Free Allegiance Forums

Because not all controllers are identical, and other people are perfectly capable of adjusting their controller/joystick sensitivity.

Jimen wrote:QUOTE (Jimen @ Jun 16 2010, 07:41 PM) Because not all controllers are identical, and other people are perfectly capable of adjusting their controller/joystick sensitivity.

Even though controllers are not identical, the software SCALES them such that they exhibit the same characteristics when used. My joystick has a short throw compared to others (tilts only around 25 degrees left or right). When scaled, the results would equate to the same output as a joystick of a larger throw. Using either joystick at full deflection should deliver the ships full turning capability.

IF I adjust the joystick sensitivity so that it is less sensitive, then it reduces the ship turning rate when I fully deflect the joystick. Comparing the difference between turn rates using quadratic and linear options, they should both yield the same max turn rate. However, they don't. This suggests that either the axis is not being scaled properly or the linear/quadratic rate is not being properly caluclated or maybe it's not factoring in the field of view properly. Just can't put my finger on it.

jbansk wrote:QUOTE (jbansk @ Jun 16 2010, 02:27 PM) IF I adjust the joystick sensitivity so that it is less sensitive, then it reduces the ship turning rate when I fully deflect the joystick.

I'm no joystick expert (mouse flyer since 2001, yo) but isn't this normal behavior? It certainly is in console games - by decreasing the control sensitivity, you are decreasing the amount of on-screen movement for any given joystick movement.

Jimen wrote:QUOTE (Jimen @ Jun 16 2010, 08:41 PM) I'm no joystick expert (mouse flyer since 2001, yo) but isn't this normal behavior? It certainly is in console games - by decreasing the control sensitivity, you are decreasing the amount of on-screen movement for any given joystick movement.

I believe you're right. That leads me to believe that the problem lies in the quadratic / linear calculation or perhaps the field of view. I'm wondering if it has to do with the collection of axis values from my stick.

I found that if I increase the turn rate so that it takes 8.15 seconds to turn 360 degrees using quadratic acceleration that if I change the acceleration to linear, the turn rate is increased and it takes then 7 seconds to turn 360 degrees.

I wonder if this isn't flawed. I would think that no matter what turn rate I impliment, the linear / quadratic accel rate would then be calculated using the extreme axis value I created. How can the maximum turn rate be altered by just changing the acceleration rate?

???

jbansk wrote:QUOTE (jbansk @ Jun 16 2010, 03:09 PM) I believe you're right. That leads me to believe that the problem lies in the quadratic / linear calculation or perhaps the field of view. Since this is new to the Dx9 version and didn't exist before, I'm wondering if it has to do with the collection of axis values from my stick. Maybe I am now able to turn at full ship capability where before I wasn't able and just didn't realize it.

What problem? Isn't this intended behavior? If you increase the sensitivity, then your turn rate should decrease REGARDLESS of how far you're shoving your stick, shouldn't it? You seem to believe that you are in a magical fairy world where you can crank the sensitivity way up while still getting the absolute turn rate, and I'm pretty sure that just doesn't happen...unless you use a mouse. Granted, I could be wrong, but until someone who can figure out what the hell you're talking about shows up, I'm going to ask one simple question:

QUOTE (jbansk)IF I adjust the joystick sensitivity so that it is less sensitive, then it reduces the ship turning rate when I fully deflect the joystick.[/quote]
ISN'T THIS EXACTLY WHAT IS SUPPOSED TO HAPPEN?

You have a minimum and maximum axis value(s) coming from your controller. The linear (straight line) and quadratic (curved line) rates dictate the accelerated speed the controller will move between both those min / max values. Using quadratic rate, the closer you get to the maximum axis value (full throw of stick) the more sensitive it becomes and the faster it reaches the maximum value. Using Linear, the acceleration rate from minimum to maximum axis value remains constant. The sensitivity does not change from small inputs to large ones.

If I increase the upper axis value on any given axis then BOTH the linear and quadratic rates should be using that maximum value in their formula to calculate their acceleration rate. The linear curve (sheesh) is a straight line between the min and max axis value. The quadratic curve is a real curve from the min and max axis value. They both are just lines between two constant values. If the maximum remains the same (meaning if I increase the axis value to reduce sensitivity) then my ship should turn at the same rate when using full axis value (full deflection of stick) REGARDLESS of which acceleration rate I select.

Am I misunderstanding what linear and quadratic rates are? I would think if I change an axis value to make my ship turn at 70 dps it should turn at 70 dps regardless of what accel rate used.

Are you checking turn rates for the screen or the G indicator? Because it doesn't matter HOW FAST your screen turns, the ship isn't truly turning with it. It's turning to catch up to it.

PS - I think it's EPIC there can be this much thread about something this silly.

jbansk wrote:QUOTE (jbansk @ Jun 19 2010, 10:28 AM) I found that if I increase the turn rate so that it takes 8.15 seconds to turn 360 degrees using quadratic acceleration that if I change the acceleration to linear, the turn rate is increased and it takes then 7 seconds to turn 360 degrees.

I wonder if this isn't flawed. I would think that no matter what turn rate I impliment, the linear / quadratic accel rate would then be calculated using the extreme axis value I created. How can the maximum turn rate be altered by just changing the acceleration rate?

jbansk wrote:QUOTE (jbansk @ Jun 19 2010, 01:03 PM) Am I misunderstanding what linear and quadratic rates are? I would think if I change an axis value to make my ship turn at 70 dps it should turn at 70 dps regardless of what accel rate used.

Hmm. Have you tried setting up your joystick to act as a mouse, and then turning on virtual joystick? You'd need to set it so that mouse position is relative to joystick position, not mouse movement relative to joystick position, but it might fix your problem.

As far as getting different turn values at full-deflection depending on whether quadratic or linear turn rates are used, then if its reproducible on other systems I'd call that a bug in either the linear or quadratic turn-rate code. Or both. This requires testing by mouse too.

Suggestions for testing:
Make a ship with 0 pitch/roll, and can only yaw. Stop somewhere. Spin as fast as possible using the mouse in linear mode. Fraps it, and use fraps to measure effective turn rate. Do the same with the mouse in quadratic mode.

Compare the fraps results. If one turn rate is higher than the other, we've got a pretty serious bug.

Repeat the same test with multiple joysticks.