the third party payware Carenado PA44 and M20R, their upwind wing will lift during takeoff or landing and almost cause a wing strike.

This may be what you were getting at, but the effectiveness of the elevator is too much, I pull back on the yoke a centimeter or two and the nose pops up five to ten degrees. It is so difficult to gently lift (or drop) the nose.

I also notice this jumpy movement with the rudder. I cannot smoothly keep the aircraft coordinated. I smoothly apply left or right rudder and I do not see a change, I then apply a little more and the nose bounces five to ten degrees to its respective side.

The yaw is also an issue on the ground at higher speeds, a slight application of rudder pedals and the aircraft is darting for the grass.

I have the sensitivity set way down and do not remember if the roll is also an issue.

IRW the object is for the passengers to not be able to feel the movements.


More than a bit, the 747 for exemple is just gliding over the runway, it could become an ekranoplan.

The A320 FBW mod, is much better in term of landing and ground effect

The flight dynamics on the Boeing 748 and Airbus A32N are working in reverse. The center of pressure is in front of the center of gravity even with the CG on the most forward limit, the center of pressure moves aft with a increase in angle of attack, there is a constant up force on the horizontal stabilizer instead of downforce. All of it is working exactly in reverse, this also means that the aircraft behavior in ground effect is reversed, causing endless floating and ability to fly far below stall speed without exceeding critical angle of attack. Its just a mess :sweat_smile:


Also taxiing on the ground is quite a challenge. Steering is sometimes almost impossible, with huge turn radii

1 Like

More votes please!

I really would like to see them speak about the flight model.

What are their plans for the future and what is their opinion about all that criticism regarding flight model.

1 Like

I know. I’ve been complaining about this since launch. I think they have improved this with the X-Mas patch. But planes are still too responsive. The reason could be control surface authority as well as lack of inertia.

I just hope they will put in the effort to review the physics model fundamentally.

I agree very much about the sensitivity. When flying 3rd party planes like Carenado the experience is much better. Remarkebly, without FSUIPC it wouldn’t be able to keep the plane straight on the runway because the default rudder sensitivity is that horrible.

That’s an interesting side note. I don’t yet own any payware addon planes. But if they’re able to tackle that issue, then at least it’s not globally impossible to get better results.


To quote Aersoft’s Mathijs, whose company produces software for both home desktop and professional, commercial flight simulators, nothing is impossible in MSFS 2020 at the moment, and the SDK now allows you to do whatever you want.

Here’s what he said:

As explained in another topic, the SDK is not really the bottleneck at this moment. The SDK will catch up but right now goal is to get the code we have to run smooth and reliable.

Also as explained in another topic, getting C++ code to run in the sim is not really a problem. You can make a electrical or pneumatic system system as complex or deep as possible. Beyond what any P3D add-ons has now if you so require. It’s just code. Of course there will be problems as variables from the new sim do not give the same value as on P3D, but you ask Asobo for help and you will get is as long as you are in the correct cycle. You might need to wait for the next update of the sim to get it running on the sim customers have though. But as I said, when needed Asobo/MS will update when needed.

I will give you an example. We have a professional project intended to train people. It is about the electrical system of an aircraft that shall not be named. Very serious stuff. I mean it models the reduction of load when a footwell heater fails. Pure C++ code and and while the output to the displays needs to be redone, it all works without any issue in MFS. 100% the same code as in P3D, just needs new variables in input and output. it can’t be used in MFS as you can’t get professional licenses but if the customer wants proof the code runs we are happy to show it.



Weather radar does not seem to be possible to implement?

Partly true and partly untrue. This is what Mathijs said when asked the question “Will there be a working weather radar that is accurate unlike default aircraft?” :

At this moment not possible. But the data is there and as soon as get the information it will be added.


Keep in mind that there’s a huge controversy as to whether the default MSFS weather is accurate or not. Example: the areas of the world in MSFS where I mostly fly, weather is very accurate. But for others in other areas, not so accurate.