[Game Developer's editor-emeritus Brandon Sheffield explores the difficulties that arise when attempting to bring a retro-styled video game to mobile devices, marrying older aesthetics with newer input methods and design.] One of my current game projects is a 16-bit arcade-style racer for mobile devices, which puts me in the position of trying to marry an older aesthetic with newer input methods and design. Along the way, I've been playing (and watching videos of) lots of games from the era, to try to catch hold of the vibe, while also seeing what works in modern times. Through the process, I've found there's a lot that can be learned from what's good and what's bad about older games.

Holistic design

With older games, graphical techniques were a lot less sophisticated (in certain ways -- I will still never understand how those assembly wizards managed to create The Adventures of Batman and Robin on Genesis hardware). Because of this simplicity and unity of hardware, you could create a game that felt completely self-contained from start to finish, from its menus to its music. Everything was being drawn in the same way by the same part of the hardware, there was usually just one sound chip, and the hardware's limitations often guided what could and couldn't be done. In many modern titles, the variety of shaders, textures, and lighting arrangements that are possible make it tougher to keep menus, characters, text, and environments feeling like they're 100 percent part of the same world. For a retro game on modern hardware, this gets more interesting. For my racing game, it would be far easier to create a 3D road -- but how would that fit with our pixelated cars and backgrounds? If you look at Final Freeway 2R on iOS and Android, the 3D road feels very smooth compared to a proper 16-bit approach, and the ability to move the camera as you turn makes the pixels of the cars behave in odd ways as it rotates in hardware instead of being redrawn. That's clearly what they're going for -- they've used a variety of techniques to try to emulate that old look, but for my purposes it can't match the dirty, sketchy roads in Road Rash II on Genesis hardware, which make you feel like the road is the game. Our current road approach is to slice the screen into small horizontal chunks, and as the camera moves forward, it looks for new chunks to render. This includes small bits of road, but also spaces out our sideline environmental dressings, which live as billboards along those horizontal lines. This emulates the approach of the older games, but is done for every scanline, since modern hardware is much more powerful. While this may not wind up being our final approach, it's interesting to see how neither fully old nor fully new approaches work perfectly when recreating the vintage look on new devices. The main lesson here is that with a smaller game, we have an excellent opportunity to create a product that looks very integrated and internally consistent, but it can be challenging because newer techniques make it easier to create something that looks almost as good but sacrifices that holistic feel. It's actually tougher to do it the old way!

Floaty control

You'd be surprised how many of those old 16-bit games that we remember as being so sharp and precise actually had rather fiddly controls. In good games like Sub-Terrania on the Sega Genesis this was purposeful, and part of the challenge. Your ship was hard to control and had a lot of inertia, but was incredibly quick to turn, meaning if you got used to the unique playstyle, you could master it. In less-successful games, having too much inertia in a platformer or imprecise hit boxes in a beat-em-up ruins an otherwise enjoyable experience. We see this comparison today, where games like Super Crate Box on iOS have simple controls (left, right, jump, and shoot) that at first feel imprecise, but have a quick learning curve, and ultimately inform better play. This works because the environments are static, single-screen, and easy to navigate. You don't need to wall-hop or do anything complex -- just jump, land on a platform that you know is always there, jump over an enemy, and shoot. On the other hand, there are games that try to be full-on platform game experiences, with triangle jumps and floor stomps, and I've yet to see a one that didn't frustrate me into wishing I had an actual controller. The lesson here is to make sure your control method informs your design. The simple environments in Super Crate Box are a good example of this. With scrolling and complex jumps, the game just wouldn't work as well as it does. Another great example is Super Hexagon, which is a twitchy, hardcore game that only lets you rotate left or right. It's precise and accurate, meaning whenever you die, you know it's your fault.

That old-school feeling

More important than anything else is the old-school vibe. It's hard to pinpoint what exactly triggers that particular feeling in a player, but there are some common threads in the more successful older games. Pattern learning through failure: Most good vintage games show you how to succeed by showing you how you can fail. Missing jumps, knowing that spikes show up over red pits but not blue ones, realizing you can duck to escape enemy fire -- these were the teaching methods of old. (And when you die, it's back to the start of the level with you.) It might feel slightly unfair the first time, but usually if you're quick enough, you can learn the pattern as it's happening without too much actual punishment. This kind of approach to learning and difficulty has fallen out of favor in modern games, which have frequent checkpoints and variable skill levels for casual players. But I think the hardcore nature of these older games contributes a substantially retro vibe, and as games like Super Crate Box, Super Hexagon, and Super Meat Boy prove, good control plus punishing difficulty and pattern learning can really engage players, even today. (Also, adding "Super" to your game title boosts sales.) Repetition bonds you with a product -- that's why kids still learn SAT words with flash cards. We remember the games from our childhood so well in part because we had to play through them so many times in order to beat them. Distinctive music: Many games these days (especially in the triple-A space) give music the backseat, and choose to let it linger in the background without major melodic hooks. Super Mario Bros.'s music was fantastic because it added ragtime to an action game. Streets of Rage 3, composed by Motohiro Kawashima, is an early study in progressive, experimental electronica. The unexpected sounds give these games an edge. Appropriate control: As previously hinted, precision isn't necessarily the hallmark of good control, but the control method has to be appropriate for the game. There were six buttons (eight if you include select and pause) on the SNES controller, but you didn't have to use them all. Also, your game's genre may inform your players' preferences; I'm a particular fan of sticky, inertia-free jumps in platforming and action games, but with flight or driving games, I like a bit of wiggle room. Easter eggs: One thing that is more difficult to represent in larger games is the little subtle nuances that made 16-bit games so charming. Hidden paths and secrets in these living worlds led to a great feeling of discovery. In Bonk's Revenge for the TurboGrafx-16, enemies on a ship cook food and wander about if you leave them alone. Sure, you'll find a developer's face in a portrait in Uncharted 3, or an interesting poster in Call of Duty, but a better example would be Halo's hidden, hard-to-reach cavemen. In the arcade racer Outrun 2, when you stay at the starting line instead of driving off with your peers, the flag waver will do stretches, breakdance, and a host of other actions while you wait. There's no reason to do this -- 95 percent of players will never see it -- but that kind of dedication to the world really helps make it feel alive, and gives the player a great feeling of discovery.

Retro future

For the last decade or so, retro-inspired developers have pondered what makes these games feel the way they do. Ultimately, the vibe may come down to individual nostalgia -- what resonates with that game's director? To rekindle that feeling myself, I've been playing Genesis and TurboGrafx games daily, and encouraging my team to do that as frequently as they can as well. It's been immensely informative, and I can't recommend this enough for anyone trying to make a retro title. I don't know if we'll be successful in taking that vintage vibe and bringing it into the future, but the learning process is already showing me how much vintage titles have to teach about present and future games. This article was originally published in the November issue of Game Developer.