Settling Balls(Almost There)

Artwork Animations
1

I have been animation for some months now, and i have been doing ball animations :rolleyes: for sometime now. Really basics ball animation. I mean so stretching, and no settle, just trying to get realistic timing and spacing. This is my latest project, using everything that i have learned: http://dl.dropbox.com/u/18498207/0000-0096.flv

I have not had the opportunity to blender much due to high school -___-

Please tell me how i can make it even more realistic. And Yes, how to start to attempt the settle. How would i even start that rocking motion? I can do the squash simply but the settle i will need assistance.

Thanks So Much, and Happy Blending.

2

Am I understanding you correctly when I take ‘settle’ to mean ‘comes to rest’? (Methinks I’ve been exposed to too much physics).

I’m more or less a beginner myself, thanks to school and what not. Anyhow, back on topic. How can you make the animation more realistic? It looks pretty good as far as I can tell. What I would like to know is whether you’re taking what material an object (in this case, the ball) is made up of before taking into account the extent of deformation. If it’s softer, I’d expect it to deform more, lose energy much more quickly, and so as a result not bounce as much. Assuming you are talking about a typical tennis ball, it will deform somewhat, and bounce a good many times before coming to a rest (rises to 3/4 it’s previous height with each subsequent bounce - atleast, that’s one way I’ve thought of gauging it).

As far as your animation shows, I don’t think the ball deforms a lot on contact with the ground? I tried playing it frame by frame, but I couldn’t see any significant ‘squashing’ as you put it. Try watching a few videos of real balls being bounced on different surfaces frame by frame as a sort of reference. Also, I like the fact the ball was spinning as it would be; however, how fast it is spinning (seemingly constant or perhaps it’s just me) contrasts with how the ball is actually losing energy. For some odd reason, my mind is telling me balls reverse the direction of their spin with each subsequent bounce due to friction; could someone verify that? I’m not quite sure if I remember correctly.

As far as helping you with the ball coming to a rest, I’ll do my best. Since you know the ball is losing energy, it’s easy enough to understand that at the very end, using the last dregs of energy it has, the ball desperately sways backwards and forwards before coming to a stop. I’ve attached a Paint illustration (you could call it) of what I mean. I’ve tried to explain stuff clearly so that it comprehensible for everyone (not everyone has done Physics, or alternatively, is good at it) but then again, perhaps I might just end up confusing matters …

If I’ve made a horrible mistake, I beg your pardon. (I realise my explanation was very physics heavy and might be completely useless).

All right, I’ll stop now. ;_;

Edit: I realise I’ve addressed more of the theoretical side of the issue, and less so of how to go about it. I’ll see if I can do something about that later, otherwise I leave you in more capable hands than mine. Good luck.

Attachments

Rolling ball.jpg1024×576 60.6 KB

3

If the balls center of mass is at the actual center, it will not rock back and forth, but instead roll forward, and air friction and the opposing force applied by the floor will stop it. You seem to have it spot on really, although I’m not too sure on how quickly it looses its velocity at the end.

4

Ah, thought it was something along those lines. About it losing speed, I can potentially see many factors, er, factoring in when concerned with how quickly it loses velocity. And yet, for the purposes of a simple animations, I think we can safely keep these factors simple enough. Keeping in mind soft- and hard- body dynamics, let’s just say a lump of dough will fall to the ground and stick there (one extreme) and a sold rubber ball can theoretically bounce on and on (another extreme). So I’d hazard a guess - between zero and forever ;).

5

Wow, I was really tired when i wrote my last post. I am seeing all these different sorts of grammatical errors. Whoa…
Anyhow, i meant to say that i was not focusing on squashing and stretching but simply timing and spacing for now, Zaeche.

Also, i understand what u mean and it makes total sense. Don’t be hesitant to go physics heavy on me cause I actually love physics and mathematics, and tend to understand things better that way. :wink:

So what your saying it that the rotation of a ball reverses with every bounce, or does it simply decrease?
Along with the rotation, the settle will have to do the following, all with the same general curve you showed me:

  • location(back and forth, digressing with each movement)
  • rotation(back and forth, digressing with each movement)

Am i correct? :confused:

6

Rotation may only reverse depending on which way you are throwing it and which way it is turning, and how fast the throw was. It usually starts spinning in the direction it is moving after it hits the ground (because of the friction applying torque on the ball in the direction the ball is moving when it hits the ground).

For example, a ball spinning anti clock-wise moving to the right and falling from a height, after its first collision with the floor its rotation will either be reduced or completely reversed (note the reversal of the speed does not have to be equal to the original rotational speed. for example if it was rotating at 100 rpm originally, it does not mean after the first collision it will be doing 100 rpm in the second direction).

After the first or second collision however the speed at which the rotation is at will be affected only a little by the collisions with the floor.

Also, if the ball is rotating perfectly with the floor, it should move Pi*diameter of the ball for every rotation.

Its best to just imagine or get a real ball, or something cylindrical and try different way in which the ball would hit the ground, and what friction would do to it. Or you could actually record the ball and play it back in slow motion :P.

EDIT: Almost forgot that if we look at the anti clock-wise spinning ball falling to the right, if it was spinning very fast, it would actually have enough force to change its direction of movement to the left, but the rotation speed would still be reduced, but not inverted.

This also allows us to theorize that looking at the scenario again, but this time with the original lower rotation speed, the movement speed will be reduced but not inverted as well. Arg this all may be hard for you to understand, but I find it hard to splat my brain into this text xD.

7

tl;dr: A good game of Cricket is what this topic needs.

8

Yours was longer xD. Never played cricket before D:

9

C’mon guys, lets be nice.
Im actually taking my physics final tomorrow :yes:
Thanks for all your help. I’ll try what you said as soon as possible.
Please subscribe so when i add a new animation, you guys can tell me what i did wrong :smiley:
I only say that because i love your critiques and they make sense to me :wink:

10

I’ve subscribed. Oh, and I have finished all my exams for now :D. I have a lot of free time :P. I should also tell you I haven’t been very active with blender in like… the past year, so I might not be able to help you with certain things, but in the old days I was quite the blenderer ;D. Infact, I’m going to go play with blender now, lol. Hopefully I can relearn my old skills.

11

This link has a good discussion on bouncing balls from an animator’s perspective. It also talks about how ball motion ties in to all sorts of other motion, eg walk cycles.

http://blog.fjasmin.net/2011/01/what-does-bouncing-ball-can-teach-us.html

I’m with Moffboffjoe. You could sit down and try to work out all the physics involved, drag, turbulence, various friction forces, mass distribution, rotational inertia, how level the ground is, how soft the ground is, how sticky the ground is etc… Or you could just film a ball being dropped and study that instead.

12

O, how I can relate to that.

Except for the part of being quite the blenderer in the old days - I was still learning. And I’ve already wasted a month of my holidays with procrastination :(.

13

Thanks for subscribing. And you guys should get back on blender again, its not a waste of time.
Even though its free, its epic. Really the only reason I used it, and now im hooked :wink:

Bender007, thanks for the info but i know almost everything there already.
I have been doing ball bounces for some time, but thank you anyway. I
appreciate it! Are you an animator as well? :eek:

Alright guys:cool:, i was digressing but now i attempted to start and questions flooded my brain.
Could you guys explain the theory of rotation.

What i got so far is that the ball will rotate abnormally. Then the ball will hit the ground and bounces uni formally, does this sudden change in ball rotation change the way distance and height of the ball bounce from being some what proportional to the first?

This is so fun, im learning more about ball bounces then i have in months :yes:

14

From what I can work out, if a ball was rotating faster than it should be for the speed it is moving at, the distance of the bounce arcs will become larger (and the balls rotation would be reduced, maybe to the point in which it matches the movement speed). However I don’t think the height of the arc will change. I’m not too sure about this though, and will need someone to confirm it.

EDIT: Here is a picture for further explanation if needed:

Ball examination.jpg1024×610 81.3 KB

15

If you know everything that’s fine. Good luck.

16

As promised, sports really do help. Have a look-see: http://en.wikipedia.org/wiki/Topspin

I am familiar with topspin in cricket and squash - I’ve used it too (devil of a spin). Because of the way it is spinning, it serves to drop the ball more quickly through the air. Moreover, after it bounces, the ball moves further along in terms of horizontal distance rather than vertical distance - a side effect of the spin: it keeps the ball low.

@moffboffjoe: The height of the arc shall vary indeed. I do not know if you’re a quickie like Broad or spinner like Giles but from my experience, anyone can add a bit of spin to an old tennis ball. In fact, I’d encourage everyone to try it sometime; it’s a fun experiment if you’ve never tried it before and it’s also a nice time killer.

17

Nice title, moffboffjoe: Looking Deep into your balls. :smiley:
So essentially all ball bounces are un even?

Because there is topspin in all balls. Some more that others, ya no?:confused:
But lets say there was a ball with major topspin. Would it bounce something like this?

or will only the first bounce have topspin and the subsequent bounces, like moffboffjoe said, will essentially correct themselves( I mean correct the rotation and location to make the Magnus effect only happen in the first ball bounce)

I think im getting somewhere. I learn new things every time i read this post
Magnus effect is pure brilliance :wink:

PS Bender007
I did not mean to sound rude. Thanks for the info a lot.
I’ll refer it to tons of people, and its even in my folder for ball animation and physics sites in chrome. Please don’t take the comment the wrong way :frowning:

18

I don’t know if that image is truly correct, as I think the effect wouldn’t be so obvious on the second or third bounce, although it would still be there, just reduced (but the speed of the ball will be increased if its rotating very fast). I might not be correct as I’ve got a bit of a hangover from last night lol.

19

I do believe you’re right. The effect all but dies out after the first bounce after which it is a normal bounce pattern. At least, that’s the case if I recall correctly.

20

Don’t sweat it. I shouldn’t be so precious… :smiley:

Either way, just happened to stumble on this YouTube clip of a guy dropping a bunch of balls, which may be of some interest. No massive spin on them unfortunately, but it’s a nice selection and does have a few throws in there which might help at least with basic spin and settling.

[video]http://youtu.be/sKJegbjS4N8[/video]