Philip,
Near as I can tell, there is some problem with finding specific
pixels, and assigning them to the mouse position. If you drag-resize
the Flash window (I'm using a PC), making it a bit bigger, the shaking
stops. Also, if you maximize the window, the shaking stops.
I put a trace() in Away3D's away3d.containers.View3D.findHit():
trace("View3D.findHit: stage.mouseX:" + stage.mouseX + ", ...y:" +
stage.mouseY);
and it seems to jump an extra pixel when the mouse is moved.
Here is what I think is happening, but I don't have time to track it
down: The mouse movements are mapping between onscreen and 3D
pixels. The mouse movement onscreen is *less* precise than the 3D
pixel location, so the onscreen pixel ends up tracking to two pixels.
Further, I would guess that every time the mouse onscreen *location*
changes, it fires a new "mouse move 3D" event. When the 3D point
translation calculation places the onscreen mouse at the wrong pixel,
it notes that the current onscreen mouse point != the stored onscreen
mouse point, and fires a new event.
This could be a rounding error, or it could be a onscreen mouse point
to 3d mouse point calculation error. It also could be the way you
have set up your material, so that's another place to look.
Can you (or someone) else finish tracking down this bug?
On Oct 16, 10:19 am, Philip <[email protected]> wrote:
> Hi delfeld,
>
> I've rar'd the source and uploaded
> to:http://www.milkisevil.com/dumpbox/away3d-shaking-filters-on-moviemate...
>
> I'm wondering if anyone else has this problem when trying to achieve
> the same (adding filters to sprites within a MovieMaterial)?
>
> btw... the source is an FDT project, but just run the src/Main.as as
> the application entrypoint.
