WIP: Naturalis glute soft physics driven morphing

Hey guys and gals.

Firstly, I want to emphasize the "in progress" part in this work in progress. It looks relatively good and works relatively well, but there are a few critical issues with it, listed further below.

In the latest v1.4 alpha, glutes only have g-force morphs which are driven by the accelerations of the glute joint, and are entirely unaffected by colliders, whether soft physics colliders or hard colliders.

In v1.3 and earlier, glute morphs were driven by deformations in the soft physics mesh, similar to how the breast morphs work right now.

What you see in the demo is a marriage of these two systems. It took a lot of trial and error to come up with how to implement this (although I must again emphasize that it's not complete).

Besides enabling collision to drive morphs alongside the g-force morphing system of Shapeshift, the goal of this new system is to allow different directions of force to result in different collision morphs. You can't see this in the demo, since it's just modifying the values of the existing g-force morphs, rather than applying entirely new morphs. But it will come into play later when morphs are implemented that target different g-forces as "baseline" g-forces from which the morph values are calculated.

For example, I plan to implement in TittyMagic the following: when leaning forward and breasts droop forward, pushing the breasts from underneath will not simply undo the droop, but instead cause a different type of morphing that's appropriate when the drooped shape is the starting shape. Each direction of force corresponds to some set of g-force morphs which produce some combined shape that acts as the baseline for a set of soft physics driven morphs - allowing very complex and fine tunable morphing that should look great and feel responsive in any pose.

What are the problems in the implementation so far? Well, the major ones are:

• The angle of the leg has too much effect on the glute morph values. Thankfully this problem won't exist with breast morphs, but is still important to fix.

• The calibration doesn't produce accurate results for some reason. Even if the model is put in the calibration t-pose after calibration, the angles are off by many degrees. This results in the soft physics driven morphs to be applied on top of the g-force morphs, exaggerating the amount of change in shape in different poses or different directions of force even when there's no collision happening.

• The system depends on aggressively dampened g-force values as part of the morph calculation. Without it, the morph values jitter in an ugly way and it's very hard to get rid of that completely.

There's a chance that some of these issues could be fixed with a different design where instead of driving morph values using angles relative to the glute joint, I would hook them up into the depth physics implementation. In the depth physics system, I calculate the depth of each soft physics joints from the center of the mesh (for a video of that see the WIP post from Nov 2023). It could sidestep a lot of the issues and complexity in the current system. My next steps are to do a quick test to see if this has any potential, then to try to fix the above issues, and then to polish the morphing system for release and add a basic UI for it.

The original plan for this month was to focus on breast morphs, but I have to get this release out first! So hopefully that won't take too long to do. The breast morphs need to be built on a system that's close enough to final that there's no risk of needing a radical redesign that would require creating the morphs all over again from scratch. And with a fully fledged system in place, the morphs will be much easier to test within VAM.

Thanks for the support, and thanks for reading if you made it this far! :)

-everlaster