Target Bone / Smart Bone Bug

Discuss Moho bugs (or suspected bugs) with other users. To report bugs to Smith Micro, please visit support.smithmicro.com

Moderators: Víctor Paredes, Belgarath, slowtiger

Post Reply
dkwroot
Posts: 677
Joined: Thu May 02, 2013 6:56 am
Location: USA
Contact:

Target Bone / Smart Bone Bug

Post by dkwroot »



Essentially, what seems to be happening is that when a bone is under the influence of a target bone it doesn't rotate around into negative degrees, but instead cycles back to 359. The problem is that smartbones are designed to use negative degrees and in some cases this is unavoidable. When these two forces collide, you have a situation where a smartbone will move under the influence of a target bone until it reaches the 0 threshold and then suddenly pop back to the original position of the bone instead of progress through the smartbone action.

If anyone else has AS 10, I'd like to have this verified.
User avatar
heyvern
Posts: 7035
Joined: Fri Sep 02, 2005 4:49 am

Re: Target Bone / Smart Bone Bug

Post by heyvern »

I have encountered this, but somehow I was able to work around it and I'm not sure exactly how precisely. It was sort of an "accident" based on the way the bones are parented and the initial rotation of the thigh bone. I was working on a different type of rig that used a knee and foot target.

With my rig, this means that both the calf/shin have two different ways to "rotate" so I can control how much rotation is applied to the thigh. It's not determined by the foot target only. If I push the foot way up the thigh doesn't rotate at all unless I translate the knee target.

Another thing that I noticed looking at my file is that my thigh bone smart bone action never goes above or below the "0" flip point caused by the target. The thighs are parented in such a way that the angle at 0 frame of the SB is 0 degrees, which gives me 180 degrees of rotation in the SB. It never reaches that "flip" point caused by the target bone IK solver.

For example I always have a body root bone that is rotated 0 degrees with a pelvis or hip bone child of the root rotated to 270 degrees. The thigh is a child of the hip bone and is rotated "down" to around 0 degrees give or take. This means there is 180 degrees on either side for smart bone actions without causing the flip.

These are just possible "work-a-rounds" and I agree that this is an issue. Even with that solution it still limits the amount of rotation for a smart bone of this type (an arm would be more problematic). It would be better if the IK solver could just "spin" the bone continuously rather than flipping back. I'm not sure though if this is possible. Most IK solvers work this way.
Post Reply