Dev Diary #12: How we got here, and where we're going

REMINDER: We're launching Steam Greenlight and Kickstarter campaigns for Legacy of the Elder Star on September 17, 2015, and right now we're running a Thunderclap to get the word out. If you'd like to support us with a tweet, Facebook post, and/or Tumblr post, click here. Thanks! <3

We're launching on Kickstarter two weeks from today, so I thought this would be a great opportunity to take a look at how we got here and why we're making what we're making. We're going deep behind the scenes today!

Turns out action games are really my thing

So, I worked in commercial game dev from 2004 to 2013, mostly designing action games like the PS3's Warhawk reboot and its spiritual successor, Starhawk. Fun stuff! But in March 2013 I'd kinda had my fill of studio life for a while, so I decided it was time to go indie.

I started working on this game that was supposed to be like a tactical Diablo-with-spaceships. You'd click around to move and target things, and hit number keys to fire "skills" (read: spaceship modules) off your "skill bar". Get loot, level up, kill tougher things... you know the drill.

This... did not get far. Most of the work went into the hotbar, and a complicated set of resource management rules. By the way, the ship you see there is actually just cut out of a screenshot from EVE online, and yes, it was meant to be placeholder art.

This... did not get far. Most of the work went into the hotbar, and a complicated set of resource management rules. By the way, the ship you see there is actually just cut out of a screenshot from EVE online, and yes, it was meant to be placeholder art.

I messed around with it for a few months but I kept feeling bored, and that was weird and new, because I'd never really felt bored by game dev before. Frustrated, sure, but never bored. It took me longer than I care to admit to realize I was bored because I was making, basically, a strategy game.

Now, I've got nothing against strategy games! I grew up on Warlords and Command & Conquer and Total Annihilation and Civilization and yes I know some of those are turn-based and some of those are real-time but that's not really the point right now. What I had to figure out – what I did figure out – was that while I enjoy playing strategy games, I actually find making them to be pretty mind-numbing. The part about game dev I love is the part where I'm iterating on how something feels: the recoil of a gun, the weight of a jump, the impact of a collision. Strategy games can have that, sure, but they don't focus on it.

But action games focus on it. So around October 2013 I scrapped Diablo-with-spaceships (please, though, somebody make this so I can play it!) and started thinking about action games.

So, shmups on iOS are kind of awesome

I don't even remember how this happened, but around that time I discovered a bunch of CAVE shmups on the iOS App Store. I didn't actually know about CAVE or any of their games prior to that, and if you're groaning right now because I wasn't introduced via the definitive arcade versions, then you're my kind of people.

But... it turns out the iOS versions are kind of awesome anyway?

DoDonPachi Resurrection HD

DoDonPachi Resurrection HD

ESPGALUDA II HD

ESPGALUDA II HD

They work like this: you touch anywhere on the screen, and when you move your finger, your ship moves the same direction and distance. You don't have to be touching your ship itself, which means you can interact with the game and still see what's going on: you just touch a little bit below, or off to the side, or wherever's convenient based on what's happening right now. The 1:1 control feels brilliant, especially on an iPad where your finger takes up less screen real estate and you can read all the enemy patterns so much more clearly.

I'd done some experiments with mobile game dev before and gotten really familiar with the similarities between touch input and mouse input. In fact, it's super easy to emulate touches in a PC application (e.g. while you're working on it, and don't want to deploy to device every five minutes to test things) by treating the mouse cursor as finger position, clicks as taps, and click-and-drags as swipes. So I'm sitting there playing DoDonPachi Resurrection with this 1:1 touch control and suddenly my brain connects the dots.

This could be mouse-controlled.

A prototype so hot it'll fry your eyeballs

This is what Legacy of the Elder Star looked like in its infancy, before we had a name or an art style or even much of a game design beyond you play it with the mouse:

One of our earliest playable prototypes.

One of our earliest playable prototypes.

