AuthorTopic: I Want To Be A Mighty Pixel Wizard  (Read 9899 times)

Offline 0xDB

  • 0011
  • **
  • Posts: 873
  • Karma: +0/-0
  • Dennis inter-is.
    • dennisbusch_de
    • http://pixeljoint.com/p/1287.htm
    • 0xdb
    • View Profile
    • 0xDB

I Want To Be A Mighty Pixel Wizard

on: February 16, 2015, 03:54:54 pm
current revision(left one) #09 (thread started with #06 (right one) for comparison):


Still working on it... too excited to keep it secret any longer... must share process/progress! :)
 

todo:
improve balance (de-symmetrize shoulders and arms, tilt&twist girdle)
improve fabric on legs/shrinking cloth issue (shortened whole robe)
tweak shoulder movement, raise them after arms were about 45 degrees pointing down while on their way up
improve balance more between top/bottom of the figure
folds/shading
revise colors
(optional: pixel a nice background)

It is taking forever but it is coming along nicely, slowly but steadily.
« Last Edit: February 21, 2015, 08:59:03 am by Dennis »

Offline Conzeit

  • 0100
  • ***
  • Posts: 1448
  • Karma: +3/-0
  • Camus
    • conzeit
    • View Profile
    • CONZEIT

Re: I Want To Be A Mighty Pixel Wizard

Reply #1 on: February 17, 2015, 01:08:23 am
hey there! that is a very nice animation you got so far.The pacing makes the movements feel real and I am especially impressed by how clean every one of your steps look. very pixel minded and you defintively have the patience for an animator :p

1- what is he doing? it looks like a little dance. He looks like some old master, and from what I have here I would assume some kind of dance master I'd guess a Sufi or Sufi analogue (monks that meditate by dancing, mostly spinning) if that's what's he's doing then the general movement is fine, but he kinda has that stereotypical kung fu master haircut so I suspect he might be supposed to be attacking. if he's doing some kind of spinkick the leg should definitively stick out more than the arm ever does so that he has good attack range.

The rythm. if it's supposed some kinda qui-gong like or sufi like dance I guess the slow pace is ok, if not....well tell us what it is so we can tell you how to tweak it :p you did a good job on speeding up while he is spinning, good sense of weight there.

2- Balance. you showed us every steps shows very clearly your thought process, that is very nice.
Now in the first step, we can see  the bone structure clearly, and he does keep his balance perfectly, but you did that in a bit of a strange way. The spine stays completely straight all the time this is a bit robotic.

To keep balance any time you tilt something to the right, something else has to tilt to the left.
so for example when you walk, you usually tilt your shoulders one way and your hips the opposite way, just the like one leg goes forward while the opposite arm goes forward.

this is a good example of what's odd about the animation, the guy never does this and it seems a bit robotic.


it comes from a link to a series of videos on phisics for animators (from a phisicist)
https://www.youtube.com/watch?v=zjhRxeR9GXo&list=PLdvNrK-c96RH2MEobcVQqhp8j26Wr1gVk
while the vids are a bit longwinded and kinda go on too long on tangents for my taste they're  very informative.


http://www.scribd.com/doc/26829442/Animator-s-Survival-Kit
The animator's survival kit surely has something on this but I couldnt find it by skimming

3. folds. well this one is a bit complicated. I know this instinctually but dont have the best language for it. First thing is that cloth gets lifted when someting interrupts it's normal fall. when you lift it as he lifts his two legs it's alright, cause both his knees are pulling it up. but later only one the leg to (our) left lifts BUT the whole tunic lifts a lot. Really, the left side should've lifted, it's true that all of the tunic is the same object so the right side should lift a little, but if this guy is a master of this dance thing then his tunic should have plenty of slack for him to do this without the whole fabric being pulled up (which I guess is what you were going for but it kinda looks like it all shrunk to a skirt). instead just lift the part to the left a whole lot and leve the right part almost alone.

also, the way the general shadows of the lower portion sort of gravitate towards where the monk's body is not there isnt really realistic. Your face shading is pretty ok, think more like that. The starting point is simply hanging tunic. This should look more like just a hanging curtain, you dont need to point out where the monk's legs are at all. just wrap that around a (for our purposes here) round shape and you should be ok.

