Does the 3D model affect the flight dynamics in DCS? Or is it mathematically simulated? In other words, could you apply the FM of the F15 to a 3D model of a brick and have it fly correctly?
If DCS simulation program do work like in everything else I’ve maintained in the last 15 years, yup. You got it right. No need (and actually littler sense) in doing it in any other way.
I’m curious because I’ve notoced some community mods where it’s just basically a copy paste job of the f15 on the f18/f4/f16 etc
I perhaps explained myself poorly.
There’s no logical (or engineering ) reason to have the Flight model directly related to the 3D model.
It should be (and so far has always been, unless we talk about some minor Indie game) always two separate entities.
What you may need is to have the 3D model to respond to every modification you do to the flight model-
i.e.: if you stop calculating a wing’s lift coefficient and all that shebang 'cause it has been shot off, you’d better remove that wing from the 3D model.
It makes understanding the sudden change in the plane handling characteristics much easier.
Or moving surfaces for what it matters.
EDIT: Virtual Thunderbirds did the same with Lock On. Put an F-16 3D model on the MiG29.
Works mighty fine, unless you change the cockpit too- you can see what I mean here.
PS: Also the undercarriage height is a tad off @1:23.
Of the sims discussed on this site, I believe that X-Plane is the only which does this.
I’d love to see some reference for that, no disrespect meant.
The only game I know of that does that is Kerbal Space Program if you use the FAR mod. It computes all the aerodynamics based on a voxel model that is built from the rocket/plane shape.
Pretty impressive actually, but it costs a lot of computing power, even for one not very complex plane.
Are the calculations proprietary to each engine? For example Prepar3d/fsx/fsw/DCS share some commonality or totally different? I’d like to take a peek at what goes into making a flight model.
It really depends on the Simulated Environment.
We’re really approaching the ugly field of programming and it stops being fun at all.
(Speaking from direct experience, here)
Each game’s simulated environment has its own set of variables used to track real world physics.
So if you’re familiar with the four forces diagram for flight
those are four variables.
If you’re not looking to create a complex simulation cough cough Ace Combat cough
you’re fine with making some tables with static data for each variable, for each plane.
Of course if you know a bit about real life physics you may know that the actual formula for lift it’s not simply a number from a table but rather something like this:
which can be still simplified with a number of tables for each valor but usually complete products have no problem in deal with in realtime- modern processors are pretty hefty and can shoulder the burden soo easily.
Even Falcon 3.0 with a simple math co-processor could deal almost with that in real time.
We can assume then that most of the real physics is probably calculated in very similar manner by all games but still different enough to make stuff very complicated to be shared directly.
Planes are a completely different thing though.
It’s safe to assume that it’s impossible to simply take a plane from a game and “drop it” into another and expect to make it work.
Digression If that was even remotely feasible RAZBAM could have all the money in the world by simply bringing all their products from FSX to DCS… Or just the A-6, which is the only one that makes sense!
Anyways… probably FSX and P3D are the most similar and easy to share stuff between as they essentially have the same common ancestor. It doesn’t mean that’s easy, but simply feasible.
The only case where stuff could be “drag & Drop” and it would work was Combat Flight Simulator 2 and Flight Simulator 2004 both made by the same company in very short time.
You can see where this is going.
I’m not much of an explainer or -Heaven forbid!- a teacher so I might have complicated your ideas here…
But if I haven’t and still have some Q’s- ask. I’ll be here for a while.
This is good stuff, thanks. Assuming you had the correct dimensions, shape, weights, etc., could you not quickly adapt them over? I’d be interested to see what the SDK looks like for some of these sims. Really I’d just like to see how hard it really is to make these things. Not in a arrogant way, but more of a educational manner. Would be cool if one of the Devs would do a stream of the coding process or something. Let us look under the hood.
From the 3D model side, it would appear that the models are a little more plug and play. Build the model in a 3d program and then import it in. Is this correct?
You can visit X-Plane.com to see their basic explanation of how Blade Element Theory is applied in the sim. I haven’t a clue if X-Plane processes extra antennae or stuck gear doors. I doubt it.
X-Plane differentiates itself from other simulators by implementing an aerodynamic model called blade element theory. Traditionally, flight simulators emulate the real-world performance of an aircraft by using empirical data in predefined lookup tables to determine aerodynamic forces such as lift or drag, which vary with differing flight conditions. These simulators sufficiently simulate the flight characteristics of the aircraft, specifically those with known aerodynamic data, but are not useful in design work, and do not predict the performance of aircraft when the actual figures are not available.
Blade element theory improves on this type of simulation by modeling the forces and moments on an aircraft and individually evaluating the parts that constitute it. Blade-element theory and other computational aerodynamic models are often used to compute aerodynamic forces in real time or pre-compute aerodynamic forces of a new design for use in a simulator employing lookup tables.
With blade element theory, a surface (e.g. wing) may be made up of many sections (1 to 4 is typical), and each section is further divided into as many as 10 separate subsections. After that, the lift and drag of each section are calculated, and the resulting effect is applied to the whole aircraft. When this process is applied to each component, the simulated aircraft will fly similar to its real-life counterpart. This approach allows users to design aircraft quickly and easily, as the simulator engine immediately illustrates how an aircraft with a given design might perform in the real world. X-Plane can model fairly complex aircraft designs, including helicopters, rockets, rotorcraft, and tilt-rotor craft.
Needless to say - I’ve spent almost three days flying this…for…well…reasons…!
…plus FighterOps. Still waitin…
To answer the original question. In DCS the 3D model does not affect the FM. 3D model is separate lets call it module to the FM model.
In X-plane the 3D model directly affects the FM. So in XP the C130 just cant fly as F15, what is easily posible in DCS.
Today, afaik, XP is the only sim with this aproach. In the past there was also FighterOps. Combat sim in development which didnt make it to any form of release. So saad…
could one then rip the 3d model from DCS, and then upload it into xplane? Assuming the model is accurate, I would guess it would fly properly in x plane.
My usual line of thought is “If it was that easy, someone would have done it already”.
But nobody claims I’m right, of course.
To give this discussion some perspective, if you’ve ever fiddled with X-Planes plane maker, then you surely know that the flight models in XP that are solely based on the 3d model are, well, not good. Technology just isn’t there yet.
To sobek’s point, Carenado is having a devil of a time figuring out why some aircraft are behaving the opposite of real life in ground effect. I.E. they seem to get sucked toward the ground instead of floating on an air cushion effect. If the behavior was based solely how the 3D model was designed, then all of their aircraft would be affected. Then that brings into play how other objects interact with the FM.
I didnt mentione it, and I need to agree with sobek. The 3D model directly afects the FM however it is not the only contributor.