I've been animating a set of 2D images against a long strip background (flower, bird). I have been splitting up the animation between Pose Mode and Object Mode. I set the character's gestures in Pose Mode, then I animate their motion along X using Object Mode. I have never done this type of 2D animation before and chose this method because it seemed logical. Is it better to do all of the character's gestures and X movement in Pose Mode? Thanks. Zaffer
If you are just doing all the animation on the armature (pose and object mode), then I see no harm in that. In fact I think there is a little benefit over animating it all in pose mode.
Here I have the X axis movement key framed on the armature object (it is 40 frames long). The small bounce along the Z axis is keyframed on the bone in pose mode (its animation only lasts 10 frames). Doing it like this enables me to separate the animation, and keyframe the bounce in local space, and then use a Cycle modifier in the graph editor to make it loop.
However if you are animating the mesh object too, then that is a problem.
The reason for this is because of the way parenting works in blender.
See parenting is one way, meaning that you can move the parent (the armature) and both will move, but if you move the child (the mesh object) it separates from the armature. When you bind an armature to an object you are parenting it among other things.