The Art Style Pros and Cons

First off, I’ve released a comic book. John Michael Jones Gets A Sword For His Birthday collects the first year of John Michael Jones Gets A Life in dead tree form.

You can get it for cheap on Amazon here, for much less cheap but in color on Amazon, here, or for cheapest of all in color, but as a PDF instead of a dead tree on Gumroad, here. Either of the dead tree versions will give you a discount code to get the Gumroad PDF for free, a feature I don’t expect will see much use for this comic, but it’s there.

So, as I experiment with game development, I have tried out a few art styles and perspectives. I’m going to list them and the pros and cons of each here. I’ve done this before, but not in an organized way. Here we go:

1. Pixel Art

Pros:

I like it a lot. Appeals to nostalgia. Quick to produce. Pixel Art games will tend to run on lower-end hardware. Upscaling low resolutions doesn’t look worse when you get a 4K monitor; it looks the same. Fun things I want to do with my game art like palette-swapping shenanigans are a piece of cake.

Cons:

Oversaturated in the marketplace. Using non-chiptune music is jarring (I love chiptune! But most of the music I have access to for scoring my games is not chiptune). Game is constrained to the 2D plane, except for some hacks that are a pain to implement. Character costume changes that are not palette swaps require me to recreate the animations from ground zero.

2. Hand Drawn Art

Pros

I absolutely love how it looks. Lets my art style shine, which maximizes marketability. Can use with non-chiptune music without jarring anyone.

Cons

Much more labor intensive, meaning games and worlds have to be smaller to compensate. Will not run on lower-end hardware. Palette swap shenanigans are doable, but much more annoying. Will look noticeably worse when upscaled onto a 4K monitor. Not only am I constrained to a 2D plane, because of the extra work to animate multiple directions, I’m basically locked into a side-scrolling perspective.

3. Low Poly 3D

Pros

Less saturated than pixel art, so more marketable, but not as much as hand drawn. Still appeals to nostalgia, though for Millennials and older Zoomers more than GenX/Y. Not as efficient to produce as pixel art, but way more than HD. Can use non-chiptune without jarring. Has access to a Z-axis, so I can put jumping in my top-down games without shenanigans. Can run on a potato, but furnishing upscaled textures for the Elite Gamer with the 4K monitor is simple. Palette shenanigans are a bit harder than pixel art, but way easier than HD. Costume swaps are much, much easier than either.

Cons

I like the art style, but of the 3 it is my least favorite. I have far more practice in 2D, with the attendant far more skill.

Conclusion

They are all good and all have their uses, with 3D constantly taking the role of “first best at nothing, but second best at everything.” However, which one is truly best will depend heavily on what sort of game I’m trying to make.

There are four game concepts that I rotate through. One is a virtual pet that runs on your phone. A mix of HD and pixel art is ideal for that, as seen in the currently existing prototype:

Although a low-poly virtual pet would make certain concepts, like varying monster size based on genetics, more feasible.

The second is a Sonic Metroidvania, starring Jump the Shark or Sparky the Dragon. HD art is probably ideal for that, with pixel art being a close second. But I keep abandoning that project because I don’t have a notion I want to pursue that is different enough from existing Sonic or Metroidvania games to justify years of my life, which is what it will take to make that game. I have a few interesting tweaks to the Sonic formula, but nothing groundbreaking. And no story I want to tell that’s compelling enough to justify a game that is otherwise a clone.

The third is Link’s Awakening style gameplay, but with Megaman X elements such as dashing, starring a robot whose parts you can swap out as you collect other parts.

3D is hands down the rock-star of this game concept. It gives me a Z-Axis, allowing for player jumping in the Zelda perspective without any hacky trickery, and it makes swapping character parts a breeze compared to the other styles.

A subset of the third (and technically fourth) game concept is Hat Trick, which is a story that could be a top-down adventure game, like Zelda, with Megaman elements. Hat Trick as a game or series of games does not strongly lean to 3D over pixel art, but it does militate away from Hand Drawn. And Hat Trick is a strong enough story to justify making a game that doesn’t do anything new, but is just a clone of another game.

Hat Trick can be forced into a Metroidvania, but it naturally tends towards a top down Adventure or a JRPG.

The fourth game concept is a JRPG that borrows gameplay components from Chrono Trigger, the Mario RPGs, and Pokémon. And this last one fits every art style equally well — I think using a side-on perspective like that of a beat-em-up overcomes the limitations of the HD art style. However, it prefers either pixel art or low-poly 3D, to enable a top down view with breadth of worldbuilding.

