Random doodles: climbing skeleton WIP

[battlestar]

Thanks, they look better fully saturated.

[battlestar]

Animating a skeleton, need some help in checking if this looks right before spending a lot of time filling things in... The arms don't look quite right to me, don't know if it's because it doesn't have frames in between or what. Thanks!

750 750 750 750 200 500 500 750 5000

1-9 fast.gif (144.63 KiB) Viewed 5910 times

1-9 slow.gif (144.63 KiB) Viewed 5914 times

[battlestar]

more falling rocks

[Frogger5]

When rocks fall, they don't do it quite that fast. And when the hit the ground, they tend to bounce and roll across the ground a little (or simply shatter.)

[battlestar]

I think rocks fall pretty fast. Unless they're bouncing off a cliff. Also the falling speed varies depends on the distance it has fallen... I appreciate your input, and please show some more evidence to convince me.

I agree small rocks tend to bounce and roll across the ground a little, the smaller one I did does that. The bigger sized ones doesn't seem to do that (unless they're on a slope) and they don't always break apart (the large ones bigger than a horse seem to) according to my reference: (start from 0:44) http://www.youtube.com/watch?v=vPrxRxufF6k The two bigger rocks up there are roughly the size of someone's head, so I'm having trouble imagining them bouncing or breaking apart readily. It would again be helpful to show some convincing counter evidence, thanks.

[Frogger5]

Wait, are you seriously using a My Little Pony cartoon as a reference?

Here's a googled link of someone explaining the physics of falling objects: http://wiki.answers.com/Q/How_fast_do_things_fall

The thing is, you can only make out the rock about four times during the animation, which in the world of animation implies that it is falling at a phenomenal speed, as though it is being propelled by a force other than gravity. Contrary to popular belief, gravity is not a powerful force at all. The only reason we don't float around above the Earth, is because there isn't a more powerful force around to counter the force of gravity.

However, let's suppose that the rocks did somehow fall from an incredible height, or are being thrown down by a particularly strong troll.
Rocks don't fall that fast and then just sit there as though nothing happened. If they fall on dirt, then they create a crater (unless the rock is small). If they fall on rock, then they either bounce, or crack (or in the case of a weaker rock, shatter).

Some real life videos to make my point:
http://www.youtube.com/watch?v=-FoadOYN ... re=related
http://www.youtube.com/watch?v=2czbp9V9 ... ure=fvwrel

[battlestar]

Well I did search around on the web and it turns out all I could find was that for decent side-view falling rocks.

Thanks for the reminder but I do know about the physics of gravity lol.. And there's no way I'm about to calculate the animation speed from a formula. Going to have to do this by eyeing it. The 150 speed one looks about right, 200 seems slow.

These rocks are supposed to be fairly sturdy cave rocks falling on cave floors. They aren't supposed to make craters or shatter. So far I haven't turned up any good bouncing rocks references. I'd have to see some examples before being able to animate it.

[Frogger5]

Find some rocks of varying shapes and sizes, hop onto your roof, and drop them on your driveway.
Ok, maybe not like that, but you get the idea. Dropping rocks is a fairly trivial task, and it doesn't necessarily need to be to scale for you to be able to use it as a reference.

Also, in order to make the falling appear slower, then you'll need to add more frames to avoid jerky animations.

[battlestar]

The bigger round rock could roll some distance further; the small rock would roll further too after already having bounced a few time; the flat rock would bounce up and flip and bounce in a different direction.

Additional frames would be added when the framework is fixed.

Under the circumstances, I could get a good idea of flat things bouncing, but for round large objects there's no way for me to test that without being really awkward.

Any thoughts on the climbing animation?

Spoiler:

http://www.youtube.com/watch?v=5gj2tiqS ... creen&NR=1
http://www.youtube.com/watch?v=XNSpvxSG7Fc
http://www.youtube.com/watch?v=bw6t-4bSABM

[homunculus]

lol, i perfectly understand why you want comments on the climbing before spending excessive time on the sprite : )

And I do have some.
1. The hip width sometimes disappears, considering the length of the leg above the knee (whatever it is called in English). Unless this part is meant to be foreshortened, but in that case the butt sticks out and that takes the center of mass away from the wall. People usually climb with their hips and whole body close to the wall.
It might be counter-intuitive while drawing (at least it is for me), but vertical things are slightly foreshortened in Wesnoth perspective, so horizontal leg above the knee should appear longer than vertical leg below the knee.
2. The arms seem bit longer than in the standing frame.
3. While climbing to the top of edge, the head and arms might be a bit more towards the center of the hex, so that the center of mass would be more over the edge. How far is a tricky question, as the terrain perspective and unit perspective are different.
You might consider climbing to the top of the edge by stretching one leg out sideways first if that looks good, because I would rather think that this is the way such move is done in reality.

