Wednesday, April 22, 2009

Real Time Simplicity: A Talk with Rudolf Kremers and Alex May

Dysons swarming an enemy tree. They say that limitation breeds creativity.  In the world of game design, this axiom often goes ignored.  We regularly see games with gargantuan budgets and five-year production calendars fail because they feel stale just a few hours in.  With every rehashed engine and recycled plot device, design stagnation threatens to infect the entire industry.

Enter Dyson.  Rudolf Kremers and Alex May entered TIGSource’s Procedural Generation competition in May of 2008 and were given a mere thirty days to complete their entry.   The result was a masterfully distilled RTS experience, with a slick visual design and engrossing sound.  They won second place, continued updating the game, and eventually submitted it to the IGF, where it received a nomination for the Seumas McNally Grand Prize.  Word on the street is they’re in digital distribution talks with the likes of Steam and Direct2Drive.  That’s called climbing up the ladder, folks.

We’ve written before about why Dyson’s simple, elegant design makes it a worthy entry into RTS canon.  But as for how it evolved from a humble, forum-competition prototype to an indie success story, we still had some questions.  Luckily, Alex and Rudolf happen to be nice guys and provided us with some answers.

Charge Shot!!!
: Tell us the story of
Dyson's development. How long did it take to produce the game? What was the most difficult or time-consuming stage?

Rudolf Kremers: The original version took us exactly one month to produce as that was the scope of the competition it was created for. We are still developing it as we speak, to be finished sometime after the summer. I think that will take the total development time up to about a year.

Alex May: Yep, over a year (original competition was May 2008 I think) of mostly spare-time work.

CS: Many notable indie games launch with a narrative about the sole author/developer and his five-year development journey. You guys worked as a team and managed to make Dyson in a relatively short amount of time.  How did you two team up, and how did you divvy up responsibilities to maximize effectiveness in a short development cycle?

RK: After Alex sprung me from that jail in Columbia and our nano-implants were returned to us it was easy really.   Ahem, additionally, we made sure that we let the design AND the tech do a lot of the work for us, which was in perfect keeping with the goals of the competition anyway. Small seeds contain large amounts of complexity.

AM: We cut features until it was doable. We were merciless. It would have been impossible without the implants. The initial month deadline really helped in that regard, actually. In fact lots of indie games had original prototypes that were made in much less time than that (see Kyle Gabler's recent global game jam keynote on prototyping in 48 hours:

CS: Early in your development blog, you made reference to some of Dyson's influences - the Little Prince, Dyson trees, etc.  Are there any influences that are more subtle?  Any games you looked to for inspiration?

RK: For me personally it is a general feeling of trying to capture the beauty and sense of wonder that is hidden in hard-core concepts from science fiction, space exploration and science, in an original game friendly setting. It is all about old but vivid obsessions from youth with this kind of stuff.

AM: I had a lot of graphical inspiration from other games over the years. I was making procedural tree simulations a decade ago. A cloud in the sky would turn into a storm just like in Fire & Ice by Andrew Braybrook, and lightning would shoot down onto the trees and they'd catch fire just like in Populous 2. A fire engine would drive onto the screen to put them out, then more would grow. The simple visuals were inspired by vector games like Gravitron and Geometry Wars, but I didn't want the neon-on-black look. Blueberry Garden and Circle were two projects that helped me break out of that visual trap.

CS: What, if anything, are you trying to communicate with Dyson's stark visuals?

AM: To an extent the graphics are a function of the game's simplicity. I like the idea that everything the player can see is something that matters in the simulation or the game. Most games these days have a graphical fidelity that is many orders of magnitude greater than the fidelity of the interactive system. This is silly. It's stupid. Some of the best games are deeply interactive, to the extent that if you can see it you can interact with it in some way. It breaks down when you have things added to the game that look great but don't affect the player in any way, or worse trick them into thinking there's more to the game than there is. Our game is still very simple in terms of interactive entities, and the graphics reflect that.

CS: Dyson's clean graphics and slick aesthetic both contribute to its low-key vibe, but what really sells this aspect of the game is its subtle soundtrack. Tell us about Brian Grainger, aka Milieu, composer of the game's music - did you ask him "give us something chill," or did he just look at the game and make music to match?

