The Flight Model Thread

[maradine]

How I've longed to bring this to you! It was cold out there!

Copypasta'd from my work on the beta forums, for your PSU pleasure. Discuss!


I'm serious about PS2 flight. I want the model and control scheme to be deep enough to satisfy the study-sim nerds, but accessible enough to please the PS1 "floating camera" hipsters. I recognize that never in the twain shall meet, but every step either extreme makes into the middle is useful.

In order to have this discussion, we first need to be able to objectively state the characteristics of various parts of the engine, and the various actors within it. Specifically, since we have asymmetric balance, it is useful to know exactly how each aircraft behaves relative to the others.

Presented below is my first pass at gathering that information. Some definitions:

Soft "Stall" Speed - the absolute minimum speed the aircraft can be brought to via tiny applications of the deceleration key and still maintain level flight.

Hard "Stall" Speed - the absolute minimum speed the aircraft can be brought to via hard braking and still maintain level flight.

I was surprised to learn that these are repeatably different speeds. Yes - you can fly all three fighters at less than a walk if you're gentle about it.

Rotation Rates - these were all measured by timing multiple 360º rotations and averaging out. Shockingly, all the fighters rotate exactly the same, both idle and at speed. The Reaver, for instance, will turn with either competitor degree-for-degree.

Translations - how fast the craft gains or loses altitude in level flight via application of the "space" and "ctrl" keys, both from an idle and at speed.

Idle Vertical Drift - how fast a craft below its stall speed loses altitude.

Obviously, the Afterburner speeds will be subject to module loadout, but it's interesting to see who has the better beta module hard-coded right now.

So - the numbers:


Note: Following image is 2 patches old - will be updating tonight.



Where to go from here? You can help me by -

• Validate! Confirm my numbers aren't ********.
• Collect! Is there something that the different craft do very differently? Get me numbers and a method for reproducing them.
• Suggest! How do we make each fighter handle differently but fairly?
• Test! Just fly the **** things and say what you see. Use different input modes. Use a gamepad. Wire up a HOTAS. Whatever.

Enhancement Suggestion #1
Implement Slip
Slip is the tendency of an aircraft to yaw into its lower wing during a roll-initiated uncoordinated turn. Slip is a desirable aspect in a flight model as it allows small heading (and therefore, aiming) corrections to be made without engaging the rudder. The slip force should be small compared to the yaw generated by rudder engagement. Slip should only be applied to aircraft that actually have some form of aerodynamic surfaces, but applying it equally is probably simplest.

Implementing slip cheaply in PS2:

1. Calculate absolute roll attitude "X", where level flight is 0 and wings aligned vertically is 90. (this is already performed and displayed).
2. Calculate speed ratio "Y" where Y is current vehicle speed divided by rated maximum.
3. Tune with constant "C". where larger C induces more slip in identical flight regimes.
4. Calculate: Slip Force "S" = X*Y*S
5. Apply S as an additive yaw moment (over and on top of whatever control inputs are being applied)

Examples:

A Mosquito moving in level flight (X=0) at maximum rated speed (Y=1) has
no slip (S=0), regardless of tuning (C=?).  Note that this is also true in inverted level flight.

A Mosquito moving at a 30 degree bank (X=30) at maximum rated speed (Y=1)
with the tuning constant at 0.1 (C=0.1)has a slip of 3 degrees per second into the lower wing (S=3).

A Mosquito moving at a 90 degree bank (X=90) at half its rated speed (Y=.5)
with the tuning constant at 0.1 (C=0.1) has a slip of 4.5 degrees per second into the lower wing
(S=3).

A Mosquito at a 90 degree bank (X=90) and motionless (Y=0) withe the tuning
constant at 0.1 (C=0.1) has no slip (S=0).  How it stays up at all in this attitude
is a completely separate topic.

[Hamma]

Wow.. well thought out post.

[VaderShake]

Hmm.....looks like some math is involved.....count me out!

[julfo]

Very interested to see the updated numbers. Suspect they will remain mostly unchanged.

[maradine]

That is my suspicion as well. If I had to guess, they are going to make the remainder of the handling identical and cert. out the differences. I guess we'll know soon enough, but it would be nice to have some developer insight into where they're going with it.

[Kipper]

Poor frame rate prevents me from doing much of anything, especially flying - but I do plan to do a fair bit of it.

I agree slip needs to be implemented, it will make things feel more realistic and it's not newbie unfriendly since its quite a small effect. Of course they should support analog throttle and rudder controls so that you can use pedals and do coordinated turns.

I'd like to see a transition between hover and forward flight modes, say on 25% throttle or something. If you want to hover, fine, but if you want to go fast, you should not be allowed to stop in the air quite so seamlessly.

Stalling in a steep climb can be a danger, but it can also be a manoeuvre (or just fun if you can do some aerobatics)

[maradine]

I think it's important to note, for those that are just joining us, that axis mapping is not yet implemented.

The flight controls or non-optimal for mouse users, and non-functional for joystick users.

[Xyntech]

I wonder how viable it would be to have the mouse control yaw at slow speeds and smoothly transition to controlling roll as the aircrafts speed increases. Only available as one optional control scheme of course, with key bindings available to roll and yaw as well.

The slip mechanic would definitely be a welcome addition. I'd also like to see the hard flight ceiling turned into a soft flight ceiling where aircraft stall out past a certain height.

[maradine]

As opposed to the cement bounce we currently have?

Actually, another 500 meters and I think I'd be pretty happy.

[Hamma]

I still think the Mosquito needs to be a bit faster to go along with the empire theme.

[Rolfski]

Are these nrs still correct?

[Ghoest9]

Back in beta a seemingly real pilot came to similar conclusions just based on the way the various aircraft handled and how it felt with respect to real jets and helicopters.

They didnt listen to him.

All I know is that the modele feels nothing like any aircraft i have been in or video I have seen shot from inside an aircraft.

[Hamma]

Flight model has changed quite a bit since early beta. Personally I love flying, I don't expect a flight simulator.

[Bravix]

Originally Posted by Ghoest9

Back in beta a seemingly real pilot came to similar conclusions just based on the way the various aircraft handled and how it felt with respect to real jets and helicopters.

They didnt listen to him.

All I know is that the modele feels nothing like any aircraft i have been in or video I have seen shot from inside an aircraft.

That's probably because there aren't any aircraft like these in real life

I really don't have any major gripes with the flight model. I wish ascending was slower and descending was faster (gravity is a bitch) but there isn't a whole lot I'd change.

It's important to keep in mind that, in real life, these aircraft would never fly. Except maybe the Scythe, cause it uses funky space-magic. Mosquito and Reaver don't, however. They supposedly just use aerodynamics....which is bull because, again, these things wouldn't fly. Not in the way the game presents them at least. There would always need to be a downward thrust to keep these things up, because those pitiful excuses for wings certainly wouldn't create enough lift to hold the aircraft up.

[RSphil]

interesting. my main problem though is still no proper joystick support. nice to see other sim nerds about. have to keep it simple though for all those out there who cant fly that well. i like the way things are now bar the lack of stick support.