Advanced RFL setup. questions

Hey all.

I was watching an old school video and was relating it to some of the newer information that I’ve come across via question and answer sessions on various boards, watching Digital Tutors or other tutorials showing placement of the foot joints inside the mesh as opposed to snapping to the grid. Let me start here.


Now one of the questions that pops into my head is, this. In the pic here




You can see that the original ball and toe joints were drawn directly snapping “X” to grid and holding shift while drawing the toeEnd joint. Ok, now that poses a question that I’ve asked various TDs and that is, “why do I see some folks(the the DT link inserted) place the joints, regardless of the shape or thickness of the foot, snapped to the grid while others place it centered inside the mass of the foot?” I get the middle center for proper deformation,holding volume etc. I also get why they snap it.

So here’s the thing, I want to add the advanced reverse foot lock setup but if I follow the direction and snap to a toeEnd joint in the joints that are place inside the mass instead of the snap to grid toeEnd, the reverse setup will be in mid air, and somehow I don’t think it will “roll” or “bank” properly nor “toe and heel lift.” SO. if I know I’m going to be adding the RFL system, should I intentionally snap those joints(ball and toeEnd)?

Just curious :).

Thanks!

Darn. No responses? I still haven’t found a solid conclusion to this theory. My concern with not drawing snapped to the grid is that the rotations won’t be authentic, but then again, other tuts, not school based, show the drawing inside the mesh, regardless of how it aligns to the grid. Just seems logical to me if it’s not drawn straight and snapped to the grid. Just curious :).

The bones that do the deforming need to be in the center of mass, more or less. The bones that fake the heel and toe lift pivots need to be on the ground. They don’t need to be built grid-snapped except for convenience - they do need to be built so their local rotation axes make sense (ie, so the heel and toe lifts are clean and don’t require two or three axis inputs to look natural