Turbulence modeling.

  • Good day, a question for developers. Is the turbulence discretely modeled (i.e. as a 3d flow field, where for example the left wing could be in an updraft and the right wing in a downdraft), or is it simply modeled as a single "gust" event affecting all the parts of the aircraft at the same magnitude?

  • To my knowledge it's a vector field, but I don't know for sure. I've experienced that one wing got an updraft and the other got none or a downdraft, especially near thermals. Turbulence itself I find is bit too simplified, it is equally distributed over time, it can't bank the aircraft to 45deg and just leave it there, at least right now. I find it a bit monotonous at the moment, doesn't really change all that much. But I think once we implement a better weather system the turbulence model will be brought up to a whole new level :D

    - Jan



  • Compared to Aerofly FS1, the turbulence model in AFS2 is quite poor, merely a random and uniform shaking field of the aircraft.
    The AeroflyFS1 wind was actually interacting with the mesh, causing updrafts and downdrafts in moutain flying. The model was perfectible, with very local and laminar influence on the flow (that could even be displayed as a field of vectors). For instance, a strong wind facing a ridge didn't cause any kind of rotor behind.
    But at least there was something.

    When a weather system comes available, enabling linking to real weather engines like Active Sky, turbulence may be modelled by automatically placing local turbulence effects, just like in FSX/P3D...
    The realism of the result depends then mostly on the external real weather engine quality.


    Config : i7 6900K - 20MB currently set at 3.20GHz, Cooling Noctua NH-U14S, Motherboard ASUS Rampage V Extreme U3.1, RAM HyperX Savage Black Edition 16GB DDR4 3000 MHz, Graphic Card Gigabyte GeForce GTX 1080 8GB, Power supply Corsair RM Series 850W, Windows 10 64 bit.