Hi there,
I work with mostly motion graphics animation and in my workflow the general rule is: less keyframes is always better, and more manageable.
I would, for instance, animate this arcing ball bounce without the keyframes on the top of the arcs, and manipulate the curves handles to make the ball go up. Same for the translation, I would usually just add three keyframes: one for the end, one for the start, and one for the overshoot, then adjust the handles to achieve the motion I wanted.
Now here's my question: is that just a silly way of working in the world of character animation? I'm coming into this course knowing that I'll find a ton of differences in the styles of animation I normally do and what I'm trying to learn, and I wonder if this is the first one?
Or are both styles valid?
Hi Nathi,
That's a really good question and I think its something quite common to think of when working with these foundational exercises. I've once seen someone just plug an equation in on the rotational transform for the rolling ball exercise. This all works, its working smarter not harder right? But when you start getting into more complicated exercises like pantomime shot, acting shot or anything that combines 2 or more actions. You will start seeing, the more you rely on the computer doing all the tweening there is a swimmyness or a very transitional flow to the action that the audience will feel and animators will see. You can see it a lot in tv animation, because of the fast rate of production. Animators have not time to support all the key poses with tween poses. If you frame by frame you can literally see many shots are just sliding from pose to pose. And by no means will I tell you, you need to have a pose each frame and DOWN WITH THE COMPUTER to handle tweens. But you'll find when you are under the situation; You have a character doing this full body action with 30, 40....100 controls all firing at once and then you realize the timing feels off. You need to slide everything back a few frames. You'll be wishing that you had the keyed pose to slide around over a stretched fcurve.
"Everything is a bouncing ball." and the ball in this case is teaching you how to see and lock in those golden moments of action.
I hope this helps, happy animating!
That makes a lot of sense! I can totally see the flexibility in being able to move the keys around. I guess I'll have to get two animator hats: one for fast motion graphics projects, one for character animation.
Every time I see a graph editor in Blender with a salad of lines it makes me wonder how on earth can one manage all of that?
I'm hoping by the end of this course I'll have a little better grasp on that.
Thanks for the great insight Phil!
Good questions Nathi.
There are always more than 1 way to skin a cat.....although why people would want to do that to the poor cat is beyond me.
Sometimes less keyframes is better/easier and sometimes less keyframes is harder than it needs to be.
It really depends on the situation AND the animator (and to a lesser degree, the software).
Imagine this curve below.....(it's not to scale and I have just quickly sketched this out so it's not 100% accurate but hopefully it illustrates the point)
An animator could place keyframes at each one of the blue marks.
Or at the orange marks
Or a combination of both.
The result is the same so which is better?
The way that got you to the end result the quickest and easiest.
(This will be different for everybody)
When I was a younger animator, I used to use as minimal keyframes as possible. And adjust the handles on everything to make it all flow nicely together the way that I wanted. My animation was alright, but it took longer than it needed to.
These days I use more keyframes in the spots where I know I don't want to be messing about with the curves and I can just tell the software what to do.
And I use less keyframes where I know the software is probably going to do what I want automatically - or only require small tweaks.
Work Smarter not Harder.
In your example that doesn't have the keyframe at the top of the arc. That is certainly possible. But that's only something I do on the really small bounces.
For me, it's easier to have that orange keyframe there - and that way if you need to adjust it, that is what you adjust. The other side handles are easy because I know what shape the curve should be (I've seen a lot of animation spaghetti)
As for the X loc in your example - I would start out the way you suggest, but then I keyframe the contacts and make the connections vector to straighten them out while the ball is in the air.
|
|
Hope that helps you think about your workflow.
Very nice! Thanks for taking the time to literally drawing it out for me. It sure does help, and it does make sense that each scenario will dictate what the best way to go about it is. Practice and time will eventually let me know I guess. Thanks again Wayne!