RK: Brian is this brilliant independent composer who also runs a record label. He is very good and I am fan of ambient music, which he has made a fair bit of. I had approached him about his music as a fan, but when Dyson was developing into something really nice I suggested to Alex that we ask him to use some of his music. Not only was he happy to oblige, he then started to create new music and audio for the game, which is more than a little bit cool. I am glad you commented on this by the way, I think it is absolutely essential to the atmosphere and mood we want from the game.

AM: Not many people seem to realize the importance of the music to Dyson as a game. We've got a lot of work to do on the sound effects but it wouldn't be anywhere near as good without the music Brian's made for it.

CS: Dyson has had quite a long development process (with good reason, obviously).  It started as an entry in a time-crunch competition, but has continued to evolve in the months since.  What about the game continues to engage you as developers?

RK: In some ways the game's potential scope, as opposed to its viable scope, is enormous. There is a certain amount of tension between what you could do with a game like this and between the minimalist design/development ethos we adopted for Dyson. We are in a situation where we can only achieve so much within limited time, and yet we have a massive canvas to fill. Dealing with these two extremes is very enjoyable in my opinion.

CS: From a developer's standpoint, what do you feel Dyson lacks?  Given more time or resources, is there anything you would change?

RK: The game is unfinished and we have detailed plans to finalize the game, which we are implementing now. As it stands it is unbalanced and misses some key features.

AM: We've got the time and resources to make a great game by Autumn, so that's what we're going to try to do. We'll be trying to bring out the real flavors of the game, so that people who want to really appreciate it on a deeper level can do so. Right now it is quite a superficial game.

CS: To follow that up, you've made continual improvements to Dyson - the explosive pods, an essential defensive mechanism and a big part of the game, strategically, were only added relatively recently. Looking beyond your most recent build, do you see yourselves making any big gameplay-changing alterations? We would personally like to see a "random map" mode where you set the parameters for the level you'd like to play...

RK: It is a popular request and extremely likely to end up in the final game. NOTE: This is not a promise.  The gameplay will still be enhanced, at times in a major way, but they are all logical additions.

AM: The more we mention now, the more we hype up features that may not make it into the final game. I think some kind of random level generator is bound to end up in the game, in some form or other. I mean, right now, each of the 6 scenarios in the game is randomly generated each time you play, so to an extent that's already in there.

CS: Define "logical additions" - I know you're trying to avoid promising any features that eventually get cut from the game, but how big are we talking here? Different types of trees? More attributes for the Dyson, or the planets from which they spring?

RK: Logical additions are gameplay mechanics that are appropriate or respectful of the gameplay that fits Dyson and the world in which it takes place. You will have to torture me to get me to say more!

CS: You've said that you have no plans for a multiplayer component for Dyson - we'd desperately like one, but we understand the difficulties you face from a programming perspective. Do you think that you might follow Dyson up with something more multiplayer-centric, or is this something that's on your radar?

RK: I have a multiplayer focused game in the works, which takes some elements from Dyson and puts it into a more suitable multiplayer format.

AM: I've got plans for a multiplayer game, but it won't play anything like Dyson, or much like anything else you've ever played.   Someone made their own multiplayer version of Dyson, if you want to check it out you can find it here:

CS: Looking at Quantum, I see much more of a Geometry Wars influence there - lots of glowing neon on a dark background. I assume this is what you were trying to avoid when you chose the game's current color scheme?

AM: Yes, that's exactly what happened. I wanted to make a glowy vector game, because that would make the graphics easy to generate in a procedural manner. But I was sick of seeing neon on black as a color scheme and there were several games around in development at the time that gave me the idea to basically invert the colors.

CS: Dyson was made as an entry into the TIGSource Procedural Generation competition.  If you ran your own such design competition, what kind of parameters would you set to inspire creativity?

AM: There's a great thread about this in the TIGForum. I don't like aesthetic themes as much, like "game must include pirates" - if you try you can still be very creative, but most people would just focus on piratey imagery and not incorporate that into the game design. The best ideas for me were things like Non-violence and Earth Day, and concepts like evolution or immortality. But really, any framework at all seems to get most people thinking about interesting ideas, so it's all good.

