Help - Search - Members - Calendar
Full Version: Katt's Bootcamp - Born-again Newbie
Hash, Inc. Forums > The A:M Exchange / Resources / User Groups > User Groups / Projects > Animation Boot Camp
kattkieru
I come and go with AM every few months, but I need to really start buckling down and getting it under my belt for an upcoming project, so I thought I'd go from the very basics and claw my way up. I'm doing these mostly because I want to be able to make that mental shift to the AM workflow as needed, but also to brush up on my animation skills (and possibly to create a bit of demo reel content).

Anyway, first up: The ball bounce. How exciting! I should also go finish off my car model. happy.gif;
martin
Looks like you got that down.
John Keates
I'm not so sure... it seems odd to me that the ball stretches at the top of the motion, almost like it is acting under its own volition a little too much. I would have it compress to a sphere as it gets to the top maybe with some wobble in there.
kattkieru
@Martin: Thanks for stopping by! ^o^/

QUOTE(John Keates @ Oct 5 2006, 04:23 PM) *

I'm not so sure... it seems odd to me that the ball stretches at the top of the motion, almost like it is acting under its own volition a little too much. I would have it compress to a sphere as it gets to the top maybe with some wobble in there.


I agree. I tried to fix it before but the way the boot camp ball is rigged (with the ball stretching from its base, and not from its center) attempts to fix it have made things worse. The ball tends to hover in space for a moment. I might go back and see if I can change the rig a bit or alter the motion so that it is less stretched at its zenith (I reckon I could do it by adjusting the position from the ground creatively), but for now I figured I could move on ahead. Thanks for the advice!
robcat2075
Hi Kattkeiru, welcome back to the boot camp!

QUOTE(kattkieru @ Oct 5 2006, 03:42 PM) *

I tried to fix it before but the way the boot camp ball is rigged (with the ball stretching from its base, and not from its center) attempts to fix it have made things worse.


Remember that you can key that ball to be anywhere you want on any frame you want. Don't accept what the rig gives you if that isn't right.

However, since bouncing ball squash happens because the ball has hit the ground, having the ball squash from the base is the best way to go.

On a general note though, not every rig can do everything. There are some parts of character rigs (like the hips) that automatically give wrong motion. You have to compensate for that.

Some more detail on the movie

[attachmentid=21136]



kattkieru
QUOTE(robcat2075 @ Oct 5 2006, 05:43 PM) *

Hi Kattkeiru, welcome back to the boot camp!

Remember that you can key that ball to be anywhere you want on any frame you want. Don't accept what the rig gives you if that isn't right.


I hope you didn't think I was making excuses. happy.gif The reason it was weird was totally how I keyed the ball. It was just an issue of not understanding the rig. I came up with a better way of working with it for the second animation.

I *really* appreciate that you drew on my quicktime; I'm a visual learner and you just taught me a load of good points in a handful of frames. Seriously, thanks!

Now, this current video I did before your post, Robcat, so it tries to fit in the suggestions of John, but the issues you noted are still there. I've done more than a little animation at this point, but I've never played with squash and stretch, ever, really, so that part's all new to me.

I think I'll move on to the next one, even though this one is imperfect. I like the weight differences, but the squetch is still not quite right, particularly on the heavier ball. I also will be incorporating your notes into the next ball one I do.

Thanks! ^o^/
kattkieru
Wait, one question: what does, "bounce two not alive balls of different materials across the screen (watch your arcs)" mean, exactly? Similar to what I've been working on, or ...?

I would guess that an alive ball moves of its own accord but a not-alive ball moves only by inertia?
Dhar
QUOTE(kattkieru @ Oct 6 2006, 08:12 AM) *

I would guess that an alive ball moves of its own accord but a not-alive ball moves only by inertia?


You guess right.
kattkieru
I figured on turning my balls from #2 into the balls that travel across the screen, so I went back and redid #2's squetching. I think I did a much better job this time, although they might be squetching a bit too much. Not that the task says exactly *how* heavy or rubbery they are.

Question: with "live" balls I could see the angle of the ball rotating a bit -- kind of like how a person jumping across a chasm sort of leans forward during the jump -- but non-living balls keep a steady up vector, right? In other words they just sort of bounce the same way as I've got them doing now, but with motion along another axis?

Next trick is moving them across the stage. But first: pretending to sleep...
Dhar
The timing looks good, but the ball needs to stretch out again after it leaves the ground. Only one frame of squash needed and that's when it's on the ground.

As for your question, whether the ball is live or not, the direction of the jump will also determine the direction of the squetch.
kattkieru
QUOTE(Dhar @ Oct 17 2006, 09:09 AM) *

The timing looks good, but the ball needs to stretch out again after it leaves the ground. Only one frame of squash needed and that's when it's on the ground.


You know, I had it like that before uploading and didn't like the way it looks. Here's a version with it put back. Are you sure it's better?

QUOTE
As for your question, whether the ball is live or not, the direction of the jump will also determine the direction of the squetch.


Okay, cool. Thanks! That's the next thing to try out. I've also already modeled a pixar lamp, but it's not rigged yet. Thinking about the best way to get it to stretch, although I might cheat and just stretch the entire object.
Dhar
The squash can only happen when on the ground, never when off the ground.

The basic idea is the ball stretches towards the ground, then 1 frame fully stretched when it first touches the ground, next frame is squashed (still on the ground), next frame stretched and off the ground on its way up. (The Animator's Survival Kit, p.92 - I think).
kattkieru
QUOTE(Dhar @ Oct 17 2006, 12:09 PM) *

The squash can only happen when on the ground, never when off the ground.

The basic idea is the ball stretches towards the ground, then 1 frame fully stretched when it first touches the ground, next frame is squashed (still on the ground), next frame stretched and off the ground on its way up. (The Animator's Survival Kit, p.92 - I think).


I need to get that book. I'm just so broke right now. happy.gif;

Okay, tried again. I guess I had to play with it a few times before I understood, if I'm understanding now. I don't think robcat's video sunk in before. How's this clip?

And if I haven't said it enough times yet, thanks again to everyone who's looking. happy.gif
Dhar
Much better. But more importantly, what do YOU think? smile.gif

One last tweak, put no more than 2 frames on the ground; one when the ball first touches the ground (stretched), and one when it is squashed. After that it should be off the ground.
Jeetman
It's looking really good Kat!!!

I'm really getting that feeling of different materials. The blue ball denotes a feeling of a soft pliable thin rubber ball while the red ball is a harder and thicker rubber. I get a really good sense of weight.

I agree with Dhar about the framing (I have the Animator's Survival book also biggrin.gif ). A great resource that is a must have especially for A:M users is Jeff Lew's DVD. If you haven't seen it here's a link:

http://www.jefflew.com/


Again, great job!!!

George
kattkieru
QUOTE(Dhar @ Oct 17 2006, 03:31 PM) *

Much better. But more importantly, what do YOU think? :)


Hehe, good question. Feels kind of odd still, to be honest, but that's probably just me needing to get used to the visual language of squetching.

QUOTE
One last tweak, put no more than 2 frames on the ground; one when the ball first touches the ground (stretched), and one when it is squashed. After that it should be off the ground.


Done in this new version, but to be honest I can't see the difference between it and v4. Maybe it's a little snappier? Anyway, I gotta do that one with motion across the stage now. ^_^;

@jeetman:

Thanks! ^o^/ I actually have a copy of his DVD around somewhere. I've been meaning to rewatch it now that I'm using AM again because it's a really good example of using AM "to the fullest" as it were.
kattkieru
Went ahead and tried the move across the stage. There are two versions. v1 is just a simple motion across with no attempt to match the curve of the ball across the stage. (No rotation.) v1a is the same animation with slight rotations about the x-axis. I uploaded both because I think v1 looks better but v1a is what was suggested.

Side note: same compression as before, but these seem to playback jerky on my machine. I'm trying to figure it out. If they playback on other people's machines then I'll just upload the full-side version; they're only 500k each.
Jeetman
Hi Kat,

If you think about the illusion you are creating by stretching and squashing, you'll come to agree why 1a is a better example.

when you stretch you're over exaggerating the movement within the path something is moving. For example: If you were to animate a character doing a broad jump, you might consider a squash as the character sinks down for the maximum thrust and then stretch out just as the character leaves the ground.

