No tricks... just the right data...

[jcomm 5,890]

Probably one of the two things that attracts me more in Aerofly ( the first being that somehow it reminds me a lot of Flight Unlimited 3 ) is that the flight dynamics modeling used in this sim allows to plausibly model the behaviour of an aircraft, being it a B747-400 or and Extra 300, with their associated inertias, the Extra being able to stop an aileron roll in a fraction of a second just like it's real counterpart and the 747 showing the realism I can't find in other 747s regarding it's inertia but also it's pitching moments due to thrust changes, and all of it is done solely by pooring the right data into the aircraft configuration files.

 

No fancy gauges programming, puffers, scripts, exteranl FDM or whatsoever are required.

 

And, I think we must always point this out, it's an "early pre-release", just the concept, of what I believe can easily become a powerful civil flight simulation platform.

[Murmur 3,941]

There's this myth that you put the exact geometry and data and a certain flight simulator produces a 100% exact flight model, but this is not true, neither be it X-Plane, DCS, or AeroflyFS.

 

Every one of them requires certain types of "tricks", be it empirical coefficient modifications, hidden wings, or any other correction.

 

Of course, the difference is how easy it is to tune the flight model to match it with the real one, in terms of basic (performance and stability and control derivatives) and secondary (stall/spins, propeller effects, etc.) flight dynamics.

 

I had a look at the AeroflyFS 2 documentation on flight models, and although it's not 100% complete, I found it very interesting. Conceptually, it looks similar to the X-Plane approach, with the difference that it allows a more granular and direct control over certain aspects: effective aspect ratio, downwash, ground effect, propwash, flight controls effectiveness, aerodynamic coefficients of fuselage and other bodies, etc.

 

This is the approach that I always advocated, because it seems to have the advantages of both the predictive flight models (blade element) and the descriptive ones (table lookup). However, at the moment the documentation seems incomplete.

 

That being said, I noticed at least two things lacking:

 

1) I could find no mention of Mach effects. So apparently it's limited to incompressible (low Mach) flight regimes?

 

2) I think aeroelastic effects are not included (this is also true for X-Plane). However, some (not all) of them could be modeled "tweaking" flight control effectiveness.

 

Maybe these two issues are due to the fact that AeroflyFS 2 derives from a RC simulator, where mach and aeroelastic effects can be disregarded.

 

So the only big thing lacking for it to be considered a good general purpose flight model, is the Mach effects.

[jcomm 5,890]

And, in it's approach it resembles of lot some aspects of both FU2 and 3 and Fly!.

 

As you put it, a mix of table lookup and multiple lift/drag/thrust generation bodies assembled to form a single entity.

 

Regarding your observations on "Mach-range" effects, now it's the time to ask at their support forums, since they're at it and really want to make the platform grow in the right way.