Jump to content

Roll control in aircraft reversed when pitch is inverted


StormFury07

Recommended Posts

This post is a copy from the Discord server: (1078874827047313488)

StormFury07:

In airplanes, when the pitch controls are inverted. Roll control also becomes inverted. So if "S" is pitch up, and "W" is pitch down. "E" is roll left, and "Q" is roll right.

In a plane pulling back on the stick is pitch up "S", and pushing on the stick is pitch down "W". Left on the stick "Q" is roll left, and right on the stick "E" is roll right.

In the image below I am rolling to the right but I am pressing the "Q" key. 

image.png

 

 

Immabed:

Were you inverting the controls so that pitch up would be 'S' and pitch down would be 'W'? Because if so, that indicates the wing part made a mistake in picking the default pitch direction.

I found the same thing, in some positions, wings pitch incorrectly by default. I believe this is when the wing attachment point is ahead of the center of mass, but the control surface is behind the center of mass. I believe that is the bug. 

What should be done for part config is individual inversion for pitch, roll, and yaw, as well as individual authority limits for pitch, roll, and yaw. As it is now, especially with the pitch direction glitch, some planes are unflyable due to inverted pitch, and in general aircraft are much harder to tune than KSP 1.

 

Stormfury07: 

Yes, I was selecting inverted controls so that 'S' would be pitch up, and down 'W'.

So the pitch direction on the wing is wrong. That doesn't excuse the fact that even if I wanted 'W' as up and 'S' as down, that the roll should also be inverted.

 

Mech:

SAS also messes up and causes your aircraft to continually do a barrel roll.

 

StormFury07:

This as well, it SAS can't handle the fact that it's in a positive feedback loop. If it rolls left lets say, SAS will roll right ('E' Key) but that just makes it roll left harder!

 

Immabed:

That's always to be expected when the control input moves the craft the wrong way. 

I agree that each axis should have its own inversion controls, since inversion only exists to fix incorrect auto-detection anyways.

Link to comment
Share on other sites

×
×
  • Create New...