Fixed Wing Flight Model (dev branch)

[x3kj 1247]

it was in 2 tickets on the old feedback tracker related to plane handling iirc and possibly also in this thread, if you want to search for it.

 

it was a band-aid back then (A2) and they removed it because it's not physically correct to use autorudder to fix the flight model.  But there was no improvement ot the flight model following that, which brings us to the current situation of an unrealistic flightmodel and also worse handling then before.

[John Spartan 89]

if I may add to this....
so from what I understand as an issue described with keyboard/mouse controls about being sluggish and incorrect. Till certain extent its correct but let's not forget that on same forums people previously asked for this change.... "please disable auto rudder", what triggered current behaviour [compare vs A2]

so lets look at the FM in overall:
- first of all this is not DCS/WT/FSX/SF2 or BF2 (thank good), so we don't want it to be too realistic but we want fun, engaging and authentic
- auto throttle and auto rudder systems that so many complained about was actually good part of this FM as long as properly configured and also as mentioned correctly is authentic - look up FlyByWire system behaviour
- lack of detail and attention to certain FM configurations in vanilla & mods is other problem that causes misinterpretation. A164 initial was "ugly" FM wise but now tweaked it's actually fun to fly - at least for me [KB/Mouse], I have made few FM's myself and only one is actually decent others need more work, it's not that easy and takes time but company can improve on that agree. 

More emphasis on sub-systems like targeting and weapons would be more appropriate point of focus than FM but at same time there are few things that can easily fix what you lads are complaining about:
- auto rudder must be brought back in but with revision and it has to be optional on config level, so based on plane model you can configure the required behaviour. 
- auto throttle needs to be a bit clearer interpretation of actual trust applied, more visual output in UI to allow pilot to have better control over the situation/aircraft
- wheels and suspension needs some attention
- and most important is documentation that explains the FM so community can utilize it to the best of it's current state

just my thought here...




 

[K20017 36]

if I may add to this....

so from what I understand as an issue described with keyboard/mouse controls about being sluggish and incorrect. Till certain extent its correct but let's not forget that on same forums people previously asked for this change.... "please disable auto rudder", what triggered current behaviour [compare vs A2]

I have flown many a times on Arma 2 but don't recall the specifics of the auto rudder.  Was it that when you held in full rudder, the aircraft would initially start to yaw fully but then correct itself even if the key was still pressed?

[John Spartan 89]

@K20017 take another test flight in A2, keep an eye on rudder anims and mouse of plane during turns, you cant miss that...
It's there, but the problem it feels a bit overdone, even thou it seamlessly trays to compliment the FM in that version of Arma.

[K20017 36]

@K20017 take another test flight in A2, keep an eye on rudder anims and mouse of plane during turns, you cant miss that...

It's there, but the problem it feels a bit overdone, even thou it seamlessly trays to compliment the FM in that version of Arma.

 

Just did a flight and certainly noticed it now.  But what exactly is this autorudder preventing people from doing?  Is it limiting their control?

[John Spartan 89]

@K20017 - in most of modern jets/planes the FlyByWire system will automatically autocorrect for pilots input mistakes or adjust to reduce workload for pilot. You will see various representations of this feature across all game platforms. Main purpose in this case is to keep planes nose in level/aimed at target automatically - just like in RL F-15 for example. Trouble there is - it's a bit to "over cooked" apart from that if revised/rebuilt it's what you lads were asking to be brought back - to resolve fly with KB/Mouse fun/simplicity. Nothing wrong with it as long as it's optional and configurable to allow flexibility for people creating aircrafts.

[twistking 204]

i can't udnerstand that noone cares about the lack of incremental throttle control / manual throttle for mouse/keyboard-users.

 

this is not ment to be a rethorical question: is there a workaround to have incremental throttle control, because this really frustrates me and is the biggest reason i dislike flying with mouse+keyboard.

this would also be very, very easy to implement. it's just mapping of to keys and deactivation of autothrottle...

