PDA

View Full Version : proposal to improve steering (I mean make it worse :-)



quiet_man
11-05-2004, 04:32 PM
Correct me if I'm wrong, but currently plane steering is closely linked to the movement of the computer flight stick

computer sticks are easy to move and any plane feels light and accurate on the controls (with correct flight stick setup)

I would vote to decouple input and in game steering with a damping (rubber band) effect depending on the (real life) force needed to move the controls

In flying conditions with light controls there would be no difference

but in conditions with heavy controls (big planes or tight turns) you would need to over steer to get fast responses from the controls

this hopefully would give more realistic/historic control response

if you please think about it Oleg

thank you
quiet_man

quiet_man
11-05-2004, 04:32 PM
Correct me if I'm wrong, but currently plane steering is closely linked to the movement of the computer flight stick

computer sticks are easy to move and any plane feels light and accurate on the controls (with correct flight stick setup)

I would vote to decouple input and in game steering with a damping (rubber band) effect depending on the (real life) force needed to move the controls

In flying conditions with light controls there would be no difference

but in conditions with heavy controls (big planes or tight turns) you would need to over steer to get fast responses from the controls

this hopefully would give more realistic/historic control response

if you please think about it Oleg

thank you
quiet_man

killer2359
11-07-2004, 01:40 AM
I'd like to add that the function of vehicle steering with mouse in the game Operation Flashpoint is a brilliant example of control coupling of the type quiet_man I believe is talking.

If you play Operation Flashpoint and drive the vehicles you'll see that the control dynamic is lovely in it's use - this would be tremendous in PF if successfully applied.

WWMaxGunz
11-07-2004, 08:10 AM
Quiet Man, try going to stick sliders and move the filter sliders to all full.

quiet_man
11-07-2004, 03:34 PM
<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by WWMaxGunz:
Quiet Man, try going to stick sliders and move the filter sliders to all full. <HR></BLOCKQUOTE>

problem is you can't change this settings midflight or for AI/other online pilots http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

I just read again a WWII pilot report about the difficultis to control his plane at different speeds

IL2 is to much "fly by wire"

Regards,
quiet_man

Nige_Reconman
11-07-2004, 04:36 PM
My force feedback seems to be quite good at forcing you to struggle against the airframe at times.

Are you using force feedback?

WWMaxGunz
11-07-2004, 08:12 PM
Quiet Man, I agree.

However if full filter is a move in the right direction then maybe it would
be good to show and ask Oleg if filtering could be built in to each axis
enough to model at all the calculus of strength versus force to get movement
over time and not quick snap to position up to strength given by stick as a
limit? (FBW, IOW)

I believe that would answer better what the old British pilot who tried out
FB and arrived at the very low slider settings (like 20-some % max) as what
he felt the feel of maneuver speeds, the handling should be.

The sim is fine if the player is slow with the stick and high filter would
not change that at all. I am wondering if full filter available even now
would be near sufficient as I run about 1/2 filter and it's still pretty
quick when I move the stick fast, but not like no filter. It may take
about 4x or more the now maximum to get what that old pilot did with the
very low sliders but at least we could reach full strength.

TheJoyStick
11-07-2004, 10:07 PM
I like that idea.

Blottogg
11-08-2004, 10:18 AM
Guys, stick forces are already modeled. Take a 109, put it in a dive over ~ 450 kph, then try to move the stick, and look down at the stick in the cockpit. At full joystick deflection, the in game stick moves less, and the resulting control forces are less, reducing pitch and roll rates.

Not all aircraft are affected as drastically, because control systems vary from aircraft to aircraft. For example, high speed roll rate is better with the P-38L than the P-38J because the L has hydraulically boosted ailerons, which are modeled in the game.

Short of a FF stick, there is no way for the program to limit the travel of your computer joystick however. Also, the Ai flight models don't have this limitation, in an effort to keep the game playable on desktop PCs.