Just noticed that you dont keep the highlights on the head the same place troughout the rotation, you kinda stick highlights to features as if they were a texture. That's not how it works. =/ they should pretty much stay on the same place on the picture but reflect according to the shape they're hitting. But to know how to do that you need to know about lightsources. can someone please tell him about lightsoruces? :p

Now, your general instinct of how to create folds is pretty good, you have them all pointing to the part where it sticks out and making arches to the other points of tension when they exist.  You should follow that more boldly when strong  points of tension appear.That should really dominate when he lifts his leg to be parallel to the ground, make that arch from the knee the basis of your shading when he lifts the leg.

I made this. maybe dissecting it will help :p

If you want to learn about folds you pretty much just gotta observe the shit out of them. just pay attention to what clothes do when you and other people move, hell maybe even on tv or movies. start with simple stuff like a lifting arm and then observe more complex stuff. fully robed people moving around (realistically) happen to be pretty complex :p

Of course, in animation there are always varying degrees of simplification.
in pixelart
King of fighters (specially the later ones) Garou: mark of the wolves. matrimelee. waku waku. Capcom vs SNk.
are all very well balanced between simplifying enough to animate and still make them look like real clothes. the list pretty much goes from most complex to most simplified,

you can find most of those games here
http://www.emugif.com/

Street Fighter 3 also has good cloth animation but it is way more exagerated. it's like the clothes take on the personality of the wearer and have very expressive motion (famous example is ryu's stance). here http://www.zweifuss.ca/

anime is usually pretty good at this too. you could also just go cartoon clothes if you think this is all too much :p

That is pretty much it. You've got a very methodical way of doing things and that's good, I'm very very rash and I have problems with scale that I have to correct all the time, it looks like you'll rarely ever have that problem so cheers for that. Also, incidentally you dont HAVE to be so pixel perfect in the first steps, no one will see that so there's no harm in putting jaggies there or even using big brushes. I use big brushes to draw rough sillouehtes (imagine a sketchy version of your 3rd step) a lot of people do :p... that's if you want to though, if you feel this is easier good for you. I kind of see you growing into making flashback style animations merged with cartoony faces from what you got here, which would be pretty peculiar and neat =) heheh. Please dont lose your enthusiasm for you work and enjoy your dancey monk anims :p

« Last Edit: February 17, 2015, 01:37:25 am by Conceit »

Offline 0xDB

  • 0011
  • **
  • Posts: 873
  • Karma: +0/-0
  • Dennis inter-is.
    • dennisbusch_de
    • http://pixeljoint.com/p/1287.htm
    • 0xdb
    • View Profile
    • 0xDB

Re: I Want To Be A Mighty Pixel Wizard

Reply #2 on: February 17, 2015, 12:09:42 pm
Thank you for the time to share your thoughts. If I had any delusions of grandeur before... they are all gone now and now that reality has got me back, I can hopefully continue to work on this without the excitement and concentrate on making it solid(or rather fluid!).

1 - What is he doing?
The intention is for it to be a controlled dance-like expression (the kind that mad wizards do for personal joy when no one is watching/judging (thinking a bit in the direction of Ballet)) not any type of attack or aggressive move.

I did not plan much ahead of what it might be(come). To be frank I just felt like animating some fabric (fabric was the theme for #pixel_dailies(twitter) last monday) and jumped right into action.

I was also thinking (that thought came after working on it for a few hours) it looked like it might be part of some very controlled ritual, a part of a complicating spell-casting sequence where he has to concentrate on being very precise/exact with the timing and movement of each body part, even with the eyes (so I added that), or the spell would fail.

2 - Balance
That (controlled spell ritual part) would also be a good explaination for keeping the posture around the spine so static/robotic. I will see what I can do about the balance though, breaking the shoulder and arms symmetry in the first part might even already do the trick to make it look more natural and less robotic.

3 - Folds
This is definitely one of the most "difficult" parts to animate. I first animated each fold individually and then spent ages tweaking the bits and pieces where the folds "interact" with each other.

It is hard to describe what I was doing really because it feels like animating the folds is more about "feeling" and imagining the fabric as a whole rather than moving individual pixels around. It also helped to stand in front of a mirror and watch how the folds/shadows kind of emerge/sink from/into the depth when fabric deforms.

And there are many different types of folds... there is a good description of folds in one of the Hogarth books (it does not deal with animation though and I found that "some things", in a small pixel resolution, do not work the same way as in static images, I am not fully certain what it is or how to best describe it but I will try: It feels like things/pixel clusters(rather than actual abstractions of what one is trying to depict) have to be tweaked in way that the eyes can track them or it will break the fluidity of the animation and end up looking wrong... it sounds counter-intuitive not to aim for realism but from my observation, the focus on pixels helps the animation more than insisting on what would be correct in a static image).

4 - Lighting/Lightsource
The light is meant to be some kind of artificial not-or-rarely-to-be-found-in-our-natural-reality almost straight from the top (think an infinite number of vertical parallel rays) light to serve clarity of form independet of position in 2D screen-space while at the same time I use light to suggest surfaces which are (in depth) "closer" to the viewer and shadows to put things to imaginary planes that go deeper into the screen (it is a foreshortening trick where one uses shadow/darker tones to suggest the contraction of a plane as it curves/rotates away from the view plane and not so much actual shading caused by an imaginary lightsource).

I was going to add highlights sparsely after I am done with the shading. Maybe something like I did in an older animation of this mech:


I am not sure if I will though as it might looks strange and wrong in this wizard animation (not too happy with it on the mech either) because of the rather huge amount (in whole pixel distances) of movement going on here. It is where I would probably prefer to keep the unrealistic "depth-suggestion-and-foreshortening-lighting" abstraction rather than aiming for realistic shading at all cost. Will have to experiment with that.

But... well... I will stop writing and resume pixelling as I feel I might otherwise be perceived as being too defensive about my work (I still trust my observations to be valid :) ).

edit
P.S.: Updated todo list in opening post for the most important issues I want to address based on your feedback.
« Last Edit: February 17, 2015, 12:51:19 pm by Dennis »

Offline Conzeit

  • 0100
  • ***
  • Posts: 1448
  • Karma: +3/-0
  • Camus
    • conzeit
    • View Profile
    • CONZEIT

Re: I Want To Be A Mighty Pixel Wizard

Reply #3 on: February 18, 2015, 04:41:34 am
it's great man, I dont get the feeling you're defensive at all...but I think I might've not been clear for speaking so longwindedly.

1-good, so I had a good general feeling for it. you should maybe look at qi gong videos on youtube, they should be pretty close to this vague spell dance thing you got going.

https://www.youtube.com/watch?v=zzD4svzG9js

3-you've studied in books what kinds of folds there are? good, that explains that you have a good general sense of how they go, you are very much right to say that when two folds meet you have to feel it trough and that's the haredst part, but what holds you back is you are unsure about how to light the whole fabric. I posted my little walking sypha to maybe give you some ideas on that. You are right in saying representation is not the same as reality, there has to be some simplicity and some continuity for things to work, which is why I gave you refferences of different levels of simplification. I'll touch that topic again on lighting :p

2-balance. I think you are starting to see I have a point but maybe proko can really convince you.
https://www.youtube.com/watch?v=_JHt6lOGfRs&index=21&list=PLtG4P3lq8RHGuMuprDarMz_Y9Fbw_d2ws
try to observe people and see this kind of thing. maybe look at a qi gong clip and see how their spine moves?.For the movements you got it isnt that necesary, but you should really tilt his shoudlers and hips at the moment he stretches his legs, that really should put some side inbalance and the spine just has to be involved in compensating that.


4-lighting. the lighting on the little walking mech is great. that sort of rim lighting works great because the light source is far away to the back so it makes sense and visually it's great to emphasize your mech's sillouethe.

What made me worry about the monk is :

a-when you rotated his head, you had every highlight stick to the same places of his body.Were this the product of a 3d program it would've been as if instead of recalculating the lighting on each frame you had calculated the lighting on one frame and then just baked (painted the lighting as a texture of) that on the model. Highlights appear where the parts that stick out meet with the lightsource in the right angle to be reflected at the viewer, therefore when the parts rotate they wouldnt be in the same angle and the highlight should disappear, so think more about the shading in every frame. I did this in that gif I posted so maybe check that out.

