Duplicate keyframes in a different layer?

[synthsin75]

As far as I've read this just isn't possible in AS. I've tried to copy from actions, but actions in a group only apply to each layer separately.

Vern, I know you like to do shape order switching on one vector layer, so maybe you'd have some idea. The only thing I can think of is just keep duplicating layers within my shape order switch. I may end up with a lot of vector layers in the switch layer, but should be able to switch what I need.

The only drawback is that I'll have to redo all the shape ordering for each layer.

Also... I'm trying to do a full 360 degree head turn using this shape order switch so I don't see any way to use a switch layer for lip-sync. (At least not without going insane,i.e. a self masking switch layer, or something) I don't think it'd be feasible to use actions either. So I'll have to do it manually with bones, I guess. I'm assuming I have to use the bones for most of the mouth rotation. Any better ideas, let me know.

Of course, I could be making all of this much more difficult than it needs to be.

[heyvern]

Not sure what you mean by copying key frames to a different layer, but no, it isn't possible. Key frames are "stuck" on the layer. If you duplicate the layer you duplicate the keys.

The shape ordering you are trying do with a switch layer might be more trouble than it's worth. There might be so many combinations to keep track of. Try to keep it simple though. If you can figure out how many shapes at a minimum need to reorder during the turn you could keep the layer number down. How many and what shapes are you reordering? If it isn't too many, Genete made a cool script that reorders shapes in a layer based on bones and shape names.

You have run into all the reasons I now use complex bone rigs and scripting for faces and heads. By using bones to control EVERYTHING I can do lip sync from any angle using actions based on bone translation.. By using custom scripting I have layers "flip" (ears for instance) automatically as the head is turned.

I don't use shape ordering in a switch for more than hands at this point. And that is just two layers, front and back.

Ultimately, I am very lazy. I like all my keys to just be bone keys... and as much on one layer as possible.

-vern

[synthsin75]

Yeah, I kinda figured as much. I have yet to see any real solutions for a complete 360 degree complex (true to life) head turn anywhere. ...And I've probably got too many shapes for a script/bone solution to handle. I only have four layers in my shape order switch, but I'm wanting to also do a tilt up/dn 360 as well, which will make it twelve layers.

[heyvern]

Something I don't think a lot of people think about is... do you really need a complete 360 turn?

My rig doesn't do that and it works pretty darn good. I get close to a complete side turn. If I need a profile or back turn I do a "one off" version for that. I may see about doing some turns on a side view to split the difference.

If you watch a lot of cartoons how many times do they do a complete turn? When they do is there a cut between the scenes? Something to think about anyway.

-vern

[synthsin75]

Yeah, I know. One day I'll learn the 3d apps I've got. But until then, I did a profile right to left turn, and once I had that it was just as easy to go on around. And since I'll have to start hiding features anyway at the profiles, I figured I might as well do the hiding for the full turn.

I have simplified it a bit. I just have to go back and redo all the shape ordering after I've done all of the actions. So only like four layers in my shape order switch layer.

One question. Is the only way to do separate but additive motions, like lip-sync added to a turn, to use point motion for one and bone motion for the other?? Anything I'm missing there?

[heyvern]

What I did in the past when using switches for lip sync and switch head turns was to not have the character talk during the turn. Then you just have an alternate mouth switch of the view you want.

I would have a default mouth position in the head turn. Hide that head turn switch layer and show the "regular" head and the lip sync mouth switch. When the head turns to a different view reverse the process. You can load in dat files from papagayo for lip sync into both mouth switch layers so they always match up.

Now days I use all bone motion for everything so my character can speak while his head turns. It's a completely totally different method though that doesn't use switch layers at all.

This brings up another question I have, how many layers or shapes do you need to reorder on the head? In my setup I only have to change what layer the ears are on. Everything else stays on the same "level" for turns from one side to the other.

-vern

[synthsin75]

This brings up another question I have, how many layers or shapes do you need to reorder on the head?

Huh, about twenty shapes, but that includes nose and ears which don't generally need to be animated. (I'm doing a full wireframe to avoid any appearant 'tricks'.) I did a version of this using separate layers and z-depth animating, but I just couldn't get over having all those 'connected' shapes divided up like that. Right now the only glitch in the full turn is the ears. Right before the back view of the head you see the front of the ear on the back of the ear, but it's very brief and I plan to hide it with hair. I could fix it if I need a bald character (just add another layer in my switch), but I'm trying to keep the keyframes confined to one frame per view so I can copy out individual angle actions. I plan to copy the full turn (or a partial) from the action, then delete what I don't need for turns through multiple angles.