I am an artist who can code rather than a programmer who can draw. To play to my strengths, I should focus on the 2D HD. But my desire to make beautiful games is less powerful than my desire to make interesting gameplay.

So I’m trying the 3D on for size, with the idea that my top-down Megaman X is the only game that demands one art style above all others, and the Hat Trick games and JRPGs can have their gameplay in 3D and still work. Because at the end of the day, looking good is the second most important goal of a game’s art. The first is serving the gameplay.

Changing my mind on art style yet again. Except not.

So this month I’ve basically settled on spending the next month storyboarding kids’ books, and working on my RPG engine in between, with individual kids’ books becoming the project of focus if I like the storyboard a month after the storyboard is finished.

How, then, shall the RPG look? Well, I have three basic options: 3D Low Poly. Hand Drawn. And Pixel Art.

Each of these art styles has advantages and disadvantages.

Low Poly: More intrinsically dynamic worldbuilding. I can make worlds with multiple layers and not worry about implementing a hacky Z dimension. I can implement jumping in a top-down game. I can switch between top-down and side-scrolling perspectives on a whim, keeping all the assets. Also, it’s more marketable than a pixel art game. Consumers repeatedly express irritation that every indie game and its brother is pixel art. When I make Drone Fu, switching parts for characters as you customize your robot will be super simple. However, I’ve done almost nothing in 3D, and so my experience level is much, much smaller.
Hand Drawn: My hand-drawn art is simply the most unique and therefore the most marketable. Building worlds that look like an illustration is intrinsically rewarding, and rewarding for the player to look at as well. And if I get paid to make games, I have more time to make games.
Pixel Art: I can produce it much faster. I personally love it as much as I love the other styles, even if customers as a rule don’t. The more games I make, the better my games will be, and producing games in pixel art allows me to finish more projects faster, even if I don’t get paid. Moreover, tilemap based games allow for a more malleable world, and the idea of letting the player dig every ground tile, chop down every tree, and so on pleases me.

In all three cases, I’m keeping the hand-drawn RPG interface I’ve been building. Reason being, it works in all three cases. Suppose I make half a dozen JRPGs in a gameboy color pixel style, and then use them to do a team up with a writer, but we create a hand drawn, or 3D RPG. The actual graphics used for the interface will have to be swapped out to fit the game in question, but the HD style and logic will not.

It works least well with pixel art. But it works well enough that I am content.

And that last scenario is kind of the deciding factor for me. I think if I build a dozen small RPGs fast, I will grow as a game maker and create something truly special. If I pick the hand drawn style, I’ll make two, maybe three games total in my whole life, and its a roll of the dice if any of them are truly great. And since I’ve decided that the books are my thing, it’s okay for me to prioritize gameplay and personal evolution over marketability.

So I guess in the manly month of March, I will be making a gameboy RPG with an HD interface. On the side. After I’ve storyboarded a little.

How I make John Michael Jones

If you are following John Michael Jones Gets A Life, either on Bunny Trail Junction or the Mad Christian Mondays Newsletter, this is a long, in depth, rambly discussion of everything that goes into making an episode. Be warned: the episode I will use for most of my examples is scheduled to run in the newsletter Monday, January 23rd, 2023, and on my own sites the next day, so spoilers. If you want to experience the story in sequence, maybe stick a pin in this post for a month and come back. Or join the Mad Christian Discord, where I release each episode as I finish it in the “studio” channel, and get caught up.

If it’s after the 23rd, or you care more about my process than spoilers, Onward!

Onward!

Captain’s Log mc•61: Animation Programs

Awesome Moments 1 is done. You can see on the project status page.

I finished at the beginning of November. I started working on finishing Awesome Moments in, I think, August? Yeah, the announcement was log m811. August started briskly, with two illustrations, four pages a day. September slowed to about one illustration a day. Then October I was lucky to get two or three a week.

This tracks with my previous observations that I can hang onto a project for about a month before it becomes inefficient. And my resolution to basically keep three or four projects in the air at all times so I can switch each month to a different one. If I had simply not worked on Awesome Moments in September, I might have completed it halfway through October.

Instead I dragged it across the finish line in November and, holy shlamoley, I needed a break. I browsed itch.io for game jams, picked one at random, and joined the first team that asked as an artist. Spent a week using some poor group of programmers and a sound guy as a testbed for theories about animating.

The result is Toasty, and here are my notes on the process.

Basically, last year, I took a shot at making an HD game by hand inking character parts, turning them into vector art in Inkscape, and animating them. At first, it promised to be at least as efficient as pixel art, but as the project dragged on, it grew less and less so.

So I gave up and switched to pixel art for a while, even though the hand-drawn stuff has a distinctive look that is hard to duplicate. But for a game jam, I gave Clipstudio Paint’s animation tools a try:

This was a great success. So I tried it out in a Jump the Shark platformer:

But this introduced some problems. Tweaks were hard to make to already established animations. I missed the resolution independence that vector art offered me. I wondered if I penciled animations in CSP, but “inked and colored” them in Inkscape, making them vector art, the lower quality in line variation would be worth the increased customizability of vector graphics. And also I made the Jump Sprite way bigger than it needed to be.

So I tried this new system for “Toasty”

..and I didn’t hate it. It was deeply hampered by the fact that Inkscape is not an animation tool, and so I had no way of knowing until I had rendered an animation out whether I had gotten it right. But I was able to do things like re-palettize the cutlery, and swap their heads for different animations. And I didn’t hate how the art looked, even though the lack of variable line-width flattened my style a little.

It also did well in the jam, gaining 5th place overall among the 52 entries, and 1st for the art.

So I tried reanimating Jump the Shark using the style from the existing game, but according to the new rules. And the whole while, I kept thinking, “man, I wish this was Anime Studio.”

Anime Studio is an animation tool I used years and years and years ago. More than ten, I think. I made some Sonics with it.

This animation is not great. I’m certainly more skilled now. But the problems with the animation are the fault of my inexperience, not the fault of the tool. The tool is fine. It’s a vector art tool that has four huge advantages over Inkscape:

Line width variation is built in and not a pain. (You can do line variation in Inkscape, but it’s weird, unintuitive, glitchy, and laggy).
You can just hide a line segment on a shape. In Inkscape, either the entire shape is outlined, or none of it is, forcing me to make multiple copies of a shape if I want gaps in a line. (There is one minor exception to this rule, but it isn’t very useful.)
It’s actually an animation tool, meaning you can see if what you’re doing works before rendering it out.
And also meaning there are things like bone-based movement and automatic tweening when you want them.

So I said “heck with it,” googled what had become of Anime Studio (it’s now called Moho), and eventually buying a license.

Here’s the idle for Jump the Shark in each of the 3 programs: Clip Studio hand drawn, Inkscape traced over a CSP pencil, and Moho:

Ignore the more pixelated look of the second two animations; that’s an artifact of rendering them at a size more appropriate for the game. As you can see, there’s less line variation in the Inkscape version, but the drawing still has a decent amount of personality, and I was able to redo the feet to make them more consistent with the character design. Moho brings back line width variation, though I keep it light to avoid making more work for myself. The tweens are smoother because I didn’t have to eyeball them. And I was able to tame the bouncing fins a little. I like them, but I went too far in the original animations.

But the biggest deal by far was the time it took me to make these animations. The original was about 2 days of constant work. The Inkscape was 1 day of work. It was actually more laborious than the CSP version, but benefited from that version already existing, and so most of the animation puzzles that come up in making a piece had already been identified and solved.

The Moho version took me a couple hours, maybe. And in half an hour I made a “tired” variant:

So, it looks like I’m going to be employing Moho moving forward.

Now, to use these characters in comics as well as games, I need to export them much bigger than Moho is equipped to handle. But Moho has SVG export. It has kinks, I’m sure — it has to! But we’ll burn that bridge when we get to it.

Anyway, I’m making a book with my kid now!

Once the print proof is prepared and ordered, I will probably try working on another game to nail down my Moho workflow; either rejiggering my Jump the Shark platformer, or else another jam where potentially I work solely as the animator. We’ll also burn that bridge when we get to it.

Protected: Wren Idles

Captain’s Log M8•T0: Ink-Slinging

Awesome Moments has ground almost to a halt. Almost. I can get out an illustration a day most days of the week. I’m only 5 illustrations away from completion, so I’m going to keep pushing forward, but while I’ve debated making a final super push of two or three illustrations a day (these only take me a couple hours to do), I’ve decided no. I’m going to give every picture my full attention, and if I try to force it I’ll be tempted to get sloppy.

The thing that has absorbed my attention this week has been trading cards. I’ve liked cards my whole life. I thought they were fun in the Amber Chronicles. I loved them in Digimon Season 3 (known as Tamers to us Digimon snobs). I didn’t really get into Yugioh or Magic the Gathering, but I wanted to.

Continue reading →

I’m making a Sanic

So, this is what I made this weekend:

To be 100% clear, this is my plan moving forward:

In July, my priority is Awesome Moments 1, the Bible Story book that rushes from Genesis to Revelation, makes the angels awesome, and draws on typology (e.g. Adam looks like Jesus).

