← Home ← Back to /3/

Thread 1021361

15 posts 6 images /3/
Anonymous No.1021361 [Report] >>1021366 >>1021381 >>1021417 >>1021421
Mathemathically speaking, how to generate curve/positioning for animation that is realistic and correct? As a curve itself is a line, there are aspects that makes it real like gravity and forces...how is this, rather than roto, is considered then? I'm kinda curious...especially since it is the correct way of animation, perse. Otherwise, how should a walk,loop,sequence are planned...with science considered too...correctly?
Anonymous No.1021362 [Report] >>1021363
:gas:
Anonymous No.1021363 [Report]
>>1021362
Anonymous No.1021366 [Report] >>1021370
>>1021361 (OP)
>>1020614
>>1018686
>>1020272
>>1020497
Anonymous No.1021370 [Report]
>>1021366
Well one seems to lead somewhere.
Anonymous No.1021381 [Report] >>1021414
>>1021361 (OP)
>Mathemathically speaking
fourier transform
Anonymous No.1021396 [Report] >>1021401 >>1021414 >>1021440
Its annoying. I want to see people's projects, share my projects, give advice, get advice. But this shit is all there is. Why why why are you entertained by reposting the same thread over and over and never actually animating a walk cycle?
Post your progress op. Post your model and how much walking youve done.
Anonymous No.1021401 [Report]
>>1021396
Hear hear. I've got 35 blocked threads right now and that's just the garbage AI and troll ones, not even the low quality ones. And I even kept the slutty peach AI one.
Anonymous No.1021414 [Report]
>>1021381
Can you please show? my talent iq test suggest im good at computer, not counting.
>>1021396
But i di- oh that was last night...not here. here
it's animated but the motion is just v hard
Anonymous No.1021417 [Report] >>1021420
>>1021361 (OP)
There is no "correct" way to animate on a mathematical level.
You would need to understand the body on a level that most people will never reach. Did you notice how difficult it is for people to build bipedal robots? They've gotten pretty far, but those robots still have an uncanny feel to them. The scientists behind those robots understand human body mechanics more than anyone, and still can't get it exactly right.

You're asking for information on something that nobody has achieved and documented yet. So why don't you adjust your expectations to something more realistic. Or, you lead the way and revolutionize our understanding of body mechanics. Those are your only choices.
Anonymous No.1021420 [Report]
>>1021417
>none documented
But animations have decided upon it, and procedura lgenerations are closest. you just need to deliver me something of it, than none at all then.
>robot is difficult
this clearly isnt diy nor is there any manufacturing question asked.
or perhaps if really too confusing, why not check kinovea and come up here with a nice diagram that is "more realistic" of an expectation. clearly, there's not much given to me yet other than "just shoot anicotine" tier answers.
Anonymous No.1021421 [Report] >>1021427 >>1021469 >>1021469 >>1021469
>>1021361 (OP)

Animate for feeling not for realism. We aim for the hyper real to sell impact and feeling of motion. Don't waste your time aiming for what is real, spend your time trying to generate how reality feels.

There's a reason a punch is windup, jump to end pose (overextended) and then most of it is just falling into the rest pose after the punch has occurred. That's the feeling of a punch. That's the art of animating a fast action. That's what our eyes catch and our mind understands from the motion.

If you have autism or something that keeps you stuck in this reality obsessed mindset, try to find ways to decondition yourself from thinking that way, its harmful.
Anonymous No.1021427 [Report]
>>1021421
...nah i'm just trying to practice man.
>reality obsessed
tell the industry man. i'm trying to have advice to follow, but i guess some newb motivators, end up in some shophouse making 2d games nobody notice, despite quipping the famous or whatnot.
>punch is
>jump to
can i get a diagram at least?
lestmaking it a private class of sort, given the tightness of time. I think curves ,good curves will help, but not even chatgpt can point me where. It's, unrealistically difficult.
Anonymous No.1021440 [Report] >>1021469
>>1021396
Try reddit
Anonymous No.1021469 [Report]
>>1021421
>>1021421
>didnt i fukin reply this
>yesterday
>fuck
>nobody documented
BUt procedural generations operate on that level. So it clearly rely on something.
>robot
noone said about manufacturing
>>1021421
>waste time for real
why are you all complaining about shit then
>autism
>suggesting autists even have capacity of choosing or, be considered
We're talking about you, the fuking customer.
Something fast, easy,cheap, that's all, please share, or fuck off with your failed seminars.
>>1021440
gotohell