Drowtales 2007 anniversary animation - segment done

slowtiger · Post by **slowtiger** » Fri Mar 30, 2007 10:06 am

White lines: I stumbled over this in my project too. Shared points of stacked shapes don't seem to be rendered exactly enough, or with no reference to each other. I just worked around that with slightly enlarging the top shape, or moving its points a bit outside - done in highly enlarged view, of course, so it's only one or two pixels in "real" distance.

My advice would be to create everything stacked with a good overlap of shapes - this also makes selecting them much easier.

But you don't need much advice - the animation and everything is flawless!

Genete · Post by **Genete** » Fri Mar 30, 2007 3:31 pm

Touched wrote:Here's a little progress update:
http://drowtales.com/~whiteradish/temp/ ... 3-7all.avi

Faen's still not in it, but Ariel acts some more.

I want to see the outtakes of the filming!!!

Good luck!

Genete

Touched · Post by **Touched** » Sat Mar 31, 2007 8:55 am

Thank you guys.

We actually had mentioned outtakes, which would be much easier this time since I'm using Anime Studio now, but time's so short I don't think it'll be possible. And I have some flubbed lines from the actresses, too! Too bad I won't be able to use them.

DarthFurby · Post by **DarthFurby** » Sun Apr 01, 2007 12:28 am

I would've loved to see this in vector format. Newgrounds would too.

Touched · Post by **Touched** » Sun Apr 01, 2007 12:59 am

Unfortunately, it just wasn't practical to continue doing this with SWF output in mind. With all the disappearing shapes I get on import to Flash, it would have taken far too much extra work. I'm going to try putting it on Newgrounds anyway, but as a FLV embedded into a SWF.

Plus, in a part I haven't shown any work from yet here, I'm using a number of raster effects and blurred particles that wouldn't be able to export to SWF.

I'll do as much output as I can in vector for other things, since I prefer vector myself, but it's all a matter of compatibility.

slice11217 · Post by **slice11217** » Wed Apr 04, 2007 10:03 pm

Touched wrote:However, you can see one of
the tricks I used for the head turn in the source file I uploaded for my contest entry. Here, though, it's a combination of bone movement and point animation, as is the body rotation. It'll take a while to explain, but the brief explanation is that:
There are no switch layers except for the mouth -- I'm controlling everything with actions and on-the-fly keyframes.

The body turn with the arms changing from front to back and vice-versa are 2 copies of the arms (since I'm using separate bone layers for each arm and the body) with visibility swapping.

I'm using Vern's master/slave bone control script to keep the bones manageable.

The hair is dynamic bones.

Hey Touched,

I've been following the thread started by Darth Furby in addition to this where he also did a head turn. On that thread, you mentioned that "actions are indispensable for this sort of thing." I've looked over his project file and I'm not really sure where he's using an action, nor am I sure that, given the actions he has, how it's indispensable. Can you elaborate on how actions are used to make a head turn? Also since you mentioned it, can you post a link to the thread you reference above with the source file for your contest entry?

By the way, I like the animation you did for this project. Since I work on Kappa Mikey and the show's based on anime, I was really interested to see your movie. Nice work.

Thanks,

Slice

Touched · Post by **Touched** » Wed Apr 04, 2007 10:19 pm

Sure, first here's the thread with the source file:
http://www.lostmarble.com/forum/viewtopic.php?t=7681
Or just the direct link to the source file:
http://www.stormvisions.com/moho/contes ... _shawn.zip

Now, I haven't looked at Darth Furby's source file yet, I just saw him mention that he's using actions for it. Actions don't help with doing a single head turn, but in a long project where you may need to do it again, that's where they come in handy. I guess I should have said "Actions are indispensable for project management of things like head turns". Because I'm working with a bunch of vector layers and separate bone layers, and Actions allow me to key as many of those layers as I choose. I'm making some complicated rigs, and Actions make them more manageable.

BTW, I'll post an updated video either tonight or tomorrow. I've got Ariel's part done, and I'm working on Faen's.

DarthFurby · Post by **DarthFurby** » Thu Apr 05, 2007 2:41 am

I'm using single keyframe actions. That's it. Nothing too fancy.