WWMaxGunz
11-08-2004, 10:35 AM
Okay Blotto; what I am saying is yes, how *far* is modelled but I think that how *fast*
is not. If I snap my joystick full to the side with karate speed and no filter then
how fast does the 3D cockpit stick reach that movement limit? I know the plane jumps.
In the real world with my maximum 50 pound pull there is resistance all through the
pull so it should take some time depending on stick forces. One second or more as
opposed to 1/10th second or so even at slow to moderate speeds. I normally don't
slam the stick like that anyway but I ain't everyone. With filter, you get to the
point anyway, just never as fast and always smoother which IMO gives better flying.

My $.02

clint-ruin
11-08-2004, 10:52 AM
There are some problems that are just unavoidable with short travel joysticks. Rubber banding might help, but have you tried the filter/deadzone settings? Filter will give you some amount of control lag.

Oleg has recommended some of the long-travel joystick mods for Il2 in the past when talking about this.

Blottogg
11-08-2004, 12:49 PM
Gotcha Neal, I misunderstood. I'm not sure, but I think there is a damping factor modeled as well, though it may not be speed sensitive. When I mash full rudder to one side or the other, it takes about one second for full control deflection, even at a standstill (with the sliders all set to "100" in the controls section.) It drives me nuts when trying to laterally correct a gun shot (the "not enough, not enough... too much!!!" PIO phenomena.)

quiet_man
11-08-2004, 01:24 PM
@Blottogg
how do you trim the plane?

it is bad for your aim if you have to move the stick between back and forth, trim the plane so that you need continous pressure (pull) on the stick to keep your aim

at the beginning, you will often lose aim the moment you press the trigger (like beginner shooting a pistol) but after some training you will be able to control the fire much better

this is also used by todays real figther pilots

Regards,
quiet_man

effte
11-12-2004, 05:27 AM
A) The controls are force based, as requested in the first post. Push your joystick fully forward and the virtual stick will move as far as the maximum allowed virtual force would move it

B) There is damping on the controls. You can€t slam the rudders between the stops as fast as your stick/pedals can go.

C) This damping has nothing to do with real life. In a real aircraft, the controls get heavy. However, any significant damping (less than currently present in Il-2) would make them feel like they are submerged in syrup. This most does not happen, at least in small aircraft with manual control systems. What limits the onset of control effects is the moments of inertia of the aircraft, the effects of which are orders of magnitude above any accelerations present in the control system.

You don€t actually slam the controls around though, but only as it wouldn€t be effective. Apart from risking flutter and damage to the structure from the resulting loads, you€d be out of the feedback loop as you can€t take in the response from the aircraft as fast as you can move the controls. You€d be overcontrolling all the time. One advantage of FBW controls is that you can indeed slam the controls hard over and you will get maximum effect as fast as possible, where as with conventional controls you€d just overcontrol and possibly stall out or throw yourself into GLOC.

Control damping was not in the early version of Aces High, an this was not a problem as far as flying the aircraft goes. It was introduced some way down the road as fast rudder movements were throwing off the lag compensation algorithms, which depend on a degree of predictability, creating erratic movement of the aircraft on the screen of other players. This is one reason to have it in place in Il-2, control spiking is another.

quiet_man
11-12-2004, 10:29 AM
@effte:
A -&gt; control limit is modeled
but imho wrong because the limited control movement is translated to the whole range of the PC stick increasing accuracy of steering
a good advantage for airplanes that had RL steering optimized for low speed

B -&gt; some sort of damping is modeled, but is independend from control forces (and I think also from plane)
I think when control forces reach full strength of pilot (like modeled by control limits) he will not be able to move the stick as fast and accurate like we in the game

C -&gt; don't know if you count WWII planes as "big", but many of them in fact exceeded "syrup" level and reached up to "beton" http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

Regards,
quiet_man

Blottogg
11-12-2004, 02:37 PM
effte, thanks for the explaination. I had a feeling it was something like that. It still drives me nut though, and forces me to unlearn old habit patterns from real aircraft. I rarely use rudder now, since the lag makes it ineffective for quick, subtle corrections.

quiet_man, the technique you describe is one I used to teach students to remain stable in formation flight. By trimming forward past level flight trim, you can vary pitch by changing the amount of backstick pressure, instead of having to alternately push/pull to make subtle corrections up or down. In extreme cases (such as military formation teams) lots of forward trim may be used (25 lbs or more). This constant arm tension lends more stability to the aircraft (until fatigue sets in), and probably builds a serious right bicep, too.