At my present rate of two illustrations a day, if I get an additional illustration in every other weekend, I should be finished by August. However, I’m not sweating it. If it takes a little longer, it takes a little longer.

This means any development on a Jump the Shark game will happen on weekends and weekends only, at least until the book is done or I drop it. I’m going to hang onto it with all I have for the duration of July, but if it’s not done in August, I may well stop, leave it for a month or two, then pick it up again and finish it.

Basically, I have ADHD, and it is more productive for me to work with it than fight it. Even on a project I love, my endurance maxes out at about a month and a half before I have to switch to something else.

So… I’m probably going to tinker with this game on weekends in July. Supposing in August I either finish Awesome Moments or drop it, and decide to mainline this project. What happens then?

Here’s my plan for this project for the foreseeable future:

Keep adding controls until I have a fun little character that can freerun around in ways I enjoy.
Once I feel the character control is more or less complete, along with level gimmicks to play off of (springs, enemies that provide specific challenges, etc), graybox some levels until I have a set that I think are pretty good.
Build a game and bring it to market.

That is to say, I don’t have a specific game design I am working towards. I’m feeling my way forward, and I intend to continue doing so for the forseeable future.

My preference in game genre is Action Adventure, so this is likely to turn into a Metroidvania, because that’s what they call Action Adventure Platformers in these benighted days (no shade meant for Metroid or Castlevania, though). But I’m also toying with the idea of just straight up implementing Alexander Hellene’s platformer design because then at least he’ll buy the game.

A Sonic Metroidvania is not very much in the spirit of Sonic, but this isn’t Sonic. This is Jump the Shark.

Anyway, at this point, this project is still in the tinkering stage. I’m 100% playing by ear. If and when a design is finalized and put into production, of course I’ll let you know.

Wren Valen Redesign

Wren Valen was a character I designed when I was single and lonely, and it shows.

But while I’ve been making children’s books for a while, I recently realized it’s really my whole thing. My ideal reader is 9 or 10. He doesn’t need scantily clad heroines. And my branding especially doesn’t need scantily clad heroines. It needs to tell parents, “these books are fine for your kid.”

Which has put me in a sort of dilemma. My wife has long said I need to tell the rest of the Wren stories. I could tell them just for her, as written stories. But there’s nothing in the plots and I have planned that isn’t suitable for children. Just skimpy costume design.

And skimpy costume design is not necessary, even though it fits the character’s personality and activities (she’s basically a sky pirate). For you see, in the first pair of Wren stories, the ones that hooked my wife, she got a shevlar harness at the end. A shevlar harness is a tight outfit that serves as the anchor for armor summoned from the aether. I’ve drawn her a couple times in the past in said harness, with various amounts of other costume over the top of it:

Which leads me to a couple of questions:

What would she choose to wear over the harness? What are the various possibilities? Which would she choose and why? And can I design the harness and/or the outfit she would wear over it to be good to animate?

After all, she went from this: to this: specifically because the skimpier outfit reads better when animated. Arms and legs are clear, the costume isn’t a jumbled mess. And in addition to reading better, it also animates better. The crop-top “woman boxer” look I’ve adopted most recently also conveniently separates out each of the parts I am liable to move independently:

When I animated John Michael breathing, I was able to scale and alter his tummy and ribs differently, as I did with Wren, but it was nowhere near as elegant because I had to ensure his shirt looked continuous:

Anyway, today I decided I needed to start investigating a Wren redesign, so I can write Wren stories in kids’ book format, and thus please my wife, myself, and my customers.

Here’s my working file, with past imagery for reference:

Harness, covered with boots and bucklets
Harness, covered with pirate outfit
Harness, pirate top
Harness, pirate bottom
Harness, light armor
Harness, dress
Harness, cloak

1, 2, and 3 look fine and are reasonable. 5 looks wrong. I’d probably want to replace the leather boots with proper greaves instead of having the greaves go into the boots. 6 is cute, but doesn’t feel like the character. 7… 7 makes sense. Wren wouldn’t want people to know she was wearing a shevlar harness, especially before she gets armor chips for it.

Right now, my favorites are 2, 3, and 7. 7 seems most likely. But I think another round of designs is in order.

Decided A) I needed to give the light armor a proper shot, which meant separating out the boots and B) I might want to retry the dress not because it will be Wren’s main design, but because she might want to make social calls. So I did the harness without the boots and bucklets, gave the armor and dress a second go, and then did a trio of variations on the cloak.