Touched · Post by **Touched** » Thu Apr 05, 2007 3:08 am

Same here. Single keyframes are much more flexible since you can make the tweens as long or as short as you want on the main timeline.

slice11217 · Post by **slice11217** » Thu Apr 05, 2007 5:57 pm

Touched wrote:Same here. Single keyframes are much more flexible since you can make the tweens as long or as short as you want on the main timeline.

So why wouldn't you just use point keyframes to accomplish this? Is it simply because you can drag-and-drop the two positions and then slide them along the timeline?

Since tone can get lost in text, I just want to clarify that I'm not trying to challenge your judgement, but rather to understand why you did it.

By the way, I've potentially got a freelance gig coming up this weekend where I may need to do a head rotation so I may be asking some more questions soon.

Touched · Post by **Touched** » Thu Apr 05, 2007 7:14 pm

Because I have the parts of the character's head split onto a bunch of different layers so I can take advantage of the layers' Z-depths to move parts in front of or behind other parts, and so I would have to go through each of those layers adjusting or copying the keys if I weren't using Actions. I can't slide the two positions along the timeline for the same reason -- there are 14 separate layers that are affected in my head turn, without counting the 10 mouth phonemes which are also affected. Actions keep that from being an organisational nightmare.

macguffin · Post by **macguffin** » Thu Apr 05, 2007 7:32 pm

Touched, I'm a week into experimenting with moho and have been looking at your what the hell2 animation with great interest.

I've got the hand of basic bones - stick onein an arm and bend the arm etc. I've used verns master/controller layer to make a more complex rig manageable.

However your set-up has baffled me and I'm sure I'm missing something really really obvious.
In the mailman head layer you have several offset bones that are effecting the face and ears to achieve your head turn. How are these bones associated with the layers?

How are different degrees of rotation applied to your face and ears layers from the same bone?

Please forgive me if this a stupid question.

Cheers

Matt

Touched · Post by **Touched** » Thu Apr 05, 2007 7:49 pm

That one was a mash-up of a few experiments that happened to work in that case. That kind of distortion isn't suitable for every kind of mass, and doesn't work for the anime style I'm using here. The way it works is that I have one big bone as the anchor in a flexible-bound layer, and a few other bones around it for adjustments. Then I have that one small bone originating in the middle of the big anchor bone, and when I move it around, it warps the head in a way that resembles a turn. Then I have several other layers that are region bound instead of flexible bound, for parts that aren't supposed to distort like that. It's basically my very simplified interpretation of the effect Vern is achieving through his bone rig, though not nearly as easy to use nor as flexible.

Touched · Post by **Touched** » Fri Apr 06, 2007 4:04 pm

Here's another substantial update. Ariel's done, and Faen's close to halfway done. Fortunately, she's more gentle and shy and doesn't need to move as much. While working on her I tried out that command that Keisern brought to my attention which I had totally overlooked -- Copy Keyframe. Perhaps because it was in the Animation menu, and I expect to find "copy" commands in the Edit menu. Anyway, it worked very well, though unless I chose "entire document" it would only copy the keyframes of the current layer, and so I end up with a bunch of channels I don't use much getting keyed too, but for that same reason I only had to delete them from the visibility channel.

So that command takes some of the burden off of the Actions which we were discussing before, though it's still more convenient to store certain things in Actions, especially if I want to use them in a separate document.

http://drowtales.com/~whiteradish/temp/ariel&faen2.avi

slice11217 · Post by **slice11217** » Fri Apr 06, 2007 7:29 pm

I very quickly did a simulation of what DarthFurby did with his project and I can see the benefit of doing rotations using single frame actions, if only to keep things organized and to keep your head from exploding with confusion. I used to merely copy keyframes from earlier in the timeline but now I may just use single frame actions.

Thanks guys!

Only one thing though: Touched, your action palette must be packed with layers like, "head 3/4", "head profile", "mouth front", "mouth 3/4", "mouth profile", etc. But I guess that's easier than trying to remember what frame the keyframe for a 3/4 head is!

Does it also mean that if you do an action for a 3/4 head that you then have to go to the mouth layer and plug in the 3/4 mouth and then 3/4 eyebrow, etc., or can you set it up so that the action spans multiple layers?