Unfortunately, this won't work well with rudder, since the problem is not a dead zone on center, nor alternating left/right inputs, but a damping factor within the sim engine itself. Plus, holding constant left or right rudder for the flight, and trying to aim by adjusting pedal pressure on one side constantly, would be too much like flying one of the 109's in the game which have no rudder trim. The real German pilots did this, but without the rubber band effect we have.

quiet_man
11-12-2004, 04:35 PM
@Blottogg
how did you set up the sliders in IL2?
Rudders work best for me if I set all sliders high, I have the first three at 60 and the others at 100%

I don't know why Oleg decided to make pedals so slow, but it shows the effect I would like to see modelled for all controls dependend on control forces

it would at least give a hint with what difficulties the real pilot was burdened

Regards,
quiet_man

Blottogg
11-12-2004, 04:47 PM
quiet_man, I've got them all maxed out to 100. Like effte posted, I think this damping was put in to avoid spiking and sim engine limitations. I'm not sure why the affect isn't as bad in pitch and roll however.

quiet_man
11-13-2004, 03:20 AM
@Blottogg
at least I know now that I'm not the only one to struggle with the rudder http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

Network has nothing to do with it, IL2 calculates the plane you fly localy on your pc and
from time to time it sends position and speeds to the server
hopping planes, collision bug or hitting the target and nothing happens, thats all related to player PCs having different data dependend on Network travel time
imho "cheating detected" should be changed to "lag detected"

I think steering is just neglected by the community and Oleg, cause it is diffucult to "measure"

and after finding the control limit "bug" (see other thread) I would say steering is one of the weaker points of FB/AEP/PF

Regards,
quiet_man

effte
11-14-2004, 11:28 AM
<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by quiet_man:
Network has nothing to do with it, IL2 calculates the plane you fly localy on your pc and
from time to time it sends position and speeds to the server
<HR></BLOCKQUOTE>

Exactly. These intervals, a Hz or two, are not near enough to provide smooth motion. Thus, the computer at the other end has to perform some interpolation (extrapolation, really). For this purpose, not only the position but also the current velocity and accelerations (angular and linear) are sent. The other computer then uses these to project a trajectory further in time. If there is no predictability in the motion of your aircraft, the difference between the actual position and this projected trajectory will be large at the next update. The position has to be corrected and you get a lag jump.

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>
hopping planes, collision bug or hitting the target and nothing happens, thats all related to player PCs having different data dependend on Network travel time <HR></BLOCKQUOTE>

You are indeed looking at the forest but you fail to see the trees that form it. http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

quiet_man
11-14-2004, 01:00 PM
@effte
you mean the rudder is slow to make flight paths more predictable?

rudder are used to stabilize the flight path or for small changes of direction, they have minor effect on flight path

no good explenation

Regards,
quiet_man

effte
11-15-2004, 02:24 AM
Control surfaces, not only the rudder.

And if you had ever been behind someone kicking full rudder, alternating left and right, with a half-second of net lag on top of that, you'd be fully aware of what this does to your guns solution.

quiet_man
11-15-2004, 09:38 AM
@effte
that's the point, why should they slow down the surface that has the least impact on flight?

0.5s pings should be avoided at all, IL2 does an exelent job on minimizing Network data but the engine does not support high ping times, even without using rudder

but to return to the topic, why not add a delay to the stick similar to the rudder delay, when you reach the limits of the virtual pilot?

Just to give you an hint of the forces youre fighting, instead of the "fly by wire feeling" we currently have

any better ideas?

Who would support writing an e-mail to Oleg?

Regards,
quiet_man

boohaa
11-15-2004, 10:05 PM
Wow...thats a good idea Quiet man.

