Cheers!
I think there are three ways to approach this. First, if it doesn't destroy your project, you could make the two comps the same size.
Second, you could use sliders to take care of the offset. Add an X, a Y, and a Z expression slider to the "parent" layer in the final comp and pickwhip the expression in the position of your layer in the child comp.
Something like:
-----------------------------
xOffset = {pickwhip to X offset slider};
yOffset = {pickwhip to Y offset slider};
zOffset = {pickwhip to Z offset slider};
x = value[0] + xOffset;
y = value[1] + yOffset;
z = value[2] + zOffset;
[x, y, z]
-----------------------------
This would allow you to correct the position in the parent comp and hopefully would only be one-time tweaks instead of lots of animating.
THIRD, (Dan Ebberts would know better than me) but you might be able to use fromWorld(toWorld())
He has a good example at motionscript.com. I'm just changing "Comp" to "World" since "World is what you use when you are using a camera:
http://www.motionscript.com/design-guide/threeD-corner-pin.html
target = thisComp.layer({pickwhip this to your parent layer});
fromWorld(target.toWorld(target.anchorPoint));
A fourth option might be to make your child comp into a 3D and rasterized layer in the parent comp. (TRY THIS ONE FIRST!)
Good Luck!
Trent
Trent Armstrong - Creative Cow Leader
http://www.dallasaeug.com