(the center of mass seems to have become a recurring mantra for me, especially in animations where, according to physics, smooth movement should only apply to the center of mass, not necessarily to the individual parts)

As for the falling animation, the speed formula suggested earlier assumes that you would do integration to get the location, but really the location formula is just square of time.
time=0 location=time**2=0
time=1 location=time**2=1
time=2 location=time**2=4
time=3 location=time**2=9
etc.

[Gambit]

Hi there. I'm just going to join in on the peanut gallery for a second.

He's not trying to draw a physics-perfect simulation of a rock falling. He's trying to make a decent looking rock-falling animation for a video game with a semi-cartoony aesthetic.

[artisticdude]

Regarding the falling rocks:

I'd personally throw some more bounce in there, especially for the smaller ones. How much bounce exactly would depend on the nature of the surface on which the rock lands... it'd probably bounce less if it fell into sand, and more if it hit earth or rock. Since it's supposed to be falling in a cave, I'd imagine the surface could be any of the above substances?

I'd also have the dust cloud form all around the rock rather than spurt out from one side, since the majority of the impact's force is coming from straight above vs from any particular side.

As a side note, I wonder if it might be possible to achieve the entire animation (with the possible exception of the dust cloud) using only just WML, via offsets and such? In the end it might be easier to just draw the animation out manually, but I really should try out that idea sometime...
EDIT: Alarantalara confirmed that this is indeed doable. Might be something worth considering, especially since it would allow greater flexibility.


Regarding the skeleton climbing:

I love it! Definitely one of the cooler things I've seen recently.

I do wonder if it might benefit from a few additional frames, though. I'd recommend throwing in a total of four frames for the climbing cycle (where he's just going hand over hand up the rock). One frame for his left hand grabbing the rock, another frame for him pulling himself up with his left hand while reaching for the next handhold with his right hand, then another frame for his right hand grabbing the rock, then a fourth frame for him pulling himself up with his right hand while reaching for the next handhold with his left. Then repeat the cycle.

[homunculus]

Ha, some interesting videos.
Also looked at some climbing techniques like Rock Climbing for Beginners- Video 5- Climbing Techniques 1 and some native climbing, and started to understand your climbing animation better.
The butt really does stick out like that sometimes, could be because of flexibility issues, but normal flexibility is the better way to go, I guess.
So the hip that I thought was fishy before, should be ok really.

Now I am noticing more things:
1. When the hand goes up, the shoulder might go up more (shoulders can move up and down individually), and the shoulder joint seems to be too far from the body (collarbone too long).
2. If I understand it correctly, the climbing over the edge part involves straightening the arms, and from this viewing angle the arms and the body should appear longer.

offtopic:
@ Gambit:

Spoiler:

Hi there. I'm just going to join in on the peanut gallery for a second.

He's not trying to draw a physics-perfect simulation of a rock falling. He's trying to make a decent looking rock-falling animation for a video game with a semi-cartoony aesthetic.

?
So, am I supposed to understand that as: a mod came down to this thread just because I briefly mentioned the simple square that could save the poor guy from wasting time on trial and error?
Not that I would be holding his hand, or punishing him somehow if he chooses not to.

Or, in case this was a flashback from the parabola vs catenary discussion earlier, my point there was just to say that reusable tiles can be done if unit offset allows it.
Why the discussion got out of hand should be clear when looking at the overlapping parabola and catenary plots in my latest post there, and trying to imagine what I might have thought about the dire need to make the difference between the two.

/offtopic

[battlestar]

Thanks for all the pointers on the climber! Those were only contact position frames to define each segment, I'll try adding a few defining passing frames to actually make it move (you never know what you can learn from ponies) and try to present it in a better way.

Some fireplace in the meanwhile... I don't know how it'll look in game.

Also, wonder if anyone savvy with terrains could do a set of ice castles? I'm still fairly clueless about terrains.

[battlestar]

Think the edge frame after the little hop still looks wrong (maybe)... And what about the timing of frames?

Planning to add frames if these look OK.

1-7 fast.gif (114.43 KiB) Viewed 5737 times

Progressing two furniture pieces...