KOF sprites are really good about doing this too. so maybe check a kof girl with long skirt or a monk from all those references I gave you :p


b-the way you shaded the guy's legs. the general shape of the lighting is light roughly sticking towards the legs. that is just not what happens with a robe. it should be more like a hanging curtain, so the shading should be ALL about the folds that are going on. it's ok to feel trough the way you do folds (you do that pretty ok) but it is definitively not ok to do that for how the light hits your figure. that is all based on the lightsource and what's blocking it. You might eventually know how to calculate that without much thinking, but it's still not a "feel" thing.

https://www.youtube.com/watch?v=V3WmrWUEIJo
Proko has a video on lighting and it would prolly help :p. dont worry about reflections for now, that usually isnt done in sprites..heh.

« Last Edit: February 18, 2015, 07:56:15 pm by Conceit »

Offline 0xDB

  • 0011
  • **
  • Posts: 873
  • Karma: +0/-0
  • Dennis inter-is.
    • dennisbusch_de
    • http://pixeljoint.com/p/1287.htm
    • 0xdb
    • View Profile
    • 0xDB

Re: I Want To Be A Mighty Pixel Wizard

Reply #4 on: February 18, 2015, 11:59:03 am
b-the way you shaded the guy's legs. the general shape of the lighting is light roughly sticking towards the legs. that is just not what happens with a robe. it should be more like a hanging curtain, so the shading should be ALL about the folds that are going on. it's ok to feel trough the way you do folds (you do that pretty ok) but it is definitively not ok to do that for how the light hits your figure. that is all based on the lightsource and what's blocking it. You might eventually know how to calculate that without much thinking, but it's still not a "feel" thing.
I tried (and failed) for a couple of more hours to just kind of wing it now but it just does not work, so I finally took a couple of relevant reference shots myself and will work from those when I pick it up again (a bit fed up with my incompetence at the moment).

I can not 100% accurately reproduce the light-source or the poses I am aiming for in the animation but for the folds, these should be good enough. I should have made refs from the start instead of wasting so many hours trying to imagine how it might look.



Offline Conzeit

  • 0100
  • ***
  • Posts: 1448
  • Karma: +3/-0
  • Camus
    • conzeit
    • View Profile
    • CONZEIT

Re: I Want To Be A Mighty Pixel Wizard

Reply #5 on: February 18, 2015, 09:22:03 pm
those refs look great man! I was a little afraid I was buldgeoning you with my brickwalls of text. lol

the first photo is pretty much what I meant with more like a curtain, it just basically hangs like a curtain :p

I think it's extremely normal that you feel frustrated with this, most animators dont bother with shit like this, even great ones like John Krickfalusi (ren & stimpy). He explains on his blog that most american features use a substitute he calls "cartoon skin" http://johnkstuff.blogspot.com/search/label/cartoon%20skin. In general only japanese who are trying to be realistic pull it off :p aside from the games I quoted the other examples that stick out are stuff Like Akira, Ghost in the Shell (esp. animations by Mitsuo Iso) or Jin Roh (Hiroyuki Iura stuff). OH, and then there's this in pixeljoint :p http://www.pixeljoint.com/pixelart/31255.htm (he had to be French, it seems they're the only westerns that can pull off what Japanese do :p) so dont feel bad, you're learning something most people cant do. The struggle is something quite natural.

BTW: here are a couple of cloth heavy SF3 gifs  from zweifuss that may be relevant. GILL http://www.zweifuss.ca/colorswap.php?pcolorstring=GillPalette.bin&pcolornum=1&pname=gill/gill-intro-right.gif and some Indian referee (from the extras page) http://www.zweifuss.ca/special/judge1.gif

The general theory I have for folds is that any point that sticks out creates sort of an asterisk * then when two points are close enough that the lines of each asterisk converges, you get U like shapes. like between the shins in your third photo. Now, the process of transitioning from straight lines to those U shapes is what's complex, because lots of strange little wrinkles appear like what's at the leg of your second to last and last photo. IMO you should put as much of that in there as you think benefits the animation, that's what you CAN feel trough, if it's accurate but it feels to busy forget it.

I think a good example of how that can be difficult is actually between your tighs in the 3rd photo. see how there's a sort of asterisk * projecting from the right side but all the lines are curved, you can see some of the left side's asterisk below the knee but not at the shins. it's just like the asterisks battle it out and the right one won :p that's why I can somewhat understand feeling it trough.

But on this case, I think I can see the reason why. See how really close to the right knee there's a sort of indentation that goes down to the height of the knee? Now switch between 2nd and 3rd photo. it -feels- like that gave the right asterisk an andvantage and that's why it takes over on the shins. Now maybe that's not what really happens but with that I have a continuity in my head that I can animate, and that's all you really need.
 
Your torso in the 3rd to 5th photos has basically two points of tension that are the shoulders, but the strenght of influence varies.

in the 3rd's torso it's pretty much just plain U-s.

On the 4th's torso fabric on your stomach is pulled so tight it's like the whole stomach is a point of tension, so you get those horizontal wrinkles, then going up a bit that transitions into the regular U shape and it starts to look weird, then at the height of when your neck starts. there's this just weird excess fabric

On the 5th's torso it's like an u that's being pulled to the left, probably because your spine is a little swayed and the fabric is being pulled by the leg too. now above the very clear u on the torso there's a lot of very weird interrupted U-s and a lot of detail there, but you dont really see that from afar, I think that's the sort of thing you can skip, specially if you cant figure it out :p. as long as you have points of tension that meet and transform eachother and disappear in a coherent way, it doesnt matter wether it's perfectly realistic or not.

I hope that little bit of rambling was a little illustrative, there are many ways you can invent how those two points of tension interact, I think what's important is that you have a reason for why the fold is there, and the continuity shows in how you animate it more so than being strictly realistic and getting bogged down by a thousand little wrinkles.

One last thing about the lightin. You mentioned simulating depth of field by darkening things as they go into the background and I agree with it, this is also what I had in mind when I made that little walking Sypha, that's what I didnt see going on on your sprite as it rotated, Sypha's hood rotates juust a little bit but I still redrew things to reflect that. So as long as you keep that in mind as the guy rotates the lighting should improve a lot. Not that this is terrible lighting or anything, it's somewhat the norm actually :p but
« Last Edit: February 22, 2015, 06:55:50 pm by Conceit »

Offline 0xDB

  • 0011
  • **
  • Posts: 873
  • Karma: +0/-0
  • Dennis inter-is.
    • dennisbusch_de
    • http://pixeljoint.com/p/1287.htm
    • 0xdb
    • View Profile
    • 0xDB

Re: I Want To Be A Mighty Pixel Wizard

Reply #6 on: February 19, 2015, 02:28:32 pm
I was a little afraid I was buldgeoning you with my brickwalls of text. lol
I hope you do not mind me not replying to everything in detail but I read it all and hopefully can apply it. It is very helpful.

This time I also have an update to share:
  • removed all the shading to work on other issues before tackling that again
  • worked on head shading a bit in the rotated frames, emphasizing roundness of the skull with more darkness and tweaked cheekbone definition
  • de-symmetrized arms and shoulder movements and tweaked girdle tilt & rotation to improve natural look & feel (balance)
  • trimmed the whole robe a few pixel rows to address the shrink-to-skirt issue
revision #07:

Offline Mr. Fahrenheit

  • 0010
  • *
  • Posts: 326
  • Karma: +0/-0
    • View Profile

Re: I Want To Be A Mighty Pixel Wizard

Reply #7 on: February 19, 2015, 08:29:41 pm
The dysymmetrization of the shoulders looks weird to me. It looks sort of forced (?) i'm not too sure :blind:. I also sort of feel hes like a master/sensei/wizard etc and he would probably have the moves down perfectly.

