You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When dragging one gizmo axis, perpendicular mouse movement causes incorrect movement. Gif below displays how dragging vertical axis sideways moves object up and down. Other two axis do not have this problem.
It is not always same axis. It depends on point of view which axis misbehaves. This happens when translating or scaling. Depending on angle of view small sideways movement may cause huge changes and cause object "fly" out of the view and sometimes finding it can be tricky.
PR #33 seems to make this less of a problem i think, but does not fix it entirely. Or i may be imagining things and it was less of a problem only due to my point of view when testing. I am no longer sure.
The text was updated successfully, but these errors were encountered:
I can confirm this issue when I have my camera at (0, 0, 0), looking at (0, 0, -1) with a y-up coordinate system trying to move an object at (0, 0, -2). X-axis movement simply doesn't work in that case, but when I move the object up slightly x-axis movement goes crazy.
The underlying issue (in my case at least) is that when you translate an object on the x-axis while the view direction is perpendicular to the plane normal, the intersection test will not give sensible results as the mouse ray is always on the plane at the camera position. The only fix that I can think of for this is to implement a different translation mode for axis-snapping that involves testing against the axis directly instead of against that plane.
When dragging one gizmo axis, perpendicular mouse movement causes incorrect movement. Gif below displays how dragging vertical axis sideways moves object up and down. Other two axis do not have this problem.
It is not always same axis. It depends on point of view which axis misbehaves. This happens when translating or scaling. Depending on angle of view small sideways movement may cause huge changes and cause object "fly" out of the view and sometimes finding it can be tricky.
PR #33 seems to make this less of a problem i think, but does not fix it entirely. Or i may be imagining things and it was less of a problem only due to my point of view when testing. I am no longer sure.
The text was updated successfully, but these errors were encountered: