Every animation book starts with the Bouncing Ball, claiming plenty of the basis of animation live within that exercise. So, can anyone imagine a better way to start learning animation than making a ball bounce around?
Most of these animation tests were made during 2010. Some are (ugly) pencil tests made by hand or using the software “Pencil”, but most of them were made using Blender 2.5x.
Besides learning a lot on rigging, I learned that animation takes a HUGE amount of time, and that is always a good idea to animate after the music was set and not the other way around! (at least if you want to sync the animation with the audio).
Oh! And by the way, the rig was based on JiriH excellent Simple Ball Rig (Link: http://blenderartists.org/forum/showthread.php?t=123001). But with the difference I did not use Bart’s script (because just animated 2D movements, so a simple ‘Transform’ constrain was enough for setting up auto-rotation).
Nice work, -from that last part, I’d imagine if you added something like tracking eyes to the balls, a simple nice story with a nice plot, some props etc, you could make a fine shot film;)
Nice balls! No mo…but do you mind if i ask you ALOT of questions?
But for now, just one, How long did it take you to learn…i have been able to do ball bounces but they have been lacking something.
I need to figure out what that something is.
Hi! I took a look to the animations you’ve been making… good stuff!!! I think I’ll ask questions to you too!!! Besides, you’re getting help from Waylow and Beorn!!!
Regarding your first question. I’m just an enthusiast noob on animation! I tried starting with characters and found the need to go back and start with the basics. So I bought some books on classical 2D animation and try to port the learning to 3D.
Regarding your second question. I stumbled upon some things that might help you.
I know you like physics, so this should make you happy! What it actually shows you is that for every bounce of your ball, it does a percentage of the bounce it did before. So maybe it would be better to start with a ball bouncing in place to check this. Another thing you’ll get from bouncing a ball in place is knowing when to settle! Try it and you’ll find out there’s a moment when you cannot make another bounce (but probably that moment will come some more bounces before of what you expected… it happened to me).
In the case of the basketball is about 50%, so I used that number for my planning:
Also look that as the bounces decrease in height, the ball advances less space (also consider that the ball goes slower each jump, because of the friction and stuff, so EVEN LESS forward movement).
Timing and spacing is everything! Beorn told you much about this.
Cheat!!! I always heard that “animation is about cheating”. And even a physically correct animation can benefit from cheating! In my case I did two things:
*If you sink the ball a little into the plane that represents the floor, it gives you a better feeling of the ball being actually on the floor. Take a look! If you put it where it should physically be (just when the first vertex contacts the floor), then it appears as it floats!
*I did some subtle minor squashes to accentuate the feeling of heaviness/gravity.
There you can see how the jumps (blue line) are flatten on the top and very sharp on the bottom (spacing), and how they decrease about 50% every bounce, the slowing down of the ball advance (red line), and the subtle squashes (green line). I don’t think my animation ended up being so good… I still feel my ball flies a bit (especially on the third bounce).
Regarding rotation: when the ball is rolling on the floor is the only moment when you know how much it rotates (3,14159 * diameter * distance). But when it’s on the air, then you have to choose. If you don’t want it to do irregular jumps, then the ball will probably rotate slowly, so the speed should increase every bounce until it reaches the floor.
Hope these nonsense helps you!!! Hahaha.
Cheers!
Le@ndro
PS: Sorry if my English wasn’t good enough! If I said something wrong or in a complicated way, tell me and I’ll try again!!!
PPS: As I said, I’m a newbie! So if someone sees something wrong, PLEASE correct me!!! Thanks a lot!
Thank you for everything(such a large post, but that means such alot of info too! :rolleyes:), so many knowledge and nice people on this site…it was not like this a couple years back. Believe me.
I have actually been working with Wylow one on one through chat…Amazing stuff really!
But he seems a little busy, so i don’t want to interrupt with a whole chat. Over the summer ill go back and steal him for myself :D!
Anyways I decided, instead of waiting till the summer, i could go back to the forum. I litterally 4got about the forum because i asked him all my questions. And plus with a forum post, all he must do is write on message then feel obligated to chat 4 minutes on end.
Oh, and thank you soo much for the images, i needed those.
So, i was asking, on a post that i did w/ Wylow and Beorn, how the ideal x-axis curve should look like…Wylow said the f-curve should be more rigid like the one on the left than the right:l
Question 1::eek:
It makes sense, and im not doubting him, but why is urs different?
And so many other people’s balls are the other way, like urs?
Did you learn it that way or did you just speculate it to be that way?
Question 2::eek:
Oh, and how did rotation curves look like, or at least your rotation curves?
I am curious to see how you constantly slow and speed your rotations?
Those are my only two questions, i think…or at least 4 now hehehehehe, hopefully its not too much to answer:o
Hi! @ristesekuloski: Thanks for the kind words! But I think I still got to play a lot more with my balls to get good (pun intended, ha!).
@Septimra: Sorry it took me so long to answer. 1) So regarding your first question, I think it’s like Wylow told you: it depends!
I don’t know why I did it that way… though I know I started all over again about three times until it looked nice enough. Mmm… I think I remember that my intention was to combine the “jump in place” Y-axis percentage stuff, with the ease-in over the end of the movement on the X-axis. The resulting curvature wasn’t on my plans.
Now that I look at it, I think it was pretty much a regular sigmoid curve, isn’t it? I remember that when I started I thought the ball “should” make half a turn over each jump, (and It does on the first three bounces, at least in this case), but then I had to slow it down. And when it reaches the floor, it rotates what’s physically correct.
I remember I did the “cartoony style” jumping loop before this, so there I had the chance to play with the rotation speed to see how it felt.
1) The location on the x-axis depends, on the kind or ball…heavier balls less linear, lightener balls more linear…Is that right? :rolleyes: 2) The rotation should be a simple curve…but then why all the extra key frames:spin: 3) I gave this too Wylow to get some critiques but i want yours as well:)
Just on the bouncing with no squash and stretch: http://dl.dropbox.com/u/18498207/0001_0135.flv
I already no, that i have to fix alot of things to be even considered perfect.
I now no that all the location interpolations must be changed
The ball rotation is off
The ball arc after contact with the floor is off
And the ball location on the x is off
But they are only off slightly, i believe
but so far i want to c what u think.
Then with your critique plus wylow and now the cheating thing b4 and the squash and stretch, the ball will bounce extraordinarily…hopefully
@thiagokolb Mmm… I started to, but it ended up being very complicated… the rig is not a good one to start with, as it has some complicated crazy stuff. But even though, the tutorial was going to be in Spanish, so I don’t know if it would have been useful to you?
@StieMO Hey! Good work there! You could polish some of the arcs and maybe take off some of the squashes, as they seem a bit too extremes, but overall it looks great! Did you use JirH rig “out of the box”? With 2.49 or 2.5x? Nice job!
@Septimra 1) I think it depends not only on the kind of ball but also on the inicial velocity you put on the ball. 2) Because I didn’t know it would end up that way! I just putted the keys and hand animated the rotations, and the final result was a sigmoid curve, ha! 3) Looks really good, doesn’t it? I watched it many times and I don’t see all of the things you mention (rotation off, arc off, etc)… maybe the two last bounces are a bit too high, they could be a bit less high so the last bounce is closer to the floor (or it looks like it’s lacking a last bounce). But I might be telling lies, I mean, it looks good!
Thx for the complement…but im a really picky person which is why i may c more things that ya.(So is Wylow 8I, but its all good)
anyway, i was wondering if you have your animation .blend files and if you would mind uploading them.
I would like to just interact with your file(if ya no wat i mean :))
@Septimra
Hi! Sorry for the long silence, Easter holidays.
So, here’s the file (Pelota08.blend (613 KB))! I had to check the rig wasn’t linked. And luckily it wasn’t! Also, this is an old version of the rig. By the way, taking a look at the animation I saw that I didn’t use auto-rotation on it! But I did left the auto-rotating bone visible, to use it as a reference when I needed to animate physically accurate rotations (in the end).
Hope that makes sense to you…
Note: If you want to turn on the auto-rotation, select the “C_Rotac” bone, and in the Bone Contrains pannel put the copy rotation’ influence up to 1 (this old version of the rig didn’t have “on screen” controllers).
Cheers!
PS: I just saw your animation! (don’t know how I missed the link!). It’s looking good! Maybe you already finished the settling??? (I wanna see!!!).
(I wanna see!!!).