Offline wzl

  • 0010
  • *
  • Posts: 298
  • Karma: +1/-0
    • View Profile

Re: I Want To Be A Mighty Pixel Wizard

Reply #8 on: February 19, 2015, 09:04:46 pm
speaking of shoulders, they're on the same height when the arms are lifted. normally they tend to move upwards a bit, the joint itself and also the muscles get flexed a bit. They look pretty stiff right now

Offline 0xDB

  • 0011
  • **
  • Posts: 873
  • Karma: +0/-0
  • Dennis inter-is.
    • dennisbusch_de
    • http://pixeljoint.com/p/1287.htm
    • 0xdb
    • View Profile
    • 0xDB

Re: I Want To Be A Mighty Pixel Wizard

Reply #9 on: February 19, 2015, 10:17:30 pm
The dysymmetrization of the shoulders looks weird to me. It looks sort of forced (?) i'm not too sure :blind:. I also sort of feel hes like a master/sensei/wizard etc and he would probably have the moves down perfectly.
Mmhh... yes it looked a bit forced to me too but I was not sure if it really did or whether that was just because my eyes had grown accustomed to how it was before, however considering this...

I was also thinking (that thought came after working on it for a few hours) it looked like it might be part of some very controlled ritual, a part of a complicating spell-casting sequence where he has to concentrate on being very precise/exact with the timing and movement of each body part, even with the eyes (so I added that), or the spell would fail.
...I decided not to try to remove the forced look but rather to emphasize it and make it as a part of the complicated spell-casting-sequence where each body movement no matter how seemingly insignificant matters, so in revision #08 I emphasize the tilted shoulders with additional facial expression (this is also a bold and comical take on "To keep balance any time you tilt something to the right, something else has to tilt to the left.". There was nothing else near the shoulders that could tilt the other way so I choosed to balance that tilt with a whacky mouth/eye/cheek distortion.

speaking of shoulders, they're on the same height when the arms are lifted. normally they tend to move upwards a bit, the joint itself and also the muscles get flexed a bit. They look pretty stiff right now
Good point, reworked the shoulders to remedy that in revision #08.

revision #08 (left) and #07 (right):

Offline ErekT

  • 0010
  • *
  • Posts: 330
  • Karma: +0/-0
  • fistful of pixels
    • View Profile

Re: I Want To Be A Mighty Pixel Wizard

Reply #10 on: February 20, 2015, 11:20:21 pm
He still seems off-balance to me. This is what I see more or less looking at your ref:



Good stuff otherwise :y:

Offline 0xDB

  • 0011
  • **
  • Posts: 873
  • Karma: +0/-0
  • Dennis inter-is.
    • dennisbusch_de
    • http://pixeljoint.com/p/1287.htm
    • 0xdb
    • View Profile
    • 0xDB

Re: I Want To Be A Mighty Pixel Wizard

Reply #11 on: February 21, 2015, 08:52:18 am
He still seems off-balance to me. This is what I see more or less looking at your ref:
Good stuff otherwise :y:
Revision #09 has improved balance.

Also started adding the folds. I re-read the Hogarth chapter on folds before I tackled that and spent some more time watching folds in motion and observed how folds are created. This seems to be more important to do than looking at static references for without that knowledge about motion, it is not possible to determine how the folds in a static image came to be (so this is relevant not just for animation).

According to Hogarth, folds are the result of kinetic forces. They are created by four main types of movement in a body which are: stretching, bending, twisting and rotating.

He says as a result of those base movements, the following nine classes of folds can occur (in all kinds of complicated combinations): stretching-folds, bending-folds, cross-folds, pressure-folds, fragmented-folds, hanging-folds, enclosed-folds, flight-folds (in wind and water) and passive-folds.

So, when doing folds, even in a still image, it is always necessary to think about motion and force for without those, there would be no folds. On earth, one force is always present and that's gravity. So as I do my folds, I constantly aim to keep thinking about how gravity tries to pull down every point (pixel) across the whole surface and whether it is allowed to fall or whether is is blocked either by something underneath or because of being attached to a part of fabric from which it hangs. That way, at least passive- and hanging-folds basically create themselves for as long as I manage to stay focused. Interruptions seem to destroy the mental model of gravity and fabric-thinking-mode so it always takes a bit of time to get back into it after a break.

In revision #09 there are stretching-folds (when the knees pull out), hanging-folds (in the starting position and while the knees return to the center) and fragmented-folds (where the stretching folds interfere with the hanging folds).

I find it very useful to think about the nine fold classes as presented by Hogarth. I feel like there is some overlap between them though and it is not often easy for me to put some fold that I see in real life into just one of those classes but while I draw/pixel/invent them, if I find myself drawing a fold that can not be put into any of these classes at all then that seems to be a good indication that the fold is probably bogus and should not be there.

revision #09 and #08:

Offline Conzeit

  • 0100
  • ***
  • Posts: 1448
  • Karma: +3/-0
  • Camus
    • conzeit
    • View Profile
    • CONZEIT

Re: I Want To Be A Mighty Pixel Wizard

Reply #12 on: February 22, 2015, 06:30:24 pm
He says as a result of those base movements, the following nine classes of folds can occur (in all kinds of complicated combinations): stretching-folds, bending-folds, cross-folds, pressure-folds, fragmented-folds, hanging-folds, enclosed-folds, flight-folds (in wind and water) and passive-folds.

So, when doing folds, even in a still image, it is always necessary to think about motion and force for without those, there would be no folds. On earth, one force is always present and that's gravity. So as I do my folds, I constantly aim to keep thinking about how gravity tries to pull down every point (pixel) across the whole surface and whether it is allowed to fall or whether is is blocked either by something underneath or because of being attached to a part of fabric from which it hangs. That way, at least passive- and hanging-folds basically create themselves for as long as I manage to stay focused. Interruptions seem to destroy the mental model of gravity and fabric-thinking-mode so it always takes a bit of time to get back into it after a break.


Thank you for sharing this dennis! I've always felt my way trough folds, and I dont have the language for all of those different conditions, so I probably stand to learn a lot from someone who can explain it in such a detailed way.
You are really bringing a lot into the table by not only listening to advice but going out of your way to learn MORE than what you're being told AND most importantly then share back with us what you learnt =D

I think I'm gonna look up Hogart too! He's got a whole book on folds, so I imagine you're reading another book with only one chapter on folds, right? please share which one =)

on your newest revision: that looks like a very detailed fold map! I think it shows that you've got a lot going on in your head, you're obviously learning a lot =). one thing that I would change about the way you're doing folds though, is that sometimes when two folds meet they dont add up, instead they erease eachother, sometimes with the weaker fold surviving but a little twisted by the other one's force...I think you might see some of that in your ref if you look for it.

in general this version looks like a fold map, which I'm sure willl be very useful for learning, but when you're done with it you should think of how the light hits things and which folds would be hidden by either too much light or too much shadow and try to erease some of them away. Not only for being realistic with the lighting but for a design reason, because as it is the fold map would be too busy for a finalized sprite, specially because you've got a very hi-contrast image here with only those two shades. Dont worry about that yet though, I think assimilating all this information is good work so do what you have to do. You could also add more shades later so that the information is still there but isnt as overwhelmingly contrasted.

EDIT: made an edit BEFORE I may get too busy :p

I went and contrasted up your refs, to something that still shows all the detail but is close to the level of contrast of your sprite.



I went and did the lighting thing on it. Mostly you had good folds going on, but I think the lack of contrast in your ref made some of the folds too regular, specially the big ones in the middle of the first frame. you might notice I removed a lot of the folds at the botton, it's ok to not show the whole line for the fold, just the corners, specially if it's in a dark shadow or really bright light

Also, I went and looked up the folds guy. His drawings seem from a universe where every fold that could possibly ever happen, does. Makes a lot of sense that he did a whole book on how to make them, because his illustrations might be famous for them, but his folds are more abundant than in reality, it's not even in the levels of roman or greek statues, it's closer to 90's comics fold crazyness. I removed a lot of weird insinuations of folds you had on the tighs, I think you were trying to communicate the crumpled fabric but it kinda looked like a glitch or something.

So, I think you might be getting a bit too many folds from trying to think like him, keep his principles in mind if they're good for you but with a grain of salt and compare to your refference, you should deffo not have more folds than your refference, specially because you're animating and specially because you only got 2 shades.

Finally, I think you're still shortening the robe waaay too much. Do you have a thing for wizard knees? :p Remember those vertical folds mean that there's a lot of excess fabric there that will get stretched when he stretches his legs sideways, so instead of pulling the robe upwards the leg will just get the excess fabric from the vertical folds. I only did it up until he lifts his leg cause it gets the point across :p


« Last Edit: February 22, 2015, 10:07:15 pm by Conceit »

Offline 0xDB

  • 0011
  • **
  • Posts: 873
  • Karma: +0/-0
  • Dennis inter-is.
    • dennisbusch_de
    • http://pixeljoint.com/p/1287.htm
    • 0xdb
    • View Profile
    • 0xDB

Re: I Want To Be A Mighty Pixel Wizard

Reply #13 on: February 26, 2015, 08:24:28 am
Hey Conceit, just quickly posting to inform you I am currently taking a break from this and will probably pick it up sometime next week again.

Offline Conzeit

  • 0100
  • ***
  • Posts: 1448
  • Karma: +3/-0
  • Camus
    • conzeit
    • View Profile
    • CONZEIT

Re: I Want To Be A Mighty Pixel Wizard

Reply #14 on: February 27, 2015, 01:14:36 am
haha, no worries. I might get busy tho :p

Offline 0xDB

  • 0011
  • **
  • Posts: 873
  • Karma: +0/-0
  • Dennis inter-is.
    • dennisbusch_de
    • http://pixeljoint.com/p/1287.htm
    • 0xdb
    • View Profile
    • 0xDB

Re: I Want To Be A Mighty Pixel Wizard

Reply #15 on: March 04, 2015, 03:59:44 pm
My mind moved on to something else which means this animation is on ice for an indefinite amount of time, might even never get picked up again, it is completely uncertain if and when. Thank you for your effort.

Offline Conzeit

  • 0100
  • ***
  • Posts: 1448
  • Karma: +3/-0
  • Camus
    • conzeit
    • View Profile
    • CONZEIT

Re: I Want To Be A Mighty Pixel Wizard

Reply #16 on: March 05, 2015, 08:20:28 pm
hehe, I was afraid this could happen if I kept C&Cing every little thing u posted  :o :P

if I were you man, I'd just finish it. Even if it meant shoretning it to just having him raise his leg and put it down in a stomp. I mean look at everything you did, all your neat little steps, your refference pics and everything feels kinda underwhelming to have it there like that. Just so that you dont have this hanging over your head for forever...I've been looking at old stuff I had done and never finished and it always feels like I should've just got it to a point where I could show it, even if it wasnt gonna be as complex as the original plan.

Your choiche though, not that it really affects me eitherway. it was a good refresher for me to write out everything I did. Hope you learned lots  :y: cheers.

Offline 0xDB

  • 0011
  • **
  • Posts: 873
  • Karma: +0/-0
  • Dennis inter-is.
    • dennisbusch_de
    • http://pixeljoint.com/p/1287.htm
    • 0xdb
    • View Profile
    • 0xDB

Re: I Want To Be A Mighty Pixel Wizard

Reply #17 on: March 05, 2015, 09:10:23 pm
Heh, please do not feel responsible, it is not linked to your C&C.

My mind does not feel overwhelmed with this. If anything, the opposite is true, it is bored by it for the time being.

It sees a clear path to the goal and it knows for in order to reach it, it would merely have to go through long hours of mechanically applying knowledge but there is nothing really interesting, nothing to discover about it anymore and what is "worse"(from the point of view of this animation as a living entity which would want me to finish it if it could express that kind of wish), something else which seems more interesting craves its' attention.

There may come a time (soon or maybe a few days or weeks away or never and I am ok with that too(I do not stress out over not finishing one thing anymore before starting the next because it is not important and I allow myself to jump between different activities)) where it wants to kick back and relax and do this type of work again but currently it is occupied and excited about something it deems more interesting (programming related) than this and I have learned from the past that I must trust my mind and allow it to do what it craves the most for when I try to force it to stick to one thing, nothing ever gets done, neither the thing it craves at the moment, nor the thing it considers boring at the moment as it gets caught in a state where it does not want to do the boring thing and does not start doing the interesting thing because of feeling "guilty" for not finishing the boring thing (I lost entire weeks, even months where I neither worked on anything nor really managed to take a break by playing games because of this odd mind state lock-up).

or in fewer words: a different itch wants to be scratched right now :)

Offline Conzeit

  • 0100
  • ***
  • Posts: 1448
  • Karma: +3/-0
  • Camus
    • conzeit
    • View Profile
    • CONZEIT

Re: I Want To Be A Mighty Pixel Wizard

Reply #18 on: March 05, 2015, 09:26:41 pm
's all good then  :P cheers ;D