That art, though. O_O

That art, though. O_O

I demoed this for some local indies in November (or maybe December?) of 2013. It was the first time I'd shown it to anyone. It didn't exactly get rave reviews (I mean, look at that art) but the mouse control did get a lot of positive feedback. It was the right idea... now it just needed a game behind it.

Moving, toward modern times

The whole "click and drag through enemies" concept came up pretty early on, but it wasn't there from the start. I had to play with the prototype for probably six weeks or so before I started to internalize that motion was going to be the differentiating factor for this game. As far as I'm aware, there's no other game out there that moves like this:

Even the iOS shmups – and by the way, it's not just CAVE, there's also excellent indie titles like Danmaku Unlimited 2 or Shogun which use the same 1:1 touch interface – but they don't take full advantage of their fluid motion: they're still fundamentally games about shooting and dodging, just like all the stick-based arcade shmups they emulate.

And that's kind of the thing about shmups, as a genre: I love 'em, but they're sure stuck in the past. Time was they ruled the arcades, and their design principles – unforgiving difficulty, a few short stages, and a focus on high scores – were oriented toward collecting quarters and churning customers. Modern video games aren't really about that (and let's just leave the F2P discussion aside for now). Instead, modern games offer tens of hours of gameplay over marathon sessions, extensive narratives and deep progression systems.

Not to diminish the value of the classics at all, but wouldn't it be interesting to see shmups... grow up?

What does a modern shmup look like?

In my opinion...

It's purpose-built for the platform it exists on. For arcades, that meant joysticks. For PCs, that means a mouse and keyboard. We chose to focus exclusively on the mouse because, as I said above, the motion we get from 1:1 mouse control is the differentiating factor for this game. Also, those prototype shots I posted above are in a vertical 3:4 aspect ratio, but modern PCs run widescreen monitors. So, we switched from vertical to horizontal scrolling to better utilize the screen space (i.e. to better fit its platform):

This is from our first horizontal prototype in early June 2014. We had swapped the player's triangle for a little bubble with a velocity-driven squash/stretch effect. You can also see a super early version of the star dash in action here.

This is from our first horizontal prototype in early June 2014. We had swapped the player's triangle for a little bubble with a velocity-driven squash/stretch effect. You can also see a super early version of the star dash in action here.

It fits into our modern lives. We crave longer experiences today than we got in the arcade, but with more and more gamers aging into their 30s, 40s, and beyond, we all have lives that tend to interrupt our gaming time. So short stages still make sense – they provide frequent "breaking points" where you can take care of life-stuff – but we're having lots of more of them in the game, and structuring things so you can take your pick of what (and how much) to play, when.

It practices modern design principles. And the elephant in the room of modern design principles is progression systems. We sure do love our XP and level-ups and loot grinds, don't we? Or at least, systems in that general vein. We're trying to avoid outright grinds in Legacy of the Elder Star, but earning and upgrading equipment, filling out special research collections, and clearing randomized, rotating bounties for score boosts are all ways we're injecting modern design sensibilities into an age-old genre.

It appeals to a diverse audience. Back when shmups ruled the arcades, video gaming was a very specific subculture, a kind of club that you either belonged to or you didn't. Today, all kinds of people play games. Importantly, that means a lot of people might stumble across our game who've never set foot in an arcade in their lives. They've never heard of CAVE or Treasure or Raizing or Irem. If you sat them in front of Crimzon Clover: World Ignition, or Hellsinker, or anything Touhou, their heads would explode.

Crimzon Clover: World Ignition

Crimzon Clover: World Ignition

ll those games are amazing, but they all have a few things in common: they're really retro and they're really hardcore. But I think the basic concept of a shmup – you move, you shoot – is something that can be accessible to most gamers, if we just package it in a way that's inviting, not intimidating, to genre newcomers.

Finding the "look"

So I thought about all that stuff a lot, and an actual game started to take shape in my mind. Around June of 2014 I brought Erik Exeter onboard to start making real art, and of course our first order of business was establishing a style.

This, I have to say, was pretty much all him. I gave three points of high-level direction:

  • Use colors!
  • No anime (I've got nothing against anime, it's just that tons of shmups use anime, and I wanted to do something different)
  • You play as a character, not a vehicle

That... was pretty much it. He sketched a bunch of stuff, I pointed at things I liked, he developed those further, and I started figuring out how to build a game around them. It's been a super collaborative process.

An early environment concept piece. Our in-game art actually uses a lot of content from our concepts directly or almost-directly, so there's not as much visual distance between concept and execution as you usually see in games.

An early environment concept piece. Our in-game art actually uses a lot of content from our concepts directly or almost-directly, so there's not as much visual distance between concept and execution as you usually see in games.

Maybe I just got really lucky with an amazing artist who doesn't suck to work with, but I feel like this is the right way to do creative work. I don't make Erik do a billion revisions until his thing looks like whatever I had in my head. In fact, I rarely even have something in my head. He comes up with cool stuff and I look at it and think "what could we do with this" or "how could we use that" and if I can see an awesome opportunity for it then boom that concept/asset/whatever is going into the game.

For what it's worth: I have no doubt that this approach would totally fall apart if we scaled up the team. Which for me feels like an argument for keeping the game small enough that we don't need to scale up the team. ;)

First contact with the public

In September 2014 we took an early concept demo to Salt Lake Comic Con as part of the Utah Games Guild indie showcase there. By this point we'd gotten a good handle on the character design and overall art style, as well as the core gameplay (shooting and dashing). Here's what that demo looked like:

This was the first time we'd shown the game to the general public, and I was pretty nervous whether they'd "get" it. Shmups are a niche, hardcore genre after all, and I had no idea if our colorful art style and mouse-driven movement would be able to overcome that.

Fortunately, people loved it! We got a ton of positive feedback and it was a huge confidence booster. Clearly, we were on the right track. :D

(Fun side note: I'm a co-founder of the Utah Games Guild, so for that event I was organizing not only showing our game, but also the whole showcase for a dozen developers and their games, too. I don't necessarily recommend this, unless you really enjoy stress.)

Game structure is a whole 'nother beast

The Salt Lake Comic Con demo was a single level: no shell, no customization, no progression. Just one level full of stuff to blow up. Our next challenge was to design the rest of the game around that experience, i.e. the game's structure.

Initially I thought we'd do it like Spelunky: four zones with three stages each, arranged in linear order, with randomized enemy patterns (which would be our equivalent of Spelunky's randomized level layouts) and permadeath. Over the next few months we set up a couple other test stages and implemented those systems, and in February of 2015 I brought the new build to Crimson Carnival, a big student party at the University of Utah.

That... was a disaster. See, in the Comic Con demo we had people respawn on-the-spot if they died; my whole reasoning for that was "it's just a demo, let them keep playing". At Crimson Carnival, death meant a "game over" screen, and when that happened, every single player stood up and left immediately. Gone were the compliments and enthusiasm from Comic Con. In their place: indifference? Disappointment?

I went to GDC a few weeks later and met up with fellow indie Jim Shepard (he of Dungeonmans fame) and told him our Crimson Carnival tale. He looked me square in the eye and said something like, "Dude, you're trying to make a more accessible shmup, and you put in permadeath?"

Well, when you put it that way... ;)

Making things comfortable

Around that time I saw some panel on YouTube with Rami Ismail, and I think Adam Saltsman, and one or two other guys... I can't remember what event it was from, which renders my Google-fu weak and sad and prevents me from linking you to it (maybe educate me in the comments?)

But anyway, Rami was talking about Nuclear Throne, and he said something like "we really want to make this a comfortable game" and that phrase has been stuck in my brain ever since.

