Jump to content
...srqbones

How Joystick "Curves" work - should be second order instead of first order.

Recommended Posts

There is another post that appears to argue the reality or non-reality of using modifications to joystick control "curves".  For the purpose of this topic, I will just accept that curves are useful, especially when trying to simulate a 3 foot control stick with a 6 inch desk mounted joystick, or when mapping the roughly 270° of head and torso motion with a 10° arc from the center of the screen using Track IR.

The way that curves have been set up to work, both in joystick modifications, as well as for Track IR modifications, is that they allow the user to set up less sensitivity when near-center and greater sensitivity when further away from center.  In theory, this works fairly well.  If your joystick is trimmed out and you are in neutral flight, and essentially you are "trimmed" when on the bandit's 6 o'clock in a dogfight, then such curves will allow you to make milder adjustments to obtain a firing solution.  Similarly, using Track IR, when you are nearer to the center of the airplane, the image shouldn't appear to jump around as much when you are zoomed in (I won't even bother discussing the "precision" mode of Track IR, as I have not figured out how to go in and out of precision mode depending upon how zoomed-in the flight simulators is).  Let's call this type of modified sensitivity a "first-order" sensitivity, dependent upon position of the controller.

It sounds great on paper, but it doesn't seem to work as well for poor pilots as myself!  I generally do not trim out in the middle of a dogfight, so frequently as I am pulling some G while trying to stay in the slot behind my target, I am pulling on the stick.  Thus, I already have a fair amount of deformation from center on my stick, especially in the pitch axis.  Similarly, I often zoom-in with the screen while looking to identify whether the bogey at my 7 or 8 o'clock is a friendly or a foe.  In both of these cases, I am no longer down on the low sensitivity area of my adjustment curves, even though my goal is to make fairly fine changes (it's really amazing how much the screen jumps around while looking towards my 7 o'clock and then progressively increasing my zoom).

I guess what we really need is a sensitivity curve based not on first-order position but on second-order change in position (speed of change in position).  In theory, this would be taken up by damping, but I do not understand the mathematics between decreasing the sensitivity or how damping functions work (the latter may, for as much as I understand, remove some of these finer position changes, where that is NOT AT ALL what the goal would be here).  In other words, no matter where your stick position is being held, relatively slow and small movements to the stick, such as when tracking a bandit, should be in fact made small.  Similarly, with Track IR, relatively small head motions when zoomed-in and looking off-center should also be made small.

As I am no game designer, no computer programmer, and no mathematician, do you 1C Studios designers think that this is a reasonable goal that can be achieved, this "second-order" basis for curves for joystick motions and/or track IR (I understand that you do not write the track IR software)?

PS – The program War Thunder has a clever view modification: when you bring your head closer to the screen, not only do you get closer to the things in the cockpit, but it also increases the zoom at the same time.  This works quite well for looking at things in the distance especially when they are not on the centerline of the airplane, such as my example of trying to identify the aircraft off my 7 o'clock.  In fact, it's so intuitive, that I find myself zooming in and out without even thinking about it, just to get a better view of an aircraft or building in the distance, or to increase my field of view when maneuvering or in a dogfight.  If this type of feature is not under intellectual property protection, you guys should adopt this!

  • Upvote 2

Share this post


Link to post
Share on other sites

The only problem with that zoom system is when you are looking side ways out the cockpit and the canopy frame is in the way. In this case I move my head to look around the frame and the camera just zooms...It annoys me.

For this reason I turn off the track IR Z axis zoom and prefer to use the hotas switch for zoom in and out. 

Share this post


Link to post
Share on other sites

I guess I don't see how when the canopy gets in the way, it works better with the Hat switch to look left then just to move your head around so you can see past the canopy. Isn't that the way real life works ?

Share this post


Link to post
Share on other sites

Very interesting in theory. I'm not sure how it would work in practice though. This type of behavior is more suited to aircraft with force sensing sticks like the F-16 or some civilian transport category aircraft. For a WWII fighter it would be harder to predict when the player is asking for full control deflection without force sensing with that model. That might lead to some unpredictability for the pilot, which is the exact opposite of what you want. I can't begin to imagine what spikes in a potentiometer would do with such a system.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...