Sew's Art Blog

(Totally wrote most of this last night and forgot to upload it entirely, heh)

Okay, completely forgetting about making the idle animations was a huge oversight on my part. I was pretty angry (new blog name: angry artist?) at myself and the situation for about a day or so. But let’s just all relax a little and look at this objectively. The ideal situation for this is to take all the pieces from our existing running animation and tweak them to be an idle animation. The benefits of this approach are many, since it means clothes will not need to be redrawn for idle animations, they’ll just magically happen and I can rest easily. But is it do-able?

The short answer appears to be no, it is not. The joints created for movement really don’t look very good stationary with subtle motions attached to them. So what does this all this mean? I’m glad you asked, because I’ve made a fairly redundant chart to help us all out!

The base animation is just what it seems like. The base nude animation, ripe for the layering of everything else over.

The Full Base Set is also straight-forward. It’s the same as the base, but for the other directions. The good part about this phase is that I only have to draw two bases and copy paste them, since the nude has no reason to need to be asymmetrical. They are each broken out into their own pieces (east head, south head, etc etc), that way each of them is kept independent from one another, even though they look the same at this point.

There is the secret ‘in-between’ step of designing these outfits that I will discuss in greater detail in the near future. For now though, the next step is to draw each outfit item 4 different time (if need be) for each direction. Some items won’t need to actually be redrawn and will just be copied over, but the majority of interesting outfit designs are asymmetrical and will need to be drawn for each direction. It’s also important to remember that this is for each item (shirt, pants, shoes, hair). This part really doesn’t take along as you might expect. I tried my best to expedite this phase by creating a generic base for each direction. The base dictates where the edge lines for each object need to meet to mesh properly with the other objects, such as upper-arm to lower-arm.

I’ve said it before, but the best part of this entire process is that once the bases are completed, they never really need to be touched or dealt with again. This is true of all bases, male/female, running/idle. So the big question now is, how far along am I in this process and how much longer until it’s finally over?

Trick question, since outfits could in theory go on forever. I don’t have a current goal for outfits before I walk away and work on another part of the game, so I figure I’ll just draw as many as can until I get tired of them. I’m actually really looking forward to lots of outfitting, working on apathy felt like doing real art again, though it was broken up quite a bit by tweaking how the base looked.  Anyway, the idle animations are a lot more simple than the running (and any sane person would have done them first), and I’m about halfway done with them now, so the full set should be finished by the next blog post. Then it’s just a matter of making them ‘fits. I should also take just a second to mention women since I didn’t add them to the chart. They are basically just a remap of the base with their own set of outfits. They will require a little bit of tweaking, I’m sure, but for the most part it’s a straight forward process of swapping out the man-ish parts with more lady-like parts. (and of course dividing them into their appropriate folders based on direction and blah blah blah).

So ETA? As long as I didn’t forget about some other huge part of character map animations, this is the pre-penultimate post on the technical of the animation. The post Sunday will be about the finished idles, Wednesday I’m planning on talking about the designing of the character outfits a little, and following Sunday we should be looking at the some finished sets.

Animations here are probably the biggest part of the game for me. The other large part is going to be making the enemies and then animating those enemies, but I’m much, much less worried about them. Why? Because they don’t need to be modular like character animations do. The enemies aren’t created in such a way that the player can manipulate how they look, so if something doesn’t look right when I make it or animate it, I can just redraw that part. There are no underlying pieces that I have to get just right so they’ll work properly with all subsequent pieces layered, because those other pieces also won’t even exist. I’ll actually be quite relieved when we get to that point.

See you Sunday!

Everything has been redrawn, characters will never be allowed to be shirtless because the waist-band idea only partly worked.

Faces and some asymmetrical tweaks pending. His head is a little too big, etc etc.

I want so desperately to be out of animation town. I realized about 20 minutes ago that I completely forgot about idle animations. Completely and totally. Oh, goodness. Argh.

I now officially have a headache.

Shame Post: There will be one Wednesday, I have no idea what will be in it right now. Idle animations? The female animation is almost finished, so that, too? I don’t even. I don’t even know right now.

I clearly did not updated on Wednesday when I said I was going to. I live in tornado alley (incidentally, right below one of those three tornado graphics), and things got a little weird this past week. But let’s get right down to it.

My plan that I was so sure was going to work, actually didn’t work out the way I needed it to. The idea was to have a base layer with all the un-clothed character animations in it, and then to have a layer directly over that one with all the clothes, so the clothes would always appear on top. The result is something like this.

Obviously having EVERY piece of clothing appear above every piece of base caused this fun layering issue. So I I had to re-think my layers and ended up merging the two. This works a lot better because I can manipulate piece hierarchy between sets. For example, with the nude set, I want have the legs appear above the main body piece. But with clothes, I’d like for the body-shirt to appear above the legs. It was a simple fix with a little folder move. The result of merging the layers ended up like this.

