Wednesday, November 05, 2008

OK, let's try this instead...

Getting back up on the horse. Ahem.

Anyhow, following up on that last (embarrassing) post, here are some sequential drawings to check out for a few movements in the Sylvester clip. The first one is kinda neat- his tongue is completely in and then the next drawing is completely out- and not just out, but down to his belly out! Check it...
 
 
Each of these drawings were held for 2 frames. I think the reason you can get away with this is that the rest of the body isn't changing quite so drastically. It's changing, but the general silhouette is basically the same.

Now we get to the happy feet stamping. Again, back to back drawings held on 2's. First, the right foot is down and the left foot is up.
 
Next, the right foot is up and the left foot is down.

No breakdown or inbetweens. Just extremes. I've tried this in Cg and it's really hard to make it look as good as it does here. One thing I learned is to not cycle between two leg position/shapes that are exactly the same, but to keep it varied as the legs hammer back and forth like that. The dry brush is the real key, though. That's what I think, at least. But what do I know? I do know that motion blur doesn't work as a dry-brush replacement.

And now for giggles, we can check out how he switches from the foot stamping happiness to the scramble run off screen...
 
  
  
  
That's crazy stuff. Those 4 drawings are back-to-back-to-back-to-back in the scene, each held for 2 frames.
And this is a 'simple' scene. The stuff with him emptying out the drawers just before this is flat out amazing. Go watch the entire short. "Canned Feud". It's on the Loony Tunes Golden Collection DVD set, first volume, disc 4. The entire short is just jam-packed with gold like this. Who's doing stuff like this these days? Who in CG animation even thinks like this?

13 comments:

Blake said...

I found the exact clip online:
http://www.cartoonthrills.org/blog/Friz/49CannedFeud/1SylvesterScramble.mov

Great Post Keith. I really love this shot. Timing in the clip is amazing a really is fun to frame by frame it. I think it's sad that we can't do this in CG without the outcome looking mechanical.

Thom said...

I'm surprised you didn't mention the drawing where a duplicate top of his head floats in the air. Or is that a video interpolation error too?

Keith Lango said...

Yo Thom,
No, it's not a artifact- that floating head is in the original. There's also a slight ghost of his ears in the following drawing (the full squash in the antic). It's an interesting use of ghosting in the scene because the other ghosting in the short is a dry brush where-as this is a piece of the previous cel left in place. I'm not sure it's as successful as the dry brush approach, though. It does strobe a bit where the dry brush doesn't.

Keith Lango said...

Blake-
Motion-wise, I'm finding that you can do this stuff in CG. It just requires a different mindset, that's all. But I've been learning and experimenting with this stuff on my own and teaching it to some students with pretty cool results. But I do admit that if you try to animate this kind of stuff with the same approach you'd use to do a more typical CG Pixar-ish style of motion it's not exactly an easy fit. The one thing that I find really helps is if the rendering style is more graphic and not photorealistic with subtle shadings and shadows and highly detailed textures, etc. This style of animation pretty much requires simpler, less sophisticated rendering.

Mike L said...

Keith,
This type of animation is what I entered the industry for in the first place. I love the loose and full feel of the animation as compared to 3d.

Here's something of a side topic...

Having finished a Bachelor's degree in what can be summed up as 3d "stuff" (certainly a generalist style degree), I am now preparing to study at a well-known online animation school. Mostly so I can do this sort of thing specifically (hindsight certainly is 20/20).
When I set out to do the degree and even this upcoming endeavor... the basis for my decision was that 2d has a very minimal future and is a very tight community with little chance of getting into.

Days older and ounces wiser... I'm seeing that 2d isn't dead but it probably is hard to get into. Now seeing this post and having been following John K for a few weeks (along with cartoon brew) I don't know...

You said previously that 2d/3d.. it's all ANIMATION... which is great... but now it sounds like to do animations similar to Warner Brothers, Hanna Barbera, or even John K himself, you have to go 2d.

Sorry that goes on with no clear organization but I hope my conflict and confusion is well-expressed. The schools that are "preparing" us for work in this industry don't really talk about this stuff...

Thom said...

