Pixel Art Characters

Back in September 2012 the indie team behind Two Brothers mentioned to me that their characters’ sprites were designed before—well, before what would normally be called concept art, but since it came second in this case, I’ll just call it their traditional art. The biggest reason to go that route is to lend a certain feel to the visual aspect of a game.

I do everything for the sake of gameplay, though, so I started from the standpoint that most characters should have 6-8 outfit slots filled up at first and characters with 9-10 pieces of clothing or 2-5 pieces of clothing would stand out and feel different! Jelia is an example of a character who I had intended to fill every outfit slot except her legs (waist and legs are different slots) because part of her initial design was that she couldn’t change clothes.

(Now that I think of it, I’ll have to go into more detail about this topic later this month in a Dreamblazers Cutting Room Floor post on the Musings side of the site. Just wait until you hear my (now-canned) radical idea to make an all-gameplay-no-story mode and an all-story-no-gameplay mode! (Spoiler alert: it’s super pretentious indie meta nonsense!))

So when you see a character with a visually busy and complex outfit design like Jig (left side)…

…or her brother Tango…

…it’s usually because I had very specific ideas for the gameplay functionality of their clothing. Characters are even balanced partly by their clothes. For example, the Characters page shows over five times as many clothes that boost Magic Power as clothes that boost Power.

An unintended consequence, though, is that I’ve made it more difficult than it should be to convert certain character designs into their 16-bit forms. Jelia turned out well because of Alex’s skill, but nothing guarantees that it’s going to keep up considering some of the designs yet to be pixelized. So that’s something I’m thinking about at this point in time.

Why am I thinking of this? Well, because of my second topic for today…

Pixel Art Monsters

Got a promising application from someone wanting to draw the monsters and other enemies of Dreamblazers! :D Actually, I got this application nearly a month ago and it nearly got lost by not forwarding to my email… so my own blunder nearly blew this opportunity, but now things are looking up. I won’t say anything more until it’s officially official (!), but—well, I’ll just say I’m really hoping on this one. =) And you’ll know the details when I have them!

Pixel Art Cohesion

With a third pixel artist potentially on the way—and knowing that my goal for the monsters is to approximate a Pokémon style—I started thinking about the cohesion of all the pixel art. Right now Becca tends to draw fairly realistic environments (to the extent pixel art can be realistic) while Alex is faithfully recreating Flora’s artwork and color choices in pixel form.

I’m usually constantly pushing toward brighter and brighter colors and so far I’ve been fortunate that nothing seems to blatantly clash, but I began to wonder if that can continue. Will I eventually break everything with the monsters? What about the special effects (like magic)? Will the menus do it?

And then I realized something just this past week. Look at this Final Fantasy VI image:

I’ve played thousands of games and I’d have to think about it, but my gut feeling is that Final Fantasy VI is probably still in my top ten of all-time. It’s a masterpiece and always will be in my book.

That said, the more I look at this picture, the more it seems like a visual mess despite each individual element looking very nice. The background clouds, lake, and trees are hyper-detailed to the extent that it wouldn’t surprise me if I found out they took a photograph by a real lake and did their best to trace it with pixels.

The heroes are a colorful animeriffic bunch with the green-haired Terra, Edgar’s pastel blue and green outfit, and Mog the cute fantasy creature, but the enemies use earth tones and darker shades for a more gritty feel. If that wasn’t clashy enough, the menus are a basic blue gradient.

And the perspective is extremely skewed in a way that I can best illustrate visually:

The background is basically a flat-angle shot from the side of that lake, but the rest of the ground is a tilted aerial perspective looking down and forward. Either of those on its own is fine, but put them together and it’s a disjointed effect of two cameras being mashed together.

But—and here’s the key thing—I never noticed any of this until now. I didn’t notice, I didn’t mind, and although I do notice now I still don’t mind. So I’m going to plow forward without concern! Apparently I’d have to mess up super badly for any lack of cohesion to be as big a deal as I’ve made it out to be in my mind and in my worries.

Game development is a pretty fascinating thing, huh?

21,008: that’s the number of spambot IP addresses now blocked from this site. But enough whining: on to more interesting things!

Much like Flora’s character portraits push me forward to do them justice with the writing and dialogue, Alex’s character pixel art pushes me forward to do justice to the animations. It’s not my natural strength, in all honesty—that would be writing, imagining, and especially crunching numbers—but I certainly love what I see from Alex and I hope you all do too. :D

Speaking of crunching numbers, I’m feeling pretty satisfied now with the fashion system rebalancing I talked about last week. I’m changing my thinking on how to handle earning new outfits, though; because several of the main characters are fairly isolated from society and have little or nothing in the way of money, originally the primary way to get new clothes would be when new characters join the party. One interesting side effect of this design is that players could appreciate new characters for the tangible benefits they bring to the table even if they don’t like their personalities. However, if it’s harder to get fashion bonuses then new clothes can also be handed out a little more frequently than before.

The enemy rebalancing, on the other hand, isn’t quite where I want it to be yet. This took a long while to get right the first time around before I decided to change the fashion system, so I might just revert the enemies to what they were before and live with a slightly easier game.

Alrighty, let’s take some exciting things one by one!

So much good news in one day! :O 55,000 Steam sales being a median for RPGs is extremely reassuring. :D https://t.co/Ry2QpL9nos

— Jelly Paladin (@JellyPaladin) June 22, 2015

So, somehow, in all the rambling and fretting and pondering I’ve been writing about here for the past couple of months, wondering what I can and can’t afford, crunching numbers on what the kinds of games that I back on Kickstarter make on average, and so on, somehow it never occurred to me to think through all the way to what happens sales-wise after a game is finished and is being sold to the public. Part of the reason why is because Dreamblazers is probably still quite a ways off, but another part is because I wouldn’t have known where to look for those sorts of figures.

Enter the article in the link above telling me exactly where to look. The article refers to 55,000 sales as the “average” for an RPG on Steam, which wouldn’t necessarily be good news, but they later clarified it was actually the median:

@GridSageGames @sinisterdesign @raiganburns yes, it's median, sorry for confusion

— Steam Spy (@Steam_Spy) June 22, 2015

I never know who’s reading this blog (other than like 650 spambots per day, but more on that in a moment), so for those not in the know about the difference, here’s an easy illustration: let’s say ten games are for sale and they sell 0, 500, 2500, 5000, 10000, 15000, 17500, 35000, 95500, and 275250. The average here is 45,625, but you can see how misleading it would be to look purely at average when in fact only two of the ten sold that many or more. The median is 12,500, which is more in line with what a person could reasonably expect to sell since five of the ten sold that many and another came very close.

Bottom line: suffice it to say I’m feeling a lot less insane now about my entire indie game developer venture. =P It’s still progressing more slowly than I’d like, but on the other hand it’s 2015 and the entire resources of the world still haven’t figured out flying cars for us, so I can only feel so bad about that.

Now for my second topic… You know who’s arguably progressing more quickly than I’d like? gamingislove, the creator of ORK Framework and Makinom. He’s so on the ball with both of his Unity assets that often I’ve just barely figured out what in the world an update with 10-14 new features does by the time he’s released another one.

Yesterday he released something that I think will let me do very interesting things with turn order in the battle system if I only 1) choose to abandon a traditional turn-based system or ATB system and 2) figure out how exactly the “multi-turns” system works. :P That’s part of why this post was a day late—this really excited me, so I rushed to experiment with it, but I haven’t quite gotten a grasp yet on how it works. Apparently it’s similar to Final Fantasy X, but there’s been a learning curve for me since I haven’t played that. One thing I do know is that now I’m capable of giving a character multiple turns in a row!

…this isn’t quite the same thing as multiple attacks per turn, though (which is what I’d really, really like). For example, Celty’s the fastest playable character in the game, so let’s say she could either attack three times per turn (not possible as far as I know) or could get three turns to every other character’s one (possible). In the first scenario, if she gets hit with the Soaked status effect that covers her in water and makes her weaker against lightning for three turns, then she’ll attack nine times before she dries off, but the enemies will also get two turns. In the second scenario, if she gets Soaked, she’ll take three turns in a row and dry off before the enemies have a chance to zap her.

So I’m not yet sure if I’m sticking with this route, but that’s what I’m looking into.

And then, of course, today gamingislove released the plugin to bridge ORK Framework and Makinom, so I’ve got even more on my plate! It actually feels good to be back in that swing instead of mostly waiting on art assets or sending out feedback about them.

One day, when I crack that median 55,000 sales figure, I’ll have to figure out some way to pay that guy back. Sure, he’s getting hundreds or thousands of customers, but it still boggles my mind that I could either pay an entry-level programmer $65,000 for a year—that’s not so far off from how much money as I’ve made in my life!—or I could have done what I did and make this one-time $100 Unity purchase. For all my struggles and all my amateurity shining through, I wouldn’t trade any of it. In fact, I literally couldn’t ahave.

Last thing for the day! The Steam article from earlier did make mention of PR being important, so at some point I have to step up in that department. For some reason there’s been a sudden massive upswing in spam comments here over the past ten days to the tune of about 1000 per day, which is another reason this post is a day late, but that’s obviously not quite the same as drawing in new eyes. :P

Not too concerned about it just yet, but finding a monster artist should bring me really close. I don’t think I necessarily need stuff like music or even special effects in battle before I can put out the most basic of basic demos, but obviously I’ll have to be able to show off monsters for the main characters to fight.

Because of my divided time, progress has understandably slowed down a bit. Not only that, but I’m now learning about Makinom since that’s going to be useful rather than continuing to build on the established stuff in ORK Framework and 2D Toolkit. So I didn’t really have anything to report this week and I had to ask myself whether it was still worth posting a devlog. These posts themselves take important time too, of course, so ultimately I decided the answer was no—thus no post on Monday.

But I’ve reconsidered. These posts are my development history and, for better or worse, I’m going to keep up weekly posts even if they’re smaller in scope than they used to be.

So rather than talking about what I did, which isn’t much in terms of tangible progress, I’ll talk about where my mindset’s at. =)

You know what’s pretty inconceivable? In the Princess Bride sense, I mean, not the literal sense? The amount of money in my bank account going up rather than down for the first time in sixteen months! This radical notion of no longer running on financial fumes has made me ask myself questions about the scope of what I can pay for.

Right now Becca is already making some overworld tiles that I forgot on our first go-around and also expanding the Seaside Slime Cove tiles with some new ideas that she and I had. If not for the job-out-of-nowhere, the Seaside Slime Cove tiles are something that I probably wouldn’t have commissioned pre-Kickstarter—one way or another, new money needed to come in before I would go for it.

But there are other things I could pay for. Other than pixel art, just a few weeks ago I briefly mentioned the possibility of a full-blown visual fashion system like those online doll programs you see people make in Flash.

Or how about this? Games like Phantasy Star IV and Fire Emblem have used illustrated sequences to portray key scenes:

(For more of those Fire Emblem examples, see here: Fire Emblem cut scenes)

I say “key scenes” because they didn’t illustrate every scene. Every scene would be the domain of games like Phoenix Wright, visual novels, or visual novel hybrids like Sakura Wars: So Long, My Love (one of my favorite games).

I can’t remember if I’ve discussed this on the blog, but I’ve definitely mentioned my approach to writing the story on Twitter:

How I write my RPG story lately: ask which characters haven't had standout scenes yet, then create circumstances that would give them one.

— Jelly Paladin (@JellyPaladin) August 15, 2014

Virtually every character gets to have at least one shining scene (or sometimes two characters at once if they’re fighting against or with each other), so maybe I could spice those up.

These are all brainstorms, of course, but the Kickstarter audience tends to want to know about developers’ ambitions—things that they’d like to do and things that they have to leave on the cutting room floor but could revisit later if circumstances allow.

…oh yeah! Speaking of the cutting room floor, I meant to link to this blog post over on my musings section from a couple weeks back.

While it’s still a big part of the game that you can “keep the characters you like and kick out the ones you don’t”, my original conception of Dreamblazers included the option to go even further than that—I wanted players to be able to switch character parties. Don’t agree with the dreams that Leaf and Celty chase? Then ignore those two and go play as Sakura and Evelyn instead. Sakura and Evelyn are about to battle against Kelly and Jun and you think Kelly and Jun make a convincing case that they’re in the right? Then switch to playing them at literally the last moment.

Of course, if Sakura and Evelyn were defeated then the story would have to end, but I liked the idea—it reminded me of the New Game+ option in Chrono Trigger where, at any given point in the story, the heroes can abandon whatever they’re doing and go beat up the final boss, leaving the timeline in total disarry.

A design philosophy like this is part of why I modeled the stat growth formulas and damage formulas after Pokémon: I wanted a player-vs.-player feel where the “antagonists” were designed in the same mold rather than being on totally different scales. In most RPGs, enemies have ten times or even a hundred times or even a thousand times as much HP as the playable characters, but the playable characters do more damage to compensate. This always make for awkward scenarios when a former antagonist joins up with the good guys and suddenly they’re no longer the imposing, indestructible force that you once battled against. Dreamblazers was meant, at least initially, for the player to be able to switch “allegiances.”

This proved to be way, way, way too ambitious, though, so it’s no longer an option. Rather than multiple concurrent stories, right now my goal is alternating stories in the vein of games like Dragon Quest IV and Mother 3: play as one group of characters for a bit, then a different group, then check back in on the previous group, and so on.

Can I even pull this off? I honestly don’t know:

I finally understand why Golden Sun 1 and 2 were separate games. Concurrently-running stories might be overambitious for any RPG…

— Jelly Paladin (@JellyPaladin) June 17, 2015

If I don't want an endless, impossible-to-complete story, then I'd better live with a lot of events that my protagonists don't observe.

— Jelly Paladin (@JellyPaladin) June 17, 2015

But that’s my current thought process and I’ve found that being able to use this technique is really helpful whenever I hit a wall in the writing process and don’t know where a specific group of characters should go next. Instead of throwing them into filler situations, why not write in a time skip without actually skipping time by leaving some characters behind to check out what other characters are doing elsewhere?

…

Huh! For a post that I was reluctant to write because I thought I wouldn’t have enough to say, this turned out pretty enriching. =) At any rate, talk to you all again at the more usual time of Monday in just a few days!