Right now IL2 has stick forces modeled just for maximum amount of deflection at varying speeds.Now add in response for speed AND for each planes diferences and youll greatly improve feel in IL2.

I hope you are better at explaining this here Quietman and wish you best of luckhttp://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

Heres an idea Ive been wondering about.It has more to do with BOB since it needs new programming.It also has to do with Track-IR.

OK what I propose here is that the screen moves in realtion to G's pulled.Now imagine diving at high speed and when you start to pullout,the view shifts more to look down at the gauges in amount dictaed by the G's expeirienced.This will have an effect that coupled with track-ir will give the feeling of pushing you head back up to look out the cockpit.

Just sit at your desk and imagine pulling out of a dive and at same time lift your head back to counter the G's.Hard nose down and youll have to push your head down....mimicing the fact taht your head will be thrown back and youll have to fight this tendency.Same goes for rolls.Nuff said now Oleg....please for BOBhttp://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

Dale

effte
11-16-2004, 01:49 AM
That€s just the point. In real aircraft, there is no limit to the speed at which you can move the controls. The only thing stopping you from violently slamming them back and forth is self-preservation. There are no huge moments of inertia in these wire and pushrod control system. There are no hydraulic dampers. You can slam the rudders between the stops and the only thing which will stop you from doing so until they break are the aerodynamic forces, which will not slow down the rate at which you can move the controls, only the amount of deflection your own physical strength will allow you to achieve.

So, why on earth would you want such an artificial limitation, beyond what is required due to limitations in the simulation and hardware, imposed on us in Il-2?!

quiet-man,
least effect? The rudder has a pretty significant effect on flight. Try mashing a pedal down in slow flight in a real aircraft sometime... or when near the stall in Il-2, especially near an accelerated stall pulling a few Gs. These are extreme situations, but as I said, even in less extreme situations rudder input with a bit of net lag on top will create very peculiar effects if nothing is done to handle it.

As for net lag, half a second isn€t much. What you do in your simulator has to go through your system, halfway around the world, be processed there, come back and again be processed. The fact that we don€t suffer more than we do from it now (trust me, it has been a lot worse in many simulations) is completely down to very clever algorithms and solutions to deal with it.

quiet_man
11-16-2004, 11:52 AM
@effte
stick forces? Thats why Oleg included controll limitation effects.

But how easy is it to keep a plane just below stall, when you have to pull constantly XXlbs on the stick? At least a minimal feedback of what's happening in the sim would be fine for me.


about rudder:
witch on would you miss more: rudder or elevator? But maybe we're both right and they slowed down rudder to save Network bandwith, CAUSE it is less important than elevator? Is this okey for you to accept?

Regards,
quiet_man

effte
11-16-2004, 02:31 PM
<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>
@effte
stick forces? Thats why Oleg included controll limitation effects.
<HR></BLOCKQUOTE>

The honour of having invented the stick force PC simulator control system falls elsewhere, but yes, the fact that stick forces is what matters is the reason for it being implemented.


<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>
But how easy is it to keep a plane just below stall, when you have to pull constantly XXlbs on the stick? At least a minimal feedback of what's happening in the sim would be fine for me.
<HR></BLOCKQUOTE>

I think we might possibly have a bit of a language barrier here. English isn't my native tongue either. http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

What exactly are you asking? In real life, flying on the edge of the stall is easy and yes, this is when keeping a constant force on the stick. In fact, it is easier to fly with precision with a steady force on the stick. For finals, many (most?) people trim a bit nose heavy so that the control can be made only by varying the amount of pull.

We have not only minimal feedback of what is happening but the best feedback anyone has come up with to date, short of force feedback which IMO is outside the bounds of the off the shelf systems on the consumer market today.

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>
about rudder:
witch on would you miss more: rudder or elevator? But maybe we're both right and they slowed down rudder to save Network bandwith, CAUSE it is less important than elevator? Is this okey for you to accept? <HR></BLOCKQUOTE>

It is not a bandwidth issue.

The principle applies to all controls alike, elevator, aileron, rudder.