[K20017 36]

@K20017 - in most of modern jets/planes the FlyByWire system will automatically autocorrect for pilots input mistakes or adjust to reduce workload for pilot. You will see various representations of this feature across all game platforms. Main purpose in this case is to keep planes nose in level/aimed at target automatically - just like in RL F-15 for example. Trouble there is - it's a bit to "over cooked" apart from that if revised/rebuilt it's what you lads were asking to be brought back - to resolve fly with KB/Mouse fun/simplicity. Nothing wrong with it as long as it's optional and configurable to allow flexibility for people creating aircrafts.

 

I see what you mean.  The issue I see with Arma 3 and its rudder problem is that it seems the vertical stabilizer has no effect on aerodynamics.  The aircraft flies crooked all the time when in real life, the vertical stab would keep the aircraft's longitudinal axis in line with its direction of flight (in a no wind scenario).  If the developers have no desire to change the aerodynamics, autorudder seems like a decent bandage to cover the wound.  

[CaptainAzimuth 714]

i can't udnerstand that noone cares about the lack of incremental throttle control / manual throttle for mouse/keyboard-users.

 

this is not ment to be a rethorical question: is there a workaround to have incremental throttle control, because this really frustrates me and is the biggest reason i dislike flying with mouse+keyboard.

this would also be very, very easy to implement. it's just mapping of to keys and deactivation of autothrottle...

I brought this up too, using the example of the Heli-AFM. You can rise thrust until the heli raises up. So why not add the same for planes? You can raise thrust till you start moving, and thus enabling you to taxi smoother. This is even a thing in YSFlight, i might do another short vid to show. But these two suggestions, Incremental Thrust and Mouse as Joystick features would drastically improve fixed wing game-play. Should be patient while they finish up Apex so maybe they'll give us there thoughts on this when they finally get un-chained from their desks by the Project Lead. xP

 

it was in 2 tickets on the old feedback tracker related to plane handling iirc and possibly also in this thread, if you want to search for it.

 

it was a band-aid back then (A2) and they removed it because it's not physically correct to use autorudder to fix the flight model.  But there was no improvement ot the flight model following that, which brings us to the current situation of an unrealistic flightmodel and also worse handling then before.

It was a band-aid for Arma 3. Auto Rudder performed, and still does perform, very well in Arma 2, as much as i find it pleasing still to fly Arma 2's aircraft over Arma 3's. There's something... odd, about Arma 3's fixed wing, but i simply can't put a finger on it. But having no auto rudder has now become a pain in the arse. -,-#

[John Spartan 89]

- so we all agree that REVISED auto rudder would be small solution towards more enjoyable FM
- throttle for keyboard needs to function similar as analogue and also display the % of thrust applied in UI

relatively small fixes but that would be more suitable for ARMA...

 

[CaptainAzimuth 714]

-revised Auto Rudder

-Incremental Thrust (Like AFM Cyclic)

-Mouse as Joystick (Can be Optional, because some might actually like the current mouse as rudder instead)

[x3kj 1247]

-revised Auto Rudder

-Incremental Thrust (Like AFM Cyclic)

-Mouse as Joystick (Can be Optional, because some might actually like the current mouse as rudder instead)

i only agree to the first 2 points, because the third - as discussed - does not change anything whatsoever. It changes the methodology to controll the 2 axis you can use with your mouse (position driven instead of velocity driven).

You can rebind rolling to the mouse instead of rudder if you hadn't noticed... I wasn't even aware that rudder on mouse is standard (is it?). I've always had elevator + aileron on mouse axis and rudder as keys since release without changing anything iirc.

 

 

 

- so we all agree that REVISED auto rudder would be small solution towards more enjoyable FM

I don't agree that it improves the FM. It mitigates the bad FM to some extend (improved handling but no improved FM). So until we can get an improved FM, it's better to have optional autorudder then not having it. The option for that should be in user controlls however , and not per plane config - because it's user preference for the most part (proven by the point that some people complained about it i think?).

 

 

first of all this is not DCS/WT/FSX/SF2 or BF2 (thank good), so we don't want it to be too realistic but we want fun, engaging and authentic

An authentic flightmodel does not need to be as complex as DCS (which has wind, center of gravity changes due to fluids, etc) or WT simulator mode (includes thermodynamics and other stuff) , nor would it have to be more complex to controll.

So yes, DCS level is not expected or wanted. But improved FM is necessary nontheless - irregardless if autorudder or not.

 

 

- wheels and suspension needs some attention

- and most important is documentation that explains the FM so community can utilize it to the best of it's current state

Why exactly wheel and suspension? This would likely be the most work for little benefit. The only time you see suspension is when you touch down on landing, that's it. We don't have IK animation either, which is required for proper suspension animation due to nonlinearity of suspension systems. Also, the FM is documented well enough imo (aircraft sample). It's just so arbitrary and not based on realworld physics that you can only trial&error the values until it somewhat fits - there is no realworld data you could use to rule-of--thumb find a startingvalue. It won't get any better then it is now.

[CaptainAzimuth 714]

i only agree to the first 2 points, because the third - as discussed - does not change anything whatsoever. It changes the methodology to controll the 2 axis you can use with your mouse (position driven instead of velocity driven).

You can rebind rolling to the mouse instead of rudder if you hadn't noticed... I wasn't even aware that rudder on mouse is standard (is it?). I've always had elevator + aileron on mouse axis and rudder as keys since release without changing anything iirc.

 

Completely wrong. You're simply not understand that last point, so let me put it in the absolute most simple terms. Currently, When you move the mouse in Arma, it effects your aircraft in the most minial possible way. In other words, there's no point in using the mouse, because the "WASD" keys give you more control. HOWEVER, the "WASD" controls while allowing you to control the aircraft, are still only an On or Off effect. You're either not moving at all, or fully using that control surface. There's not incremental movement, no precise flying. Simply, hard turns and rolls, ONLY.

 

What Mouse for Joystick does, is replace the current problem(s) with mentioned above. I opened Editor and took off in a jet, and moved my mouse in a circle. It made the plane move in a circular strafing motion about the size of a quarter. With my suggestion, once again, as if not thoroughly explained, it would allow MUCH more control of the aircraft, that of which CANNOT be achieved with current mouse or keyboard controls, because of the way it's implemented. In other words, you move the mouse once, and the plane just slightly moves, and stops. It doesn't keep going. As you said, it's velocity driven, which is terrible. With positional, it allows you do put as much turn into your aircraft as you need, and then you simply move the mouse back when you're lining up. It improves flying, it improves attacking, it improves accuracy, and overall, would improve the terrible performance jets currently have, visually, and in functionality.

 

If i need to make a video to prove it in detail, just say so, i don't mind.

[x3kj 1247]

 

Currently, When you move the mouse in Arma, it effects your aircraft in the most minial possible way.

then how come i can fly with mouse just fine (with the limitations of having rudder only as keybind and no autorudder)? That issue is on you... Adjust sensitivity if you can't controll it, or use acceleration. I have zero issues with 50% sensitivity ingame and in windows setting without any acceleration.

 

That keyboard only flying is choppy wasn't even part of the debate. Saying your preferred controntroll method is better then current mouse aim, by comparing it with keyboard-only-controll is not a valid argument.

 

As you said, it's velocity driven, which is terrible.

No it's not terrible. It's a matter of preference. If i want to stop rotating i stop moving the mouse. If i want to rotate faster, i move the mouse faster. Simple as that. Whereas with positional controll i have to look for the current position of the mousepointer in relation to the arbitrary controll boundary and then move it to the center, to stop rotating.

[Pyronick 21]

Why exactly wheel and suspension? This would likely be the most work for little benefit. The only time you see suspension is when you touch down on landing, that's it. We don't have IK animation either, which is required for proper suspension animation due to nonlinearity of suspension systems. Also, the FM is documented well enough imo (aircraft sample). It's just so arbitrary and not based on realworld physics that you can only trial&error the values until it somewhat fits - there is no realworld data you could use to rule-of--thumb find a startingvalue. It won't get any better then it is now.

Well, isn't the landing bit one of the dodgiest things in Arma 3?

[x3kj 1247]

 

Well, isn't the landing bit one of the dodgiest things in Arma 3?

Dodgiest? Maybe on custom carriers where it likes to spazz out, but not on normal ground... Compared to the deficiencies in the simulation of flying, and given that landing represents 1% of the time using the plane at best, the impact is miniscule. Arma in general is very unforgiving when it comes to hard landings, but this is more due to how the impact damage is coded (as it equally applies to helicopters). In a perfect world i'm all for improved landing gear and so on (DCS has that wonderfull bouncy feel to it when landing/starting/taxiing). But this isn't a perfect world, ressources are limited. So i suggest improvement of the things that actually have the most impact rather then sidegrades or little details that have little impact.

[CaptainAzimuth 714]

then how come i can fly with mouse just fine (with the limitations of having rudder only as keybind and no autorudder)? That issue is on you... Adjust sensitivity if you can't controll it, or use acceleration. I have zero issues with 50% sensitivity ingame and in windows setting without any acceleration.

 

That keyboard only flying is choppy wasn't even part of the debate. Saying your preferred controntroll method is better then current mouse aim, by comparing it with keyboard-only-controll is not a valid argument.

No it's not terrible. It's a matter of preference. If i want to stop rotating i stop moving the mouse. If i want to rotate faster, i move the mouse faster. Simple as that. Whereas with positional controll i have to look for the current position of the mousepointer in relation to the arbitrary controll boundary and then move it to the center, to stop rotating.

It's not a sensitivity thing, or a control thing both of which i can manage more than easily. It's the fact that it's not a constant thing. You can move your mouse, but the plane only moves for a couple seconds, and then stops. This the problem. Sensitivity won't fix this, as it will only make it move more, and stop. It should be a constant thing, hence my original example with the Dot in the Box method.

[K20017 36]

The ground handling too needs to be fixed.  You'll get great handling when you are slow but once you speed up to a certain point (60 km/h or so), you have almost zero directional control.  Then as you speed up more, you'll begin to regain some and you can try and stay on the runway.  There shouldn't be a zone you transition with terrible control, the effectiveness of the rudder should be increased much earlier.  But then again, any fixed wing aircraft doesn't feel like it is interacting with any air :(

[dr. hladik 231]

From what I've heard the Fast jet flight model is just a heavily modified helicopter flight model, which if thats the case is a real shame. I'm as passionate about fast jets as the next bloke is however I do agree there needs to be some sort of update for the flight model.

 

 

Not true.

[xxgetbuck123 945]

Not true.

 

Well that settles that then.

[VTOLAFMPLS:3 36]

FixedWingLib supports VTOL  (83)

 

 

http://www.rtdynamics.com/v3/products-services/simulation-and-training/fixedwinglib-cgf-for-c/

 

http://www.rtdynamics.com/v3/products-services/simulation-and-training/fixedwinglib-cgf-for-c/features/

[xxgetbuck123 945]

FixedWingLib supports VTOL  (83)

 

 

http://www.rtdynamics.com/v3/products-services/simulation-and-training/fixedwinglib-cgf-for-c/

 

http://www.rtdynamics.com/v3/products-services/simulation-and-training/fixedwinglib-cgf-for-c/features/

 

Just what we need for Arma! Looks great, never heard of the stuff before but either way I like it

 

Edit: Says this stuff is for C++, does that mean it could work using that Arma 3 scripting C++ thing/thread that has been going around? 

[oukej 2910]

For Apex owners - this would be also the place to post feedback about the flight model of new assets (VTOLs, drone and prop. plane) - how you enjoy flying VTOLs or how crappy it is ;)

Hit us!

[x3kj 1247]

Has the same issues as other fixed wings... Fine course corrections are barely possible unless you use alot of rudder, or you roll 90° and use elevator.

I know from my own test with my VTOL's that AI have issues with low "VTOL Power" (meaning that it takes long to gain heigth with pure vtol). They always bank left /right, until they made enough heigth. In case of low vtol power this leads to alot of sideway movement on starting, making it quite dangerous for anything in the vicinity.

Now that A3 officially has VTOL, i can finally report this :D I haven't tested if the NATO vtol suffers the same issue though.

[razor6014 35]

You might want to increase roll speed by about 10%