I think I had better sleep on it before I try my hand at round 3.

Captain’s Log LC•R2

Last week, as predicted, I did very little on the game. Not nothing, though. I spent a lot of time doing character and setting designs that will tie into the comic. I found a workflow that is almost as fast for creating “hand-inked” looking vector art as my pixel art workflow is at making pixel art…

..which re-sparked the age-old question of whether I should use HD art or Pixel Art.

There were three elements that tipped the balance for me. First was the seasonal enjoyment of Muppets Christmas Carol. My piqha, and indeed all characters in the retro-cartoony art style I’m developing, are my version of muppets as much as anything. And one thing I like about “my” muppets is that they exist in a digital world, where Mr. Henson’s exist physically. It’s something I’ve tried, and failed, to develop, in the past:

But as much as I haven’t got it figured out, I haven’t let go of it either. Even my “paper dolls” exist conceptually in my head as digital life forms. Pixel art merely makes that explicit.

The second element that tipped the balance was watching a video on Super Mario Brothers speedrunning where they talked about frame rules and manipulable RNG. Mechanics necessitated by the hardware of the time, but mechanics that I fully desire to include in my games on purpose. And the fact that they are pixel art helps thematically hint that these things will exist in my games.

The third element that helped tip the balance was a tutorial on YouTube on how to create a pixelation filter, which I immediately implemented yesterday out of the sheer joy of doing it.

I now have a glitch animation I can call whenever I want from code, as well as a fade out/fade in method that is both more elegant than what I did with Prelude to Nightmare and more Godoty: my Hat Trick fade was done the same way I would do a fade in Unity.

Along the way, I tweaked my inky caricature to be in tune with Popeye, and tweaked my pixel art caricature to be in line with my inky one.

Which is a great improvement in my eyes.

Ink and pixels will both always be elements of how I present my stuff, I think. With 3D making rare but real appearances from time to time..

But I do love the pixels.

One marginal fourth factor convincing me to go with pixels over HD was that I want my games to run on potatoes, and not require super high-end hardware.

One marginal fifth factor is that Sierra called their graphical adventure games “Hi-Res Adventures” because this was hi-res compared to a text adventure:

… and I think it would be hilarious to call my games “Low-res adventures” despite them being higher res than the Sierra high res adventures.

The one thing that was not a factor despite the fact that it ought to have been the single most important factor is that it’d take much of a week to rebuild what I’ve got so far in HD. At some point in a project, you have to commit to not starting over, even though you’ve learned so much and done so much that you know starting over would be faster and better. Because if you let yourself start over once, you’ll let yourself start over again and again and never get done.

My books are not perfect, but they are finished, and the lessons I would learn by starting over get applied to the next book.

But while that should have been the first factor and the deciding factor, I never considered that factor, as the other factors made the decision before I got to that point.

What are we going to do this week?

A game is complete when it has a start menu, sound and graphics options, an input screen (although, ideally input customization options), a credits screen, and gameplay with the game over conditions (win conditions, lose conditions, so on).
I do not release incomplete games.
Note LA•S8: Complete Game

This week, my primary goal is to turn my gameplay demo into a small complete game. Doing the bare minimum work as fast as possible to have it done.

Then, in January, the first two weeks will be dedicated to expanding the game, and the second two to polishing the game, making sure at the end of each week to end with a finished game. In this way, at the beginning of February, even if I have to cut content that I wanted to put in the game, I will be able to release a game.

So that’s the plan for this week. Make a title screen/start menu, the options and credits, and the end conditions.

The FYOOOTCHER..

If God wills, and I haven’t finished development for the Mad Christian Last Legend comic by February, as a side-effect of making this game, February will be devoted to comic development until it is ready to go. Using the game engine and comic assets together to make YouTube animated shorts (and I dunno, TikToks) will be the hoped-for side-effect of that project as well, because the plan is then to spend March and April producing a JRPG, Last Legend I.

If Bunny Trail Junction is the the rocket, then Last Legend I is the launch and Last Legend Zero is the fuel.

Captain’s Log L9·R1

Three days left in September. I’ve got my plan.

I work on the Awesome Moments Kickstarter as my primary project every day until it’s good to go. Target Launch Date for the Kickstarter is October 9th.

Once it’s ready to go, I return to Hat Trick: Prelude to Nightmare. At this time, I might start streaming my development on Twitch or YouTube.

Yesterday I made a brush holder so I could ink vertically.

And inked a Hat Trick with it. End result: good idea, didn’t actually make things easier overall.

ONWARD!