Cheers,
Fred

quiet_man
11-18-2004, 01:59 PM
@effte
at nearly all WWII planes the pilot moved the control surfaces by muscle

depending on the size of the surface and the speed of the plane, the pilot was not able to move the controlls the full way

this is modelled in FB/AEP/PF

but we get nearly no feedback about this, controls feel very light all the time, even if our virtual pilot is working like crazy

in RL I know the difference between 1 and 50kg, in IL2 it feels the same http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

Regards,
quiet_man

effte
11-19-2004, 05:20 AM
Unless pulling your stick to half deflection feels the same as pulling it to full deflection, it does not feel the same.

If it does, you will have to go get a better joystick.

Half deflection means half the available force on the virtual stick is applied*. Full deflection means the full available force is applied.

Half deflection should require about half the force of full deflection, with a reasonably good joystick.

Or, to put it another way if it is still not clear: If you apply the maximum usable force to your joystick, your virtual pilot is applying the maximum available force in the simulation. If you are applying half the maximum usable force to your joystick, your virtual pilot is applying half the maximum available force in the simulation.

If you can't feel a difference, the problem is with your hardware.

/Fred

*) Depending on the stick scaling settings, of course.

quiet_man
11-19-2004, 09:08 AM
@effte
now we speak about the same thing http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif
you are right about the stick forces

But what I mean is the difference when changing from 5 to 10% and from 95 to 100% input
in the sim you can change between 95 and 100% as fast and accurat as between 5 and 10%

if you take a "slow" plane, say P11 or I16 and a fast plane like Mustang or FW190
now fly the slow plane at "high" speed, the controlls become heavier and you need to move the stick wide, you can't stall -&gt; !!!steering becomes easier and accurate!!!

for the fast plane the speed is still slow, you can only use small stick movement or you stall -&gt; !!!less accurate and higher difficulty!!!

the result is that it is very very difficult to fight slow planes in IL2, as long as the slow plane does not reach the speed limit, it gains advantages from higher speed


in RL, if 100% is the pilot pulling with everything he has, it would be very difficult to fine tune the input

that's what I would like to see in IL2

Regards,
quiet_man

effte
11-20-2004, 05:25 PM
In a slow aircraft at high speed, chances are you will not be able to pull hard enough to stall it. You don't have enough control authority to bring it to the critical angle of attack, as you are outside the envelope within which the flight control system designers considered maximum load factor necessary. Chances are that you are above Va anyway, meaning you'd get structural damage due to excessive loads before stalling the aircraft.

In a fast aircraft at slow speeds, it will be easy to stall as the flight controls are designed to provide manoeuvrability at high speed.

If you are pulling everything you have, it might get a little harder to make fine adjustments. It is harder to make fine adjustments if you have the joystick to the stop in Il-2, if you have anything resembling useful stick settings. If you don't, you will be hard pressed to fly with any precision at all anyway.

This is why you have trim, in Il-2 and in real life. Use it wisely and anticipate where in the envelope the fight will end up.

Thinking things over in advance is a good piece of advice in most situations, be it flying, flying simulators or writing on internet forums.

quiet_man
11-21-2004, 06:57 AM
@effte
you got exactly my point, just the other way round http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif
steering in IL2 is to easy and especial for slow planes! maybe my stick setup is to good http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

I have to problems to do ground attacks with Mustang, P47 or FW190 but with the I16 I'm a sniper, no need at all for trim!

and this is because of the control lockup modeling in IL2
when you lose control authority the full stick input is translated to less control output

so the planes feel more stable and you can make very very fine corrections to your aim. This is very similar to fly by wire systems where the computer reduces control output according to higher speeds.

Regards,
quiet_man

quiet_man
11-25-2004, 09:35 AM
hmm, is the lack of responce a lack of interest?
or does it mean that everyone agrees?

should I disturb Oleg with this?

Thank you for the discussion effte,
quiet_man

effte
11-25-2004, 09:55 AM
No offence, but you really need to think through how control systems work, how they are implemented in the Il-2 series and the implications thereof. You are seeing features as problems. Give it another read. http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif