odf posted at 7:15 PM Tue, 18 October 2022 - #4446963

"One of us! One of us! Gooble gobble!"

I've been wondering if the receding pupil might have thrown Poser's match center calculations off. Hard to tell for sure, though, without doing extensive tests or having a peek at the actual code.

Dunno. Doesn't seem to effect translations, only scales. It might though. It seems to be having a hard time finding dead center so your theory might have some merit. But match centers has no problem with finding center with LF's eyes. Then again, I'd have to look, but I'm pretty sure that LF's pupil is a receding cone that's terminus is dead center of the eyeball. An odd pupil design that I always wondered about.

Incidentally your query about just using Poser's scale and matching it to the deformations wouldn't have worked easily. The scaling origin is different between Blender/Poser and the scale morphs I made so it would have to be not only a scale but a slight translation to match. Very tricky with eyes, you need everything to be an exact match or clipping happens. Even the slightest bit of clipping is a real PITA to deal with.

Working on this is going much quicker than the Nova eye stuff I did. Consider that there I had to make 3 sets of morphs for each transformation I did, 2 additional for the conforming crosstalk with the conforming lashes and brows. That was almost nightmarish... a good trial by fire for eye morphs though.

Anyway, I need to make some hidden utility sub dependencies to correct some brow texture stretching/compression at extreme dial ranges. It's a little thing, probably not something the normal range use would be relevant to, you can see it in the animation but those brows are a bit lower than Saintfox's, but I'll do it by using a default textured head in Mudbox which I'll have set as locked and comparing to a textured morphed version which I sculpt the difference onto. Just very minor tweaking to minimize it. I should have baked it in but attaching another sub is no big deal.