Control Enforcement of a Pose

[Visping]

I would like to be able to control the Enforcement of a percentage Pose from a on/off Pose. Is it even possible?

 

I am constructing a character who is talking trough the corner of his mouth. Witch corner he is talking trough is varying depending on angle to the camera. I am planning to use Dope Sheet to quickly make the basic lip sync before fine tuning it.

 

I am going for this approach:

• I make two percentage poses for all phonemes, one for each side (Right A I, Left A I, Right E, Left E, Right O, Left O etc.) and form the expressions of the mouth.
• Then I make a percentage Pose for each phoneme (A I, E, O etc.) In these poses I control the poses above (Pose A I at 100% vill set both Pose Right A I and Pose Left A I to 100 %)
• And finally I make an on/off Pose for each side of the mouth (Right LipSync and Left LipSync) This is where I want to control the enforcement of the right and left poses....

The result would be:

When Right LipSync is on and Left LipSync is off my character will speak only from the right corner of the mouth.

When Left LipSync is on and Right LipSync is off my character will speak only from the left corner of the mouth.

And if both poses are on he will speak from both corners of the mouth.

 

By this approach I am aiming for a dynamic way of changing "talking sides" while having the dope sheet to speed up the lip sync.

 

I tried to do this but the Enforcement did not stick when trying to set it in the relationship...

 

Is this kind of control not supported in Animation Master?

If not, do anyone have any suggestions for a workaround to achieve what I aim for?

[robcat2075]

Hmmm.... can you show me a screen shot of where you are accessing the enforcement of one Pose when you are making the On/Off pose?

[Visping]

This is where I try to change the enforcement in a on/off pose

 

[robcat2075]

If i said to turn on "Show more than drivers" does that solve it for you? If not, ask some more here.

[Visping]

I have searched in AM but can't find "Show more than drivers".

My first guess would be in options, but I can't find it... Where can i find it?

[Shelton]

The more drivers is on the pws. It is set of lines and allows you to access groups. I will upload a pic when i get home

[robcat2075]

There is a Show More Than Drivers icon for every item in a Pose, Chor or action

 

Turn it on for the On/Off pose that contains the pose you are trying to set enforcements for.

[Visping]

When opening the Action node I can see that when working with a relationship a temporary action is shown. In it there is a shortcut to the whole model who has this little icon.

Is this the one?

 

[robcat2075]

I'll try and make a video of it tomorrow and then you can see what I'm looking at.

[Visping]

Great! I am looking forward to it

[robcat2075]

I've boiled this down to the bare details to show the essential concept.

 

Here's the animation:

 

CornersSpeak.mov

 

Here's the explanation:

 

clip3621ControllingPoseEnforcement.mov

 

Here's the PRJ:

 

PoseEnforcement07.zip

 

When you first asked about this I didn't think it could be done!

[Visping]

Thank you Robert!

You picked up exactly what I wanted to do and demonstrated it so even I understood perfectly

I have just one question. You mentioned in the end of the explanation that this is probably not the way you would do it. Could you give me the reason for that? It would be good to know if I am heading in the wrong direction not realizing that there is a more obvious solution.

[Rodney]

That was awesome Robert!

(and your project file is the example of simplicity in presenting a solution)

 

While Stefan never suggested it was impossible to do... you managed to sneak an 'It can't be done' video in on us too.

 

Talking out of the side of a mouth (or at least asymmetrically) is such an important aspect of character dialogue that I'm really glad you guys have explored this.

 

You mentioned in the end of the explanation that this is probably not the way you would do it. Could you give me the reason for that? It would be good to know if I am heading in the wrong direction not realizing that there is a more obvious solution.

 

He's got me curious too.

I'm going to guess he wouldn't generally use a dopesheet to drive the lipsync as his approach but would rely more on direct animation of the mouth opening and closing.

In other words, in a similar fashion to how he animated the dialogue in the video without resorting to use of a dopesheet.

[Visping]

Rodney, I am under the impression that the strength in using dopesheet during lip sync is speeding up the pipeline. After the dopesheet has done its work I will have to fine tune the lip sync anyway. Would you agree?

Or do you think using dopesheet doesn´t bring any advantages?

[Rodney]

Rodney, I am under the impression that the strength in using dopesheet during lip sync is speeding up the pipeline. After the dopesheet has done its work I will have to fine tune the lip sync anyway. Would you agree?

 

 

Most definitely.

 

Or do you think using dopesheet doesn´t bring any advantages?

 

That's a dangerous question to ask me.

 

I'm a big fan of dopesheets/xsheets although they've largely gone out of fashion and are often misunderstood.

The primary purpose of these tools in the past was to drive forward the plan for animation whereas now it's largely for automation.

Those that don't use xsheets/dopesheets use other methodologies to organize and move production forward.

 

Automation can be seen as a dirty word in animation circles but isn't that the purpose of computers in the first place? To take the drudgery out of the process.

I'm of the opinion the dopesheet is sorely underused especially when it is considered only as a tool for automating lipsync.

The dopesheet can drive ANY pose... that makes it all the more powerful.

 

Having said that I would also say that the majority of automation should be used just to get the initial staging set.

The use of finer controls and character (personality) performances can then be focused on.

(As you've suggested above)

 

The question one might ask would be whether or not a pose can/will be reused.

Similarly to canned Actions they have their proper uses.

 

Imagine a set (before a character has been placed into it). There is no limit to the number of things that could be automated (as driven by dopesheets) to create better environments, crowds, background characters, etc. Anything that can be 'posed' can be leveraged/driven by a dopesheet.

 

Of course the bottom line of dopesheets is that it's use is largely for early production/preproduction; to get the flow of production moving... economically.

By automating 80% of the more laborious tasks we then can spend the majority of time directly focused on the fine tuned details of most important work.

And here's the imporant point: if any of the work (automated or otherwise) doesn't have a compelling need to be further refined... it's done.

 

Disclaimer: At present Dopesheets cannot be saved by themselves in A:M (as they were in the past).

The workaround/improvement(?) to this is to save the Action wherein you've created the Dopesheet, then reuse that Action as necessary.

That Action/Dopesheet can then be edited in any text editor, automated further or even created wholy in another program.

[Visping]

It´s good to know that I wasn´t so far out after all, Rodney

You´ve got me curious to explore the possibilities of driving other poses than lip sync by dopesheets, but that will be a later matter!

[robcat2075]

My own feeling is that dialog animation is more about the acting performance than the mouth shapes. Compared to the time put into the acting, the time to do the mouth animation is a minor factor.

 

I've never actually tried the dope-sheet method to automate lip synch, maybe it can be done great but, looking at other people's results, it seems to give an over-animated, choppy look that needs to be edited to bring it back to normal and it also has a repetitive look to it.

 

At animation school they taught us to just keyframe motion manually rather than start from something that was automatically generated so that's how I've always done it. It's the last thing I do after all the other animation is figured out.

 

If you can get a cheap copy of "Stop Staring" by Jason Osipa, he teaches that fundamentals approach to dialog. Some early editions even have coverage of Animation:Master but he was using an old version (which didn't have CP weighting) so the details of how he rigs a face in A:M are not necessarily current but the general concepts of sliders and mouthshapes and timing are still valid.

[Visping]

I think I will try my way forward and see what the result looks like. In any case the control of the pose will be useful for me. And as a bonus I now understand how to key properties. It is great to take part of the way more experienced animators reason. Thanks for sharing, Rodney and Robert!

And thanks for the tip! I have ordered "Stop staring" now