The thing about that "game over" screen at Crimson Carnival is that it was uncomfortable. Oh look, I failed. The game is kicking me out. In front of my friends. Who are making fun of me now. :(

And the thing about all those super-hardcore bullet hell shmups I've mentioned a few times up above? They're uncomfortable for a lot of people, too. They're way too hard and there's way too much stuff going on on the screen. People feel overwhelmed and incompetent. And on top of that, there's a sizable proportion of shmups out there with character designs that are just... creepy. (If you know what moe means, you know what I'm talking about. If you don't... maaaaybe don't Google it.)

We started thinking about what worked for us at Comic Con, and it was all stuff that made players feel comfortable. They respawned after they died, instead of the game coming to a halt. The game has bright colors, upbeat music, and appealing, semi-cartoony characters. It's not gritty, gory, or voyeuristic. Kids liked it, and their parents liked that they liked it.

Which is all a very long-winded way of saying: the permadeath had to go.

Wait, so you're letting in all the filthy casuals?

So this... comes up, sometimes, when I talk about this stuff. When I tell certain people that players didn't like hitting the "game over" screen and that we responded by making it so there is no "game over" screen, I get back this sort of accusation that we're mollycoddling players who should just suck it up and "git gud". And I'd wave that away if we were making a casual match-three game, but we're making a shmup, and shmups are known for being hardcore and competitive.

Have you ever watched a superplay? Look up some on YouTube some time. They're amazing. Here's one I like:

This is a huge part of the hardcore shmup community. (And actually, I should correct myself now, because most of them don't use the term shmup: they use "STG", which is a sorta-kinda acronym for "shooting game", which is the Anglicized Japanese term for the genre. The more you know!)

A big part of what the community does today is study these games in great detail, honing their strategies for dodging patterns, managing resources, timing their kills, and much more, all in pursuit of a superplay. There's deep analysis here, on par with that you find in any competitive sport. Check this out:

Competitors at this level are forged in the hellfires of extreme challenge... or at least, so goes the conventional wisdom. So when we start talking about making our game "comfortable", what does that mean for competitive play?

The way I see it, we're basically taking the same design approach as Blizzard when it comes to this issue. Take a look at Starcraft. On the surface, it's a pretty casual, streamlined RTS that's comfortable for millions of people to play. The vast majority of those players play casually, and that's all they want from the game. It's fine!

And yet, at the same time, in the very same comfortable game, you've got this incredibly vibrant competitive scene. (South Korea, in particular, says hello.) When we talk about e-sports today we're usually talking about League of Legends or DOTA 2, but Starcraft kinda kicked that whole thing off. Casual, comfortable Starcraft.

I guess what it boils down to is, the entry point for our game should be pretty casual, so a whole diverse audience of people can get into it and enjoy it. But that's the entry point. That doesn't mean we can't, say, use our bounty system to throw some very tough challenges at you; it's just that those bounties will be optional. They boost your score, so if you're playing competitively you're gonna want to opt-in. But if you're playing casually, and you don't care about your score so much as just blowing things up? Skip the bounties, keep things easy and comfortable, and just have fun.

And here's the real dream, for me: that we bring some players into Legacy of the Elder Star who are honestly intimidated by those other shmups, but because we're giving them a nice, comfortable experience, they come in and try it out. And they like it, and they stay awhile, and they start to learn the genre. Maybe they watch a few superplays and, for the first time, actually understand a little bit of what's going on. Their confidence grows, and they start to explore other shmups. They play their way backwards from DoDonPachi to Battle Garegga, or from R-Type to Defender, and notice the lineage. They start asking questions about the history. They learn what MAME is, and what a PCB is, and that there's this whole rich collector culture that's grown up around this genre.

Those players would never have gotten there if not for a comfortable game that was willing to just let them have fun. I don't really see that game out there. I'd like for ours to be that game.

So, where are we today?

We're here:

And hopefully six weeks from now we'll have the funding we need to properly finish and ship this thing. :)

We're launching Steam Greenlight and Kickstarter campaigns for Legacy of the Elder Star on September 17, 2015, and right now we're running a Thunderclap to get the word out. If you'd like to support us with a tweet, Facebook post, and/or Tumblr post, click here. Thanks! <3