My guess is that the floating head was an experiment that they decided didn't work that great (since it was never done again, AFAIK). I think it works better with eyes and I'm pretty sure this same cartoon has some floating eyes somewhere.

Even if it doesn't work, I love that it's in there. Can you imagine a director letting that stay in today? At least in CG?

Andreas said...

Mike i dont see why 2d would be hard to get into (more than 3d is) or that it has a minimal future.

The thing is that if you know how to draw 2d animation is easier. Its harder to get weight and all those principles in a software that sometimes has a mind of its own thats why i think Keith is saying that 3d requires to have a different mindset. As far iam concerned if you know how to animate you know how to animate 2d,3d,flash or whatever.

Keith Iam not sure i understand why the drawings would be held for 2 frames. I have seen it in other animations too and i was always wondering how you can get away with it i guess not having the pose change is one way but what really are the benefits for doing something like this?

mike l said...

If you guys haven't seen Jason Ryan's ramp-up tutorials on his site, they have a few examples showing how he mixes up timing to change the mood/pace. He actually does jump between 1s,2s,3s,and 4s for various reasons. I won't post the url unless Keith asks/says it's okay but it does seem to cover the exact thing being described.

Andreas,
Thanks for that. I love to draw... it's what led me to where I am..and I never get to use my love for drawing outside of sketching for storyboards or possibly thumbnailing (my non-animator boss doesn't understand the importance though so I often don't get to do this). One of the biggest drawbacks I've seen for 3d is the rig... at best it's still much more like stop-motion than hand drawn animation :/

Gary said...

No breakdown or inbetweens. Just extremes. I've tried this in Cg and it's really hard to make it look as good as it does here.

Build twelve individual Sylvester models for each second of animation, then use them as an object replacement sequence. That's what gives cartoons their lively look. The contours are moving, shimmering, strobing all over the figure. A belt buckle, for example, is not a solid object; it expands and contracts somewhat according to the action occuring in the waist area. And its contour is shimmering all the while through those changes.

Lively all over. A belt buckle. That's 2D. Try that in 3D photoreal and it looks like a bad acid trip. Cartooning requires NPR rendering.

Keith Lango said...

Mike-
Jason and I are friends and we definitely support each other, so there's no worry about putting up a link. In fact, here is is...

http://www.jasonryananimation.com/main.html

If you have the spare change I think it's well worth the money to buy and watch his tutorials. I have enjoyed watching his tutorials and seeing how he approaches his animation. I always learn a little something new each time. I'm certainly very far from knowing everything and I take every advantage to learn from others. :)

As for 2d, I think there's still a great demand for it. Maybe not in feature films (in the US) but in so many other areas of media it's still used a lot. If you have a desire to get into it then definitely get into it. I can't imagine that you would be worse for the experience. :)

Keith Lango said...

Andreas-
Re; the 2 frames per drawing. It pretty much comes down to two things- Budget & workload. You cut your work load in half, cut your clean up needs in half, cut your inking and cel coloring needs in half. Cleaning up on 1's is a lot of work. It's more work for everybody, from the animator on down to the cameraman.

The thing I like about it goes beyond workload savings. I think reduced frame rates (to a point) allow more freedom in the expressive range of the motion. Since it's less tied to the real in terms of frame rate, it allows less than real in terms of motion and physical ability. When time is less literal then motion and action can be, too. But like Gary points out, you need to be consistent in your approach, so the visual rendering needs to be less literal as well.

Anonymous said...

If there's any blender users that're good at making a plugins, what about using a particle system to simulate that "dry brush" technique instead of motion blurs?

billburgNYC said...

Yo Thom,
No, it's not a artifact- that floating head is in the original. There's also a slight ghost of his ears in the following drawing (the full squash in the antic).


I'm a little late the the discussion here, but I wonder whether that ghosting is the result of DVNR when Warner's "cleaned up" the cartoons for dvd.

Thad K. wrote about this a couple of years ago and posted some comparison frames between VHS/laserdisc releases and dvd releases with DVNR (search for DVNR on his blog).

As you observe, Keith, this kind of ghosting is quite different (and less successful) than dry-brushing. That strenghtens my suspiction that this was not the artists' intention. Wish we could know for certain what those original frames looked like.