The behind the scenes stuff was me re-figuring out how to make those buttons work for clothes and things. It’s basically the same script I wrote for changing color, I just had to edit it a little differently. It actually took me 3 tries before I got it working the way I wanted it to. The way clothes were positioned above the  base was also a cause of some frustration because the pieces didn’t quite snap the way you’d expect them to. It required more tweaking, especially the joints for the knees/elbows. I think I got it worked out, though.

Since everything seemed to be working in practice it was time for me to make sure each section was arranged correctly, and then begin moving and merging the base and outfit layers for the other 3 directions. But before I did that, I wanted to make sure everything was working properly, so I didn’t have to come back and manipulate things again. I decided to try and make ApathyTeen proper.

I am again frustrated. Everything looks OK except the feet are really bothering me and I can’t seem to get them to look right. My mind began to wander a bit. This next part has less to do with the direct art process for the game, and more to do with my current state of mind and the process of me creating things in general. (You are welcome to skip it, I italicized it for easy skippin’!)

I became a little disillusioned with myself earlier this week when I was fixing up my redone portfolio and decided to put a date on each of the images I had up there. Of the 11 images I edited it down to, eight of them were done in 2010, with one from this year and one from 2011.

My first instinct was to go to my art directories from 2011, 2012, and 2013. I was surprised when I realized these directories were mostly empty. Have I really just not been making art like I used to? My ’11 and ’12 directories combined have eleven images in them. In contrast, I have 150 images from 2010 saved (and each year before that seems to average between 60-80). How is it even possible that I did so little art the past 3 years?

I have my first art for FairWeather marked as around May of 2011, though that was around a weird time in my life where I had just moved and was looking for a job and was fairly busy. I didn’t start working on this project intently until about January of 2012, and up until that point I wasn’t really able to get much art done for anything. Since Jan of ’12 though, I have 335 saved images in my GameMake directory. Some of these are still images of plans, some concept art, flash animations, some are slight edits to other images, so I’m not suggesting that these are all fully finished pieces of artwork, but these 300 plus images are just the ones I ended up saving and sharing. I felt a little better when I realized that my art production hadn’t slowed down over the past few years.

However, I kind of feel like a fake at this point. I feel like any other artist would have been done with this project by now, or at least have a lot more to show for it. Even compared to any other project I’ve ever worked on, I’m getting an alarmingly small amount of work finished. And it’s not from lack of time spent working on it, I can prove that. Looking at the numbers, I’m perplexed at how I can spend so much time and get so little done. It was when I got to this point in the animation process that I started to understand a little better.

At this point I’m frustrated with the entire design of everything, but that reminds me of what Derek Yu wrote, “stop making excuses for starting over”. I will admit that I have restarted the animation process several times, and am always extremely upset with it, but it’s not just that a shoe isn’t lining up properly. It’s that I feel like I’ve bitten off more than I can chew on the animation end of things. I’d say “on the game as a whole”, but I don’t believe that to be true. Nothing, for the most part, about the other aspects of the game has me worried or stressed out. I want to get this done as quickly as possible, but I am seriously starting to wonder if my design and style choices for animation was holding me back.

Having said that, all that’s left to figure out is the shoes and how to get them to display properly. The framework for everything else is almost in place and pieces are easily swap-able. There are a few fringe instances. For example, I will need to add an extra piece of clothing for the pants that wasn’t necessary for the bases. It will serve as a waist line piece, fixing every pair of pants on a shirtless person looking like a pair of chaps.

It’s almost done. The skin is changer is broken in this one because of how I have the not-seeable outfits layered. But it’s really close. To give you an idea of how much time making this framework saved me, I made the entire outfit set for apathy that you see in about 20 minutes, and that includes having to adjust some of the pieces slightly so they’d line up correctly.

Work smarter, not harder.

Goal for next week: I’m kicking it into high gear to cross the map-animation finish line. Finished male & female thin running animation sets with however many outfits I have time to make. I’ll run some tests of these new graphics on the current map-tiles and we’ll see what adjustments need to be made, and then possibly move back into tile battle work.

This has given me more trouble than I anticipated. I started out with the initial sketches, as usual. I made two of them, but I knew I had to go with the 8 frame one for consistency. These took way way longer to get to even this rough state than I care to admit. The back-view running animation at this isometric angle is seriously my least favorite thing in the world. The good news is that I only have to get it right once. The bad news.. is that I’m still not great at it.

Before I dig into this too deep, I realized waaaay after the fact that the reason my 12 and 14 frame animations looked so slow and silly was because I am displaying the whole thing at 11 frames per second. Meaning anything more than 11 frames would take longer than a second to play, as opposed to the 8 frames version which would play at least once a second. Pretty straight forward stuff here, dunno how I missed it. This realization changes nothing though, as I’d much rather go with less frames to make things go by a little quicker. Just some basic stuff I’m going to need to keep in mind with some later animations.

 
On the Action Script 3 end of things I found a much easier way to program the whole thing. Variables and for-loops n’ what not. Things I’m not nearly qualified enough to cover. Just know that I’m way happier and turned 100+ lines of code into 22.

