Some considerations about the FM purists...

[Anvilfolk]

I know very little about the current FM accuracy, but I have a question related to the previous post. If you're flying with CEM/full-realism, how do those max speeds really affect flight? I mean, you have to optimally fly your airplane the entire time, and during combat that doesn't necessarily happen. I get the feeling that small differences in speed accuracy would get dissolved in flight due to CEM. Any more expert pilots agree with this?


Another more practical thing that has been mentioned, but that hasn't been given due attention: the assumption that given reliable a reliable source of aircraft performance, you can accurately model it in a combat simulator.

Even if you have reliable sources for aircraft performance (which is an open question), I would reckon it's still a monstrous task to actually simulate it. It should be quite easy to implement performance graphs or matrices or anything (as in FSX, for instance?), but if you do that you don't know how to deal with dynamic damage modelling. So graphs are really just a guideline. I'm guessing the current approach is to have a proper physics engine, control surfaces modelling, and control surface effect on movement depending on different conditions (e.g. weather). Then, as the airplane gets damaged, control surface and airplane properties change (i.e. more friction, less lift, etc), and the physics engine needs to be able to cope with this. Graphs are useless here, no matter how accurate they are.

And modelling an airplane in a physics engine to simulate what real-life performance is, instead of directly modelling real-life performance (according to performance graphs), is completely non-trivial.

I'm just saying that they might want to get real-life performance, but given that damage modelling needs to be in, it becomes an overly complex, heuristic, impossible to really understand process. It's probably guess-and-test, and with the absurd amount of variables..... I don't envy them!

[carguy_]

Originally Posted by Anvilfolk

I'm just saying that they might want to get real-life performance, but given that damage modelling needs to be in, it becomes an overly complex, heuristic, impossible to really understand process. It's probably guess-and-test, and with the absurd amount of variables..... I don't envy them!

The IL2 community agreed long ago that it is not the real life figures that need 100% attention. The trick is to simply recreate the valid differences between the planes. Who cares if the FW190A3 has 2sec slower roll rate if it is still much better than the Spitfire by a good margin? I don`t care if my Bf109 has a 10m/s slower climb rate that in real life. If I still outclimb the Hurricane in a historical manner then I don`t see the problem.