AuthorTopic: Professor walking  (Read 22298 times)

Offline robotacon

  • 0010
  • *
  • Posts: 222
  • Karma: +0/-0
    • robotacon
    • View Profile
    • It Got Dark

Re: Professor walking

Reply #20 on: June 08, 2009, 10:06:06 am
About reviewing:

I love Pixelation because of the reviewing system. I wouldn't be here if it wasn't for it. I can't stand 15min at Pixel Joint with their judging system.
You owe me nothing, you can review me any way you see fit. I will however answer back to what I think is bad advice.
I'm not searching blindly after good results. If you look at my history I follow advice and post edits. I don't agree that working iterative is wrong.

About process :

In your edit, Helm, you lost the contact points and you disregarded the running speed but I still felt that it was advice I could use because I could see around what I felt was bad practice and listen to the essence of what you said.
From ndchristie I learned nothing and TrevoriuS is promoting some of the worst animation advice I've heard.
You say that I've got the wrong workflow... which gives bad results... so you don't like my results? Perhaps you don't even like my style.

The artwork for this project is very different from what I've done in the past and it's not like any other game I've seen. It might look like pixel art but it isn't.
I haven't got a fixed palette yet but I've been using Arnes 16 colors until just recently when I made a new development palette.
If you want reasons to attack my process I can give to a handful more things like that which you might think is part of an insane process.
I haven't talked much about it at Pixelation because I want to get honest feedback without people asking why I do things backwards.

I don't think it's wrong to animate one limb at a time. I do it in this project more than any other animation project I've done but even for pen and paper animation I think straight (not even or linear) animation between two key frames one limb at a time is perfectly fine.
I think it's useful to be able to go back and make edits. Seriously... my final artwork is about as simplistic as your, ndchristie, animation sketches that you do before going into full rendering. I can afford doing several edits without it hurting me.
I look at the animation as a whole and I am not afraid to redraw things from scratch but I also aim for a unified style. You ndchristie is working on a game where everyone looks exactly the same with different clothes. That's your design decision that works for you but I think i need different types of character models so I work on them all as a whole. You think I stare at one limb at a time when I'm looking at the whole family of characters at a time.

About the art:

I think you've all been right about stiffness of characters but I think you're wrong about weight of characters.
A lot of the time I have only one pixel to play with and I want to use as little sub pixel animation on key features like the head, gun, shoes and other features.
I also want to not use that much anti-aliasing or dithering at this point because the game engine takes care of that.
I'm sure the style has more to do with voxel graphics than with pixelart.
I hate myself for giving this secret away but I'm using projection to color the characters and all the characters have their own skin that can be edited which transforms the entire spritesheet.

The challange:

Make your own original old person walk cycle no bigger than 27x27 pixels.

Offline Helm

  • Moderator
  • 0110
  • *
  • Posts: 5159
  • Karma: +0/-0
    • View Profile
    • Asides-Bsides

Re: Professor walking

Reply #21 on: June 08, 2009, 11:44:17 am
You learned nothing from ndchristie?

He told you that you can't aim while flailing your secondary arm around. I actually preffered your version with the expansive movement in the off-hand until he mentioned it. PRACTICALITY. I enjoyed the flowing animation in the off hand because it added some life to it but if it's at the expense of the animation making some sense then you should go without it. Here's one thing you can learn from his post.

Another thing: weightlessness. I have to concur with him that they don't feel like they have weight because there is no vertical movement of their centers of mass, their torsos and pelvises. The images look like they're gliding not only because they're over-smooth and without any pendulum motion, but linear motion, but also because the limbs support centers of mass that are totally static. Here's another thing you can learn from his post.

Trevorious gives the worst animation critique you've ever heard? What? His urging to you to start with keyframes? That's animation 101. It's really amazing to me that you'd think otherwise. He tells you to make keyframes and not work one limb at a time and then think about inbetweens when the movement is solid with the keyframes only. What's bad about that?

Also, people are critiquing your process because animation IS PROCESS, it's not pixel art. We can't fix the problems with your animations by nitpicking about pixel placements, antialias, dithering, color usage, you see? This is why I don't critique animation a lot of the time because I am not an expert in animation, I am in pixel art. But people here know more than I do about animation and often they post, and you should listen, instead of telling them you learned nothing from them or that they're talking nonsense.

Regardless of whether you are willing to alter your process due to the critique you get, THIS is the critique you'll get because that's what is hurting your art. If your workflow is going to be unaltered because your project is going to be how it's going to be now, then... well that's a risk/reward issue for you. The problems aren't going to go away. Buttery smooth animation has no PUNCH and to have PUNCH you need anticipation for movement, more violence in the snaps and certainly gravity forces applied to mass, that's how animation is.

Quote
so you don't like my results? Perhaps you don't even like my style.

I actually adore the style! But I don't like the animation results. Don't deflect the critique you're getting here, it's amazingly valid.

Quote
A lot of the time I have only one pixel to play with and I want to use as little sub pixel animation on key features like the head, gun, shoes and other features.

First of all you don't have one pixel to play with necessarily, creative animation could get you to work in a lot of different ways. Second, why no subpixel animation? I see binary jumps of the grandpa head or gun in the boy, up and down. If you'd subpixel it would actually look much better. I think I might try it on the grandpa later.


If your method (with projection and that stuff) precludes using the strengths of pixel art animation to benefit your animation... then work with bigger models and do proper vector smooth animation and use that as a base. We know how to make pixel art animation to look good, and if your process doesn't allow for that, it's self-defeating.

Offline robotacon

  • 0010
  • *
  • Posts: 222
  • Karma: +0/-0
    • robotacon
    • View Profile
    • It Got Dark

Re: Professor walking

Reply #22 on: June 08, 2009, 01:32:14 pm
I didn't say I didn't get any useful feedback from ndchristie in relation to the artwork, I said I didn't learn anything from ndchristie about the process. That's why I put headlines on the different parts of the post to separate what I was talking about. The same thing goes for the valid discussion about weight. It's a talk about the animation right and not the method I use? See... you didn't learn anything from ndchristie either. All he did was say I'm doing it wrong but there was no lesson to be learned, no tips on how to think or anything that I myself haven't considered and processed myself.

If I was practical about animating I would draw ten times less frames than what I've got now. The number of frames needed for each character grows like crazy only because I want smooth animations. I'm not lazy when it comes to animating but with a lot of frames to be drawn I like a simple style. That's why I've got so little rendering and so on.

*facepalm* How can you say I'm trying to deflect the critique I'm getting on my results when that is what I'm fighting for at the moment. YOU deflect from the discussion on the results to a discussion about method. If we are going to talk method for gods sake post some examples or references or anything for me to bite into. Talk is cheap.

Using key frames is animation 101... Am I not using key frames? I've got two mirror key contact frames on frame 1 and two 7 with matching passing positions on frame 4 and 10 plus two extra key frames that double for jumping frames. That's an animation technique I haven't seen anyone use in gaming before. When jumping I keep running the cycle until I hit the jump-frame and then I freeze until I land when I continue the run cycle. That makes for really smooth jumping action. Also I'm currently working on easing in and out of running using different running speeds so I have a loop for 4 pixel running, 2 pixel jogging and 1 pixel walking that I can switch between seamlessly. All that would be impossible without using key frames. I also animate by doing the key frames first and then doing passing frames and adding the in-betweens after that.

Now TrevoriuS is telling me I should draw the hip first and work from there to the torso and onward which I think is terrible advice. He says you should finish your key frames entirely before working on and not animate limb by limb. I on the other hand think it's justified to animate limb by limb. You seem to think that I don't consider the animation as a whole but I do. I plan and project how things should move and then I do straight animations which might need some touch up at the end but that is not as big of an issue at the size I'm working at. It's not the same as some people do when they draw everything but the head and say they are going to add it later on which I think is really bad practice It's not the same as drawing everything but the arms and say that you want to finish the legs first. I iterate through the entire body but I don't work one frame at a time. I work similar to what Richard Williams suggests in the Animator's Survival Kit where he discusses Straight Ahead animation and Pose to Pose animation and how you easily can combine the two different techniques first getting the general movement and then do separate runs on different body parts.

You think you know how I work and then you tell me I have a bad method when in fact I just did a crappy animation because it's not easy to animate things moving slowly. Try for yourself, it's much harder doing a slow walk than doing a fast run.   If you point out errors I've made I will most likely change them because I listen to advice. If you tell me I should change my method then you have to showing me your way of working and then I might adapt some of that but without showing any examples it's all talk.

You say you don't know that much about animation but you defend ndchristie and TrevoriuS over me so you have confidence enough to trust that they know what they are talking about and that I don't.

I think my method rules! There's not a problem that I can't fix, because I do it in the mix!

Offline Helm

  • Moderator
  • 0110
  • *
  • Posts: 5159
  • Karma: +0/-0
    • View Profile
    • Asides-Bsides

Re: Professor walking

Reply #23 on: June 08, 2009, 01:46:04 pm
Your keyframes are washed in with the tweens because you don't designate priority to them with punch and with anticipation. It's all a linear fluid mush of movement. If you like it, that's great. If you put it up for critique, expect people that don't like it to point it out and suggest ways to ameliorate. The reason your animation technique is not common in gaming is because it looks lifeless.

You say you consider the animation on the whole and people tell you your artwork doesn't look like you do. Your retort to this critique is meaningless because it will not be stapled to the end game, will it? ATTENTION PEOPLE: I HAVE CONSIDERED THE ANIMATIONS ON THE WHOLE, in the title screen. People go by what they see, not whatever process restricted you to shape your ambition around. The end result doesn't look holistic, it looks like legs moving, arms moving, stable torso. Instead of telling us BUT I THOUGHT ABOUT IT.... address it. More. Different ways. Think outside the box you've put yourself in. If you think this advice is useless because I'm not telling you in exactly WHAT ways to think outside the box then you misunderstand the theme of this critique.

I think your game will be great because it's clear there is vision and commitment involved. But if there's too much vision and too much commitment to it then you don't need critique from third parties, it's clear it's distressing you and you're stepping out of line in your responses. Your game would be greater if you were willing to try different processes.

You're being quite rude to people by dismissing their help that they offer out of kindness and in their free time. You're telling me what I did and didn't learn by another user? Being told there is an error in your work is learning something, being told avenues of approach on how to fix it is another, and you've been given both in many posts in your thread. Stop telling me 'talk is cheap'. What this board is here for is talk. We don't have to do better than you to tell you that there's something wrong with your animations.

Here's a grandpa animation I did, anyway:

« Last Edit: June 08, 2009, 02:18:18 pm by Helm »

Offline hsn2555

  • 0010
  • *
  • Posts: 112
  • Karma: +0/-0
  • Hakuna Matata
    • View Profile
    • HsnGoneWild Gallery

Re: Professor walking

Reply #24 on: June 08, 2009, 02:18:42 pm
o.o robotacon ,chill man

the old man is holding a stick, so his knee is probably supposed to bend.
there :
« Last Edit: June 08, 2009, 02:22:48 pm by hsn2555 »
Be Like no other

Offline robotacon

  • 0010
  • *
  • Posts: 222
  • Karma: +0/-0
    • robotacon
    • View Profile
    • It Got Dark

Re: Professor walking

Reply #25 on: June 08, 2009, 02:51:13 pm
Helm:
What do you mean my key frames are washed in with the tweens?

What kind of anticipation is it you're looking for? I'd say that the contact frame I call the heel-strike where the heel is just about to touch the ground is an anticipation frame.

You say it's all too linear and in your edit you removed frames to make it less linear but by doing that you totally destroyed the step length. I'm aiming for linear movement of the feet on the ground. Anything else would need me to move the character at an irregular speed in-game. Now for the Professor that might actually be a good idea because then you could make it look like he had a limp but for the running guy it's out of the question.

You and the others say the torso is too stable and I'll look into it. How much does the torso move when you run and so on. From my experience the head does not move that much when you run/walk and the head is mounted on the torso so it shouldn't move that much either but I'm willing to try it out. There's definitely some twisting motion that I might put in there. It was an error to add vertical head movement to the old Professor. I was going to something cartoony and it failed.

hsn2555:
Thats a great idea. It breaks the monotony of the upper body elegantly.
I thought it was going to look too wild but it looks great. I'll see what I can do.
I think I know how to use the bending of the knee to delay the step and then take back the loss in speed later on so that the overall speed is constant. Nope it doesn't work but I might add an irregular walking speed.
Thanks!
« Last Edit: June 08, 2009, 02:59:58 pm by robotacon »

Offline hsn2555

  • 0010
  • *
  • Posts: 112
  • Karma: +0/-0
  • Hakuna Matata
    • View Profile
    • HsnGoneWild Gallery

Re: Professor walking

Reply #26 on: June 08, 2009, 03:17:07 pm
but i still think you should redo the shading,
follow helm's shading
Be Like no other

Offline robotacon

  • 0010
  • *
  • Posts: 222
  • Karma: +0/-0
    • robotacon
    • View Profile
    • It Got Dark

Re: Professor walking

Reply #27 on: June 08, 2009, 03:27:27 pm
I apologuise for being rude, especially to TrevoriuS. Sorry.

Offline ndchristie

  • 0100
  • ***
  • Posts: 2426
  • Karma: +2/-0
    • View Profile

Re: Professor walking

Reply #28 on: June 08, 2009, 04:27:36 pm
I didn't say I didn't get any useful feedback from ndchristie in relation to the artwork, I said I didn't learn anything from ndchristie about the process.

Sorry, I thought I was being clear.  The traditional animation process involves beginning with simple wireframes, scribble- or block-men and animating though the keyframes to the tweens until you have a set motion. 
I generally begin with the spine and two axes for the hips and shoulders (whose motion will always oppose each other, because they are connected), followed by 'leading' parts such as two-part stick arms and legs and a block for the head and chest.  then come the following parts which often include hands, feet, hair, tails, etc - things that flow behind, but in some animations (say a character is dead, or being dragged) this can include any limb, and in some animations (attacks and other determined motion) things like the hands and feet can be leading.  I also tend to plot on paper a spacing*. I also like recently to pull out a basic music-writing program and set beats out for frames to see if it "sounds like a ______," because when planning frames you can never engage too many senses.

*A spacing might look something like this:
.1_.2_.3_.4_.5_.6_.7
x | - - | - o - - | - | x | - - | - o - - | - | x
the numbers above are evenly spaced frames, the x is the start of the cycle and o's are other key frames, with |'s being tweens.  I and most others use a static frame length so this is actually not a timing chart, but a sense of where in the animation the frames are.  This example shows a cluster of frames around the step with fewer around a slightly advanced passing.

Even if that wasn't clear though, I made a number of direct references to the animation that I think were quite clear.  He can't move his head forward without moving his shoulders or having an elastic spine - you should, if your goal is believability, move the body with its parts or you'll end up breaking it.  He can't move one shoulder without moving the other, which would throw off aim - either lose the motion on the off-arm or sacrifice aim by moving them against each other the way our skeletons demand--- I didn't leave a prescription here because that choice is left to you.  I would LOVE to play a game that only allowed you to fire accurately by timing your shots to when his aim is true - that sort of S&M game design has a huge following.

Talk is cheap.
No, that reply is cheap.  Everyone here owes you nothing and is making a commitment of time to your growth.  That's not cheap.  What also isn't cheap is listening to others and being respectful.

Using key frames is animation 101... Am I not using key frames? I've got two mirror key contact frames on frame 1 and two 7 with matching passing positions on frame 4 and 10 plus two extra key frames that double for jumping frames. That's an animation technique I haven't seen anyone use in gaming before. When jumping I keep running the cycle until I hit the jump-frame and then I freeze until I land when I continue the run cycle. That makes for really smooth jumping action.
This is a fantastic trick I'll admit but you should also know that it is pretty common.  Since you're an expert on my project, you should know that I did the same thing about a year and a half ago.  Squaresoft did it a decade and a half ago.  I'm sure there are dozens of other examples.  I'm not trying to bash you, you've hit on a good trick - I'm just trying to point out the benefits of looking around include not laying claim to discovering common practices.

Now TrevoriuS is telling me [...] Richard Williams suggests [...]

What you're doing is actually worse than leaving the pieces off, you've got them there, static, interacting with the other parts of the animation in ways that they will not later.  Moreover the pieces you leave static, like the spine, are central to and defining of all the body's motion.  And Richard Williams would never suggest neglecting motion in favor of the parts; that's something that needs re-reading.  What he's talking about in that section (...things that have to be there, like anticipations...) is a highly developed framework before he gets anywhere near working on the different sections, and by then he has so many guides and hit-points that he KNOWS the motion is going to work.  He even references watching a video of just his essentials.  He moves into the straight-ahead when he hasn't got any room left for error, if he's been watching himself.  You on the other hand have posted more than one piece that has nearly finished motion on the legs without a moving torso.  That's not what Williams is getting at.

You think you know how I work and then you tell me I have a bad method when in fact I just did a crappy animation because it's not easy to animate things moving slowly.
You have plenty of examples here of your process.  Internet forum is unique form of time-based media which allows actually for great insight into a person's process on a piece.  That's what most of us are actually here to study.

Try for yourself, it's much harder doing a slow walk than doing a fast run.   If you point out errors I've made I will most likely change them because I listen to advice. If you tell me I should change my method then you have to showing me your way of working and then I might adapt some of that but without showing any examples it's all talk.
I don't remember pixelation as ever being the place for some sort of fantasy showdown, winner-take-art, nor is issuing double-dare challenges the way to learn anything, but if you're asking for an example of the methods I'm proposing as further explanation of technique, I'll be happy to provide one.

