I was just finished the aimoffset tourtorial learning phase and got a peek at the demonstration， but the question is since the aimoffset is working as the blendspace， shall I make another animation asset for every basic motion like jumping walking running with the aim motion to be consisted of the aimoffset？ Because the aim motion with standing blend with running seems would drag the leg backward。
if that’s the case you aren’t blending correctly. you need to blend the 2 spaces on the pelvis somewhat like this:
It might not make sense to have an aimoffset applied to certain states. Use a bool value together with ‘Blend by Bool’ node to decide whether or not to apply the aimoffset as the situation requires.
Thank you for helping，but I am not sure how some of the objects work in your picture，like， I can‘t find the“use ecached pose” object in my locomotion diagram， and in that aim offset object， I assume you probably using idle animation with gun holding and breathing to play when character doesn’t pitch or yaw in your aimoffset diagram，but if your aiming up animation has no breathing motion， just the pose， and if your character is in the middle of two animations， you probably find joint in this situation is kind of twitch， because it holds and moves at the same time，so if I want to create an aim offset before I make the whole hierarchy in your picture, should I add breathing motion to the aiming up, down, left, right with the same timing to make sure they blends correctly? Is this the way to do it？
No, aimnoffset is additive if you configure it properly. You literally need just 8 frames or more that function as a place reference of where the bones should end up whilst the base animation and any underlying effects are applied to the animation.
you should follow a tutorial that’s on the docs for aim offset to the letter, and then experiment with changing the animation base for it.
Better yet, download and break apart a Paragon character.