Realistic Curve Settings

Thing is a real stick in flight doesnt really have a center per se. You trim it till you dont have to hold it in place, normally in pitch because you have to trim for speed, that position is then the new position it holds if you let go. ( not talking about Aircraft with Autotrim FBW like the M2K). In the Sim unless you have fancy Hyd damped Cyclic then you trim the stick it goes to its center position, in other words as you trim you move the stick till it is back in its center position. This is a problem with a hard center as it produces a tendancy to overcontrol in close formation. I used to deliberately fly out of trim to keep the stick out of the center notch with my Warthog if I was flying close, which is certainly not realistic. It also made the M2K a little tricky as it Autotrim means that you are often stuck in the center detent.

Of course if you have a soft center detent then this is not a problem. To be brutally honest its not really relevant as you fly a plane by moving the controls to get the desired response which you can do no problem in DCS whatever your controls just some make life easier than others.

Sent from my SM-G955U using Tapatalk

.

Your physical joystick is a zero order controller that controls a first order controller in game. So whatever is built in to the aircraft is a non-issue.

https://mwomercs.com/forums/topic/118355-controls-demystified/

The in-game controller is likely longer and therefore affords a higher degree of precision. With 1:1 control over the in game stick, you wouldn't be as precise. Thus you need a curve. Also, this means curve required is dependent on many things.

Curve is not used to simulate any flight characteristics. It only effects your zero order control over a first order controller.

Without FFB, trimming to center is the closest thing possible to real life. In real life you would trim until the stick stopped pressing against your hand giving it a new center. Trimming to center does the same thing, except you have to move your hand slightly as you do it.

Your physical joystick is a zero order controller that controls a first order controller in game

Not sure about that description, but suffice to stay "your stick is a toy replica." There are even greater concerns than length of throw, like force applied and damping. (Natural and artificial, depending on the aircraft.) None of these are dependent on the speed, g load, or control rate as they are in real airplanes.

Not sure about that description, but suffice to stay "your stick is a toy replica." There are even greater concerns than length of throw, like force applied and damping. (Natural and artificial, depending on the aircraft.) None of these are dependent on the speed, g load, or control rate as they are in real airplanes.

A zero order device physically moves the thing to where you want it. You physically move your stick to the position you want the in game stick to be. That'z zero order control.

The plane is controlled by the in game stick. The in game stick controls the velocity of the plane's AoA changes. A first order controller controls velocity. The in game cockpit stick is first order control.

You're not wrong at all though. But, I do think it's important to mentally de-couple the physical joystick in your room on your desk and the in-game joystick in the cockpit. One reason for this is because while the physical stick has zero order control over the in-game stick, the in-game stick has no control over the physical stick. So, this avoids people thinking they need to compensate for some flight characteristics, especially in the older planes that don't have fbw. For example, you wouldn't try to set a curve to emulate stick pressure in a WWII aircraft, because there is no way for the game to communicate that unless it is artificially saturating your in-game stick movement. The closest I've experienced to this in DCS would be controls locking out at high speed in WWII aircraft. Even if you did, your curve would still be entirely situation based, and just an overall mess.

Of course, I could be misinterpreting control orders. My apologies if I am. That Mech Warrior thread I linked is great though for anyone who swims the sea of peripherals. It really emphasizes how important it is to understand the relationship between what you're trying to do in-game, how the game processes controls input, and how you're inputting those controls.

Edit: This entire conversation and issue stems from the lack of FFB controls in the flight sim world. If we had access to proper force feedback joysticks and yokes, we wouldn't have to amalgamate flight controls in weird ways like this. IMO, force sensing setups are the closest we can reliably get to fully realistic flight controls. Though there was that guy making FFB WH bases, I haven't heard anything recently.

TL;dr: set your curve where you can comfortably fly close formation, but not too crazy where you can't maneuver in combat. Trim however you like, but remember that if you don't trim to center, you will be in a different area of your curve.

Hi everybody,

First post for me; I am Matteo from Italy. I haven't found an “Introduce yourself” thread; please tell me if there is one.

Back to the topic:

I don't think that there is a realistic curve everyone can use, because I guess that every joystick already has its own curve and so when we edit the one in DCS we are making the curve on another curve.

It seems to me that most of the times either the exponential or the logarithmic curve (or approximations of them) can be used (see the attached file).

The log. one has a quicker response for small input and the exp. has a slower response for small inputs.

I see that most of the people use an exponential curve and that doesn't surprise me.

According to the Weber-Fechner's law, for a lot of our senses, the sensation we get tends to depend on the logarithm of the intensity; you can find a lot about it online (where I found it out). Therefore, an exponential curve compensates for our logarithmic brain.

As a matter of fact, they install logartihmic potentiometers behind the volume knob of stereos: those have actually an exponential behaviour.

Anyhow, my Thrustmaster T16000M seems to have a built-in logarithmic curve.

In my T16000M, after a lot of use, the Hall Effect sensors still seem to work fine; however, the joystick has some jitter at center which, IMO, is caused by its mechanical components: the stick doesn't always stay in a precise centered position: it can wiggle under its own weight (maybe even due to wear) and so I use a deadzone of 5, which seems to have helped me. Well, you can tell from the graph attached that my hand is not that precise! Ahahah. I don't feel like I need a curve to smoothen the transictions between the dead zone and the active zone.

I find that I just want a straight line with the SU 27 and 33 because they already seem not very reactive at little stick input and I prefer a quick response when approaching the carrier: sometimes I have to correct my heading/pitch /etc. quickly. On the F15, on the other hand, I find that a curve of 10 helps me when I am shooting a target with the cannon and I should also try that with the aerial refuelling, which I have never managed to do successfully.