Good advice also does not need to come from your betters or even necessarily from a skilled mouth.  Someone who speaks from experience, training, or with regard to fundamentals will have a better chance of offering good advice, but this isn't final fantasy, you don't get double experience points just for battling a higher-level animator.  Good advice must be judged on its truth and perceptiveness alone and not the speaker's ability to implement it.  Find me someone who always does exactly what they know is best, flawlessly, and I will show you a fake.  Show me someone who is inexperienced but whose advice should still be valued and I'll show you 6,000,000,000 others just like him.

Quote
What kind of anticipation is it you're looking for? I'd say that the contact frame I call the heel-strike where the heel is just about to touch the ground is an anticipation frame.
Anticipation is self-evident, you can't just call something anticipatory and have that become truth.  As it is, you're about half-right with the contact frame; you've got some interesting stuff on that frame but it doesn't define the motion of the piece in any notable way (and it should).  For the record too you're including two frames in your description - one where the foot is off the ground and one that is contact.  These are distinct and both very important.

Quote
Now for the Professor that might actually be a good idea because then you could make it look like he had a limp but for the running guy it's out of the question.
Sometimes characters have different needs, as you've stated yourself:
Quote
I think i need different types of character models

Quote
A lot of the time I have only one pixel to play with and I want to use as little sub pixel animation on key features like the head, gun, shoes and other features.
I think you misunderstood me.  My criticism was that you just grabbed a whole chunk of animation and dropped it a pixel without other consideration and called that adding weight.  that's not the same as animating a sinking, weighty motion.  I don't think you should move that chunk of animation more than a single pixel, i think you should animate that section properly with an integrated and anticipated sink of, perhaps, a single pixel.



As for what you've referenced in my work, you're right, I had specific design reasons (evolving tactics job class system - very common) which demanded I keep the characters of a uniform model type.  Beyond that, I'm not pretending to be a genius, nor am I saying that work I've done in the past, particularly what you're looking at (stuff that's over a year old in a young career) is perfect or relevant to what pixels you should move.  There's a lot of problems with that stuff, that's why it was posted for critique.  If anything I like to look at that stuff and enjoy how I've grown in such a short time since because I have been open to the advice of those around me in the weeks and months since.
A mistake is a mistake.
The same mistake twice is a bad habit.
The same mistake three or more times is a motif.

Offline Helm

  • Moderator
  • 0110
  • *
  • Posts: 5159
  • Karma: +0/-0
    • View Profile
    • Asides-Bsides

Re: Professor walking

Reply #29 on: June 08, 2009, 07:36:28 pm
Helm:
What do you mean my key frames are washed in with the tweens?


Again, I am not an animation expert, but if you allow me to extrapolate from what my eyes see the thing is this: keyframes need more screen-time than in-betweens. That doesn't just mean that the actual keyframe needs to be on screen more, it means that it, and all the tweens that are very close to it, if grouped, make up a lot of screen time. To explain, imagine a pendulum motion. Wait, I can actually draw this real fast.



This is what you're doing. You might have started with the drawing the extreme frames, which you at that time called keyframes. But then you evenly distributed frames between the keyframes because you think that's the best way to do even movement for terms of gameplay.


(watch this in your animation program, firefox doesn't like 10mlsec frame animations for some reason)

Here I introduce timing to the same loop. See how there is some 'grouping' around the stress frames? This pendulum motion makes the keyframes get more screen time, and it enforces how human beings understand fast motion, which is a matter of survival. We don't look for something that goes evenly fast, linearly  in our horison line, we look for something that goes fast, STOPS AT A POSE WE CAN DISCERN, then moves fast again. This is how hunting movement is, this is what signals the eye to follow. We don't follow every bird in the sky with our eyes instinctively, but it's a point of survival to follow birds of prey that make their killing dives or the cougar that will stop on a dime and lounge.

However personally I find this still has too much inbetween 'fluff' that muddles the pendulum motion. More frames doesn't mean better animation. I suggest this:



Yes I am aware that in gameplay terms this will create a bit of 'glide'. So what? The gameplay expense is minimal and what you gain is more lifelike motion that convinces the eye and suspends disbelief. With your animation style I am wondering if there's even GRAVITY in effect in your gameworld... isn't this a bigger gameplay problem than a few pixels worth of glide? Have you checked to see if the glide is even visible in fast movement?

Quote
What kind of anticipation is it you're looking for? I'd say that the contact frame I call the heel-strike where the heel is just about to touch the ground is an anticipation frame.

Anticipation isn't just a name you give to the few frames before your stress frame. They have to feel as if they're anticipatory movement, to clue the brain in about what's going to happen. That's what makes animation exciting.

hsn2555's edit is really great.