The process was basically the same as what I covered before with a few exceptions. I was not expecting to run into any layering issues in this process, but boy was I wrong. Take a gander.

The head piece has the neck attached to it, and the body is resting on the layer above that. It wouldn’t have made any sense at this angle to try and do it the other way with the head on top, since the head keeps bobbing in in front of and behind his back. This would all be perfect, except in my initial sketch you’ll notice that I have the arms moving behind the head. Well, layer wise I need the arm to be above the body since in a few frames it appears above it, and then I’d also need it below the head. This is obviously impossible and has been a great cause of frustration for me (I promise, eventually I’ll have a week with no frustration and just smooth sailin’)

I’ve attempted to change the way the arm swings to make it look less ridiculous several times (and in the process completely jacked up the back arm), and I’m sure there is a way to handle this where the head and the arm never cross paths. I’ve spent so much time working on adjusting their movements that I didn’t even invest the time to draw a proper hand for his back arm. Nor have I taken the time to adjust their feet yet, though adjustments like that are really very simple if I can get the base animation to play nice.

Having said all that, we only fell a little short of our goal, though I am cheating a bit here. The bases for the bottom two animations there are obviously just mirrored versions of the upper two, and that is mostly what they are going to be in the end. Right now it’s literally just borrowing the symbols from South and North and mirroring them, but to do it properly I will have to make new symbols that are named properly and exist in their own special set. This is important for asymmetrical designs in outfits and what have you. It’s a simple set up that doesn’t require any extra art, just a little time.

Goal for next week:  Tweak the north animation to not look terrible, properly add in the east and west, adjust the feet and hands, and begin on the default costume. Oh, and since we fell short of the goal?  All of this will be done and uploaded with a special shame-edition blog post on Wednesday. Look forward to the tears!

First off, let me say that there was no good tutorial for how this process even works that I could find online, and I searched off and on for weeks. Some people covered different parts of this idea, but no one used the idea to the extent that I needed to. The new system basically works like this:

Instead of me hand-drawing every frame of animation, I realized I could just draw pieces and manipulate them. What’s even better, is that I could take any of these pieces and swap them out for a different piece that would move the same way as the piece it was replacing. Meaning, I could replace the naked foot with a shoe, and the shoe would move the same way without me having animate it at all.

So in essence, I found a way to draw each piece of clothing only once instead of eight times per direction (4 times total vs 32), whiling making the program do the hard part for me. Granted with this method I do lose a little creative control, but it’s still very possible for me to go in and edit one individual frame if I need to (For example, if the foot needs to look like it’s coming right at us in one frame, and then bending back in another)

I was still super unhappy with the way our animation looked, though. Too stiff, not exaggerated enough. I decided to study some nice dynamic running animations, then try my hand at redrawing it one last time. Don’t let your friends tell you that you can’t learn anything from anime.

I drew it originally at 16 frames and then cut it down to the two versions you see here. I did this so I could ask everyone which one they liked better, but by the time I had uploaded I already knew the answer. After I had my sketch in, I made some objects and slapped them on as a sort of proof-of-purchase so I could know for sure that this idea was going to pan out.

At this point I’m confident in the way the animation looks and feels, so I go ahead and finish up all the base pieces for this direction.

This animation isn’t without it’s problems, either. The feet and hands will need to be adjusted on certain frames to prevent everything from looking so stiff. The important thing is that the framework is here and is easy to manipulate if I’m not happy with how a particular frame looks. And since all the outfits that will be layered on later are based on this base here, it’ll only need to be edited once.

This is the part where things get a little tedious. The images are actually exported individually as PNGs to be imported into the game. It’s a simple process, but it still takes awhile, especially at 32 frames per direction set and per clothing set. Even though I’ve cut out the work of me drawing a pair of pants 32 times, I still end up having to export it that many times for the game to be able to read it. I simply don’t have that kind of time, to export each set and try out different combinations in-game. So I’ve worked on a way a preview these directly in flash. Even though I’ve messed with this idea in theory several times, when I’m trying to actually make it work everything is riddled with bugs and sadtimes.  It’s probably my fault for paying attention during programming 101, or not paying attention when Hawk talks about algorithms and.. math.

At this point I’ve basically redefined the way every object is made by separating the color and the lines from each other and adding them to different layers. This isn’t a necessary step, but it  helps a lot when recoloring specific pieces or adding new ones later. It’s also useful for the way I’m going to be layering shadows onto the base and extra frames.

I managed to get a little more done that just be frustrated at and tweak my animation , redo all the objects in that animation, and program a skin tone changer. I’ve been fooling around with different layouts for our actual game website, and I continue to go back and forth on several ideas we came up with. We thought a background featuring an island with all the elements of the game in was a good idea, but I’m really doing a super poor job of angling it and finishing it. I’m also over-complicating the entire design. This is probably the 4th island I’ve attempted to draw with little luck.

Goal for next week: Finished Animation Bases for the other 3 Directions, Have a base outfit for each of them

(I also finished* redoing my own website.)