Now in your ball example the ball is coming in diagonally. If stretch is only an illusion to exaggerate the movement along the direction it is travelling, then you should stretch the ball in the same direction as it travels or you'll lose the illusion.

In your second ball example (v1a) something didn't seem quit right. When I scrubbed the animation I saw why. If you look at the impact of the ball as it hits the ground, it's not squashing correctly. It's moving too much to the right.

If you center the squash of the ball in relation to the angle, then it will look much better.

I'm referring only to the first bounce of v1a. The rest look good.

George

Dhar
George hit the nail on the head when he said "illusion". That's what we're all attempting to do here, is to give the "illusion" of motion. If you were to follow the laws of physics exactly, you'd have the most dead animation. But it's those illusional exaggerations that bring the animation to life - Thus Disney's animation Bible: The Illusion of Life smile.gif

I noticed in v1a that the second bounce doesn't even have a squash which may be adding to the brake up - or it could be I'm missing that frame.

You're pretty close on this one.
Jeetman
QUOTE(Dhar @ Oct 18 2006, 02:08 PM) *

George hit the nail on the head when he said "illusion". That's what we're all attempting to do here, is to give the "illusion" of motion. If you were to follow the laws of physics exactly, you'd have the most dead animation. But it's those illusional exaggerations that bring the animation to life - Thus Disney's animation Bible: The Illusion of Life smile.gif

I noticed in v1a that the second bounce doesn't even have a squash which may be adding to the brake up - or it could be I'm missing that frame.

You're pretty close on this one.



Good eye Dhar, I missed that. I agree. If you add a squash at the bottom of the second bounce, it will look better IMHO.

Also, the suggestions I offered are just that. There IS no "total" right or wrong in animation. I think this is the point Dhar is eluding to.

That's what I think makes animation so exciting and frustrating all at the same time LOL. You can break the laws of physics. You can redefine logic. You can create a completely different world. The challenge is relaying your vision (or idea) to the audience in a way that they all can perceive and understand it.

George
kattkieru
QUOTE(Dhar @ Oct 18 2006, 02:08 PM) *
I noticed in v1a that the second bounce doesn't even have a squash which may be adding to the brake up - or it could be I'm missing that frame.


Yeah, I finally got it compressed properly. The squash *was* there, but that frame seems to have been dropped?! This one looks better. For some odd reason, despite my choosing 30 fps it was compressing at 24.

I also changed the first bounce a bit. It's very slight, but to my eye it makes it read a little better.

QUOTE(Jeetman @ Oct 19 2006, 08:03 AM) *

Also, the suggestions I offered are just that. There IS no "total" right or wrong in animation. I think this is the point Dhar is eluding to.


Very true. I think that everyone will see things a little differently, and that in the end you have to make a performance your own. No one can ever please everyone. happy.gif Me, I just need to work on the basics that most people accept as writ.

Attached is a new version of 0103.
Jeetman
That slight adjustment to the squash definitely helps IMHO.

Looks good to me. Great work!!!

George
kattkieru
QUOTE(Jeetman @ Oct 19 2006, 11:49 AM) *

That slight adjustment to the squash definitely helps IMHO.
Looks good to me. Great work!!!


Yay! Thanks! ^o^/ Glad that's over with. If I never see another ball it'll be too soon.

...

So on to balls with tails! cool.gif

It's amazing how much my perception of these things has changed just doing a few excercises... now when I do the first pass on a ball it looks terrible until the squetching is in there. ^.^ I think I got the intended rig usage sussed (the squetch control is hidden by default in the downloadable model, btw, which was a bit confusing), but I was having trouble with the tail motion. So I flicked one of my ties around for a bit and then sat down and thumbnailed it in Toon Boom.

I'm just uploading the (very very) short clip in case it helps someone else. I'll be back to post the Real Thing later.
kattkieru
Alright, onto the "real" animation. I really like the first bounce or two, but the settling at the end strikes me as too stiff. Think the third bounce could be a bit different too; I guess this is a first pass. I've actually been considering tying a tail to a tennis ball, but not today. happy.gif;

[attachmentid=21597]
Dhar
The bounces are fine. I think the stiffness comes from the lack of variety in the tail's movements. Let the tail flail uncontrolably and see if that doesn't "loosen up" the model.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2012 Invision Power Services, Inc.