Page 1 of 2

Actions in non-Pro AS

Posted: Sat Jan 26, 2008 10:26 pm
by synthsin75
I'm sure someones already posted this somewhere, so I'll just ask first. Would anyone be interested in knowing how to build actions in non-Pro AS? If so, I write it up for ya. I can even give pointers on how you'd modify Darthfurby's great 'head turn with actions' tutorial for use with non-Pro AS

Posted: Sat Jan 26, 2008 11:57 pm
by Genete
Would anyone be interested in knowing how to build actions in non-Pro AS?
Yes I am.

Posted: Sun Jan 27, 2008 12:12 am
by Víctor Paredes
me too.

Posted: Sun Jan 27, 2008 12:27 am
by Genete
I bet that it involves manual modification of the anme file.

Posted: Sun Jan 27, 2008 1:10 am
by synthsin75
Okay, here goes nothing... All actions really do for you is remember keyframes and names for them, making them easy to find and reuse. Non-Pro users just need to keep track of where the keyframes are themselves. First you need to estimate how many actions your going to need in terms of frames. Darthfurby's head turns use one frame per angle (just like he keys them in his tutorial), but a walk cycle would require a number of frames for the one action. Now I haven't done a full project yet, so I don't know how hard it would be to estimate. Just as a test, I did a 3/4 right to 3/4 left face turn. This uses two frames, one for each angle.
What you do is move your start time forward as many frames as you need. I would suggest over estimating since it could be complicated to move the start time after you've done some animating.
Now you just use frames before your start time for the actions. I suggest laying out your layers as Darthfurby does, but when you make your actions, do so in a separate layer/s. In other words, where Darthfurby has his actions in a Group layer (also not in AS standard), we'll do ours in a new vector layer. Basically this is where we'll 'group' our actions. In some ways this is easier because you never leave the main timeline.
So I have my initial 3/4 face built on frame 0. I didn't do it for my test, but next time I'll separate the features (eyes, mouth, nose, etc) into their own layers. I'll go frame 1 to make my first action. Just follow the great example in Darthfurby's tutorial. Just be sure to select all of your points (ctrl-A) and right click the selected point key and select 'add keyframe' to insure that all of your point motion is included. Also, you can repeat this after any changes to your action to reset the key.
All you'll need to do is keep a list of which frames (before start) contain which actions. In my test, frame 1 is my 3/4 left and frame 2 is my 3/4 right turn. To use these, I just need to copy and paste them into my active timeline.

I'll be happy to clarify any of this if needed. Also I haven't done enough with switch layers yet to know if there are any issues.

Edit: I forgot to mention that all of the layers (nose, eyes, etc) are copied into the one action vector layer. This way all of your actions are on one layer that can be easily found. You just have to sort the shape depths.

Posted: Sun Jan 27, 2008 1:17 am
by synthsin75
I bet that it involves manual modification of the anme file.
If only I had any clue as to how to do that.

Posted: Sun Jan 27, 2008 1:20 am
by Víctor Paredes
synthsin75 wrote:
I bet that it involves manual modification of the anme file.
If only I had any clue as to how to do that.
just open the file with notepad or similar :wink:

Posted: Sun Jan 27, 2008 1:22 am
by synthsin75
just open the file with notepad or similar
I mean after that. I keep flirting with the idea of learning scripting, but my main focus is the art.

Posted: Sun Jan 27, 2008 1:24 am
by Víctor Paredes
synthsin75 wrote:
just open the file with notepad or similar
I mean after that. I keep flirting with the idea of learning scripting, but my main focus is the art.
ouch

Posted: Sun Jan 27, 2008 1:25 am
by Genete
That's a great tip! :D
But you can call it 1/2 actions.
You loose:
1) Multi layer action ability (including layer z depth animation, very important on DF head turn technique)
2) Action references. When you insert a action reference you can modify the action and all the references change. With the store keyframe technique don't.

Anyway a good point for non-PRO AS users. :D

-G

Posted: Sun Jan 27, 2008 1:37 am
by synthsin75
ouch
No offense intended there.
1) Multi layer action ability (including layer z depth animation, very important on DF head turn technique)
I'm still new to this, but I thought that putting it all on one layer and sorting the shape depth would take care of this. Is depth sorting of shapes not an animated feature?
2) Action references. When you insert a action reference you can modify the action and all the references change.
Yeah, I knew that not having instances would be an issue, but it's not that hard to just recopy the action. Well, depending on how often it's been used. I guess there's never any substitute for good planning, huh?
Edit: Okay, just tested the sort shape, and your right. It is not an animated feature. I would still assume that clever use of DF's 'vector putty' would do the trick. Yeah, I guess 1/2 actions is appropriate since you'd probably have to separate the layers for any real depth sorting. :(

Posted: Sun Jan 27, 2008 1:52 am
by Genete
synthsin75,
unfortunately shape sort is not an animate feature.
-G

Posted: Sun Jan 27, 2008 1:56 am
by J. Baker
Genete wrote:synthsin75,
unfortunately shape sort is not an animate feature.
-G
But it will be whenever we get the next update.

Posted: Sun Jan 27, 2008 5:55 am
by heyvern
J. Baker wrote:
Genete wrote:synthsin75,
unfortunately shape sort is not an animate feature.
-G
But it will be whenever we get the next update.
Animated shape sorting is not in the promised update. Animated layer sorting is. I only point this out because both have advantages but are quite different.

Both can be done using scripting right now but having it in the application will be much better.

-vern

Posted: Sun Jan 27, 2008 6:07 am
by J. Baker
You are correct heyvern but if you create a single shape per layer, you pretty much have the same thing. :wink: