So, first off I know this has been discussed in many threads (yup, I used the Search bar) so I'm sorry if it sounds repetitive. I felt that this specific line of conversation deserved its own thread, and much of the info I came across was 6mos-18mos out of date, so although the questions may have been answered, the answers may not remain pertinent enough to warrant some insane necro-thread revival. That being said, I have a few questions about BeamNG's implementation of air resistance for the developers or anyone with advanced understanding of the .jbeam structure therefore complete knowledge of its variables. - Estama mentioned in a thread awhile back that a wind tunnel was in development for testing of aerodynamics and of downforce/lift in a controlled environment. Wondering if that is still a plan. - Secondly, can anyone please (I apologize for potentially coming across as needy and ignorant) explain to me how exactly wind resistance/downforce is implemented in the .jbeam? I have read in other threads that creating an inclined plane of nodes creates lift by presenting a drag surface, and that a negative incline presents an inverted drag surface and therefore creates downforce, truly mimicking reality. However, when taking apart the .jbeams from the stock vehicles, I've noticed that the tags for friction and drag appear in multiple .jbeam files for multiple parts, as well as the master .jbeam for the vehicle. Does the game take the master value as gospel or can that be over-ridden for a specific part? i.e. the body of the car is aerodynamically transparent (think Lotus Esprit) but a variety of separate rear wings/front splitters can be configured as interchangeable parts with varying degrees of drag implemented on each (race cars are often "tuned to the track" with wing/splitter adjustments offering different degrees of front and rear downforce). It would be phenomenally easier to clone parts and change the drag variable than to re-model different spoilers with different degrees of inclination/declination, which would have to be tested, remodeled, changed, etc. I know I've popped up hither and yon on the forum, but I've tried to stay back from asking a multitude of questions with regard to content creation. I've learned how to model, UV and texture from reading this and other forums, although modeling is still my weak point; I seem to need to choose between polygon density and outward quality and that still bothers me. I'm also picking up the .jbeam structure by reading and dissecting the stock vehicles, but the wiki is quite lacking in the more advanced functions. Thanks in advance for letting me borrow your brains! Any help you can offer is sure to speed up development of my first mod!
Downforce, drag, and other aero effects are determined by collision triangles, which are defined in the "triangles" section of a jbeam file. Since n/bs are usually very low resolution when compared to the flexbodies (the visible mesh), aero fidelity is quite coarse and requires some dragCoef tuning to attain realistic behaviour for different parts, especially those concerning ground effects. A dragCoef value found in a jbeam will affect all triangles within the slot it defines, so another dragCoef value would have to be assigned for the next slot. (from what I remember huehue) Poke around in high quality n/bs for an idea on the numbers you should use; the dragCoef for an entire official unibody is usually <10, and if you have the Zakspeed Capri, you'll find that the wings and diffuser would have some mental number for downforce. One more thing to mention is that collision triangles ignore the 'airstream' from other triangles, so each triangle is aerodynamically invisible to others. This means triangles at the rear of a vehicle are affected by air just as much as the triangles at the very front. edit: frictionCoef determines the amount of friction a node experiences on contact with the ground and other environment objects. It has nothing to do with aerodynamics.
