An Example of one of our rigs

[daney7g]

The character had real character thanks for sharing.

[jjwinks]

This rig is amazing! Thanks for sharing it. I feel really inspired to learn from it and apply it to my own creations.

[amanandink]

Thanks guys, and any questions just let me know and I'll try and help.

J

[kphgraphics]

Nicely done likes excellent.

[Kosafwc]

And how is it done, that there is a smooth pass between layers in mouth switch layers? I was trying to figure it out, but I just couldn't

[amanandink]

Hi Kosafwc

the mouths are on an interpolation setting for the switch layers, when that is turned on int he Layer Settings the switch layer inbetweens the keys for you as long as you have the same number of points and shapes in each mouth level.
We usually build the mouths as separate levels inside a group folder then once they are approved we flatten them into one layer, by copying and pasting the shapes down. Then use the interpolation of the switch layer to do the inbetweens.

J

[Kosafwc]

Thanks for quick answer. I was tring to do the same by thicking the interpolate sub-layers option, but it never did work. Perhaps I have different point quantity in phonemes layers, but it should have the same, as I always copy phoneme layer and just rearrange it to another phoneme.

EDIT: Made it from the scratch and this time it worked. Perhaps it's magick, I don't know. But thanks again

[funksmaname]

It's not simply about having the same number of points - they must be THE SAME POINTS repositioned so that the program knows which point should go where - so you must duplicate the mouth and then reshape it.

Now we have smart bones - I would be more inclined to use a single mouth layer and create SBs for each position - it also allows you to create an inbetween position for mouth shapes that need it, giving you a less linear automatic interpolation

[heyvern]

Thanks for quick answer. I was tring to do the same by thicking the interpolate sub-layers option, but it never did work. Perhaps I have different point quantity in phonemes layers, but it should have the same, as I always copy phoneme layer and just rearrange it to another phoneme.

EDIT: Made it from the scratch and this time it worked. Perhaps it's magick, I don't know. But thanks again

I create my switch layers on frames of one vector layer first. That way I know they are "perfect" for interpolated switches. I am editing the position of the same layer on multiple frames. When finished I just duplicate the layer as many times as needed and copy each key to frame 0 and delete the other keys. I save the original "template" layer for back up so I can make more if needed.

I started doing this so I could use onion skin to see the different positions more accurately before creating the switch layers.

[Kosafwc]

I did it by simply copying the basic layer and making adjustments to mouth shape.

Funksmaname: I tried this way with smart bones at the begining, but I have a head-turn smart bone in my character. When I was trying to turn the head, my mouth would change their shape in a strange way. I couldn't control the side view with this head-turn bone inside phonemes actions, so I decided to do it old-fashioned Perhaps if I'll upgrade to 9.5 it will be possible?

[amanandink]

Vern thats a great idea about the mouths and doing everything on one layer. The issue we found was the auto weld catching points by mistake when your making the new mouth shapes and this will obviously change the number of points.

Funksmaname I'd be interested in see your rigged mouth setup, with the smart bones, we tried a test with smart bones and moving to the different shapes but it didnt quite work for us, we had to keep inbetweening through the shapes in the middle of the dial. So if we had it set up A B C etc positions and wanted to go from A to C we had to pass through B. We also looked at Slegin setup but we needed to have standardised positions for each animator rather than letting them form the shapes.

J

[funksmaname]

Hey there,
Well, I didn't mean putting A B C together - but A A-B1 B would include an inbetween for A to B which could not be linear (like the 'buh' action in the vid below).

What you could do is instead of creating phenomes, create different control bones for things like 'OO', 'Smile/frown', 'Open mouth' and combine those to make each mouth shape. To make a repeatable mouth shape selection for your animators, save each 'phenome' as a standard action which could be single or multi-frame actions that put the bones in specific pre-defined positions? (essentially, the animators would put down actions instead of switch keys)

This is all fairly theoretical, you'd have to see if it yields good results and work with in a team environment... I'm a lone tweaker

Skip to 0:45 if you're in a hurry. I really should work out how to properly say 'Phoneme' so I don't go 'fffffffffffffffffffffffffffffffffffenome' every time