CS: Speaking of the TIGSource Competition, games using procedural generation seem to be cropping up more and more, especially in the indie world.  Could you talk a little bit about the concept and its general impact on game design?

RK: It is a massive boon in terms of ease of production. There is no difficult asset pipeline, nobody has to wait for things to get modeled and animated in Max or Maya and most importantly, it facilitates rapid iteration of ideas. In the case of Dyson it completely fits the foundation of the game design, which lies in subjects that are closely related to procedural generation.

CS: We've noticed that you have deals with Steam and other services coming up. Give it to us straight - how long before Dyson starts costing us money? What do you think a fair price is, and what do you think of the bang-for-buck ratio of indie games and DLC in general?

RK: The game has a release date at the end of July, so that is as straight up as it is going to be.  We are currently considering a $20 price point but have not yet made a final decision. A fair price is one where we provide a game that feels like it is worth the money you have spent on it. This can be completely different from game to game, or even from person to person. When it comes to most indie games I think that people generally get a fair or even very good deal. Compare it to the price of new so-called "AAA" games and decide for yourself I say.

CS: As the industry takes a broadside fusillade from the economic crisis, mega-titles like Halo are beginning to seem overblown and wasteful, while titles like Braid and World of Goo gain fresh recognition and respect. Do you think the recession could be an opportunity for indie games?

RK: I have to say I never think about this while working on our own games. I suppose providing a good game for a lower price point than hugely expensive big budget games is timely considering the decreasing budgets that people have, but I think there is place for all kinds of games. Massive budget, studio making/breaking, monster, blockbusters and games like the one we are making. On a related note however, having worked on big budget titles myself, I do think that many of these "AAA" games can be made much better and faster than they are now, the waste that goes on due to bad project management and tools and technology can only be described as grotesque.

CS: You mentioned a few things you thought developers often get wrong - i.e., investing more effort in the graphics than the gameplay. What do you think a game like Dyson says to an industry obsessed with Hollywood-style development and a production budget to match?

RK: I reckon not too much, we are so small in comparison that we fly under the radar.  But in more serious terms... I am not sure, I think it may make for an interesting interview if you are able to get somebody from that world to comment on the indie scene.

CS: Games have evolved to the point where many require eight buttons, two sticks, and some sort of goofy tilt control.  RTS games, in particular, regularly come with a litany of complex keyboard commands, which must be studied intensely for high-level play.  Dyson's two-button, scroll-wheel mouse controls deviate noticeably from this trend.  How do you think this relates to your simple, yet elegant aesthetic?

RK: We are going to make it even simpler.  One of our challenges is to provide deep and immersive strategic gameplay to people who have never gotten into 4x [see here] or RTS games, and to do so you simply cannot just adopt the old conventions of the genre and slap some great visuals on it. The design also has to fulfill its part, after the visuals have already done their job. The idea is to let the player slowly learn the game without noticing the interface too much. As opposed to the interface being a barrier that the player has to break through. Imagine only being able to enjoy music by learning an instrument first? I think for many inexperienced gamers that is what it feels like when they pick up a joypad or keyboard and mouse and are confronted with 19 pages of explanation on how to enjoy the game. I'd like to improve on that.

CS: For any RTS gamer, tempo is key - you want to be aggressive while not stretching yourself too thin. I've always been a blitzkrieg expansionist, and that's gotten me in trouble more than once in Dyson. How does the game make real-time strategists reevaluate their tried-and-true tactics?

RK: Mainly by accepting the game for what it is. It is by no means perfect, but it tries to be consistent within its own rules set, and we don't really implement anything we think is extraneous to our goals. This means that there will be some overlap between RTS and 4x ideas simply because sometimes the same problem requires the same solution, but often things are done in their own way.

CS: Bringing it back to your comments on graphical fidelity versus interactive systems in games - two of us have been playing through Resident Evil 5, and while we're impressed with the graphics, we're not so impressed by the fact that we can only interact with a fraction of what we see, and that we keep killing the same five basic zombies over and over again. What, do you think, are easy ways to help the player feel more engaged in their environment?

RK: That is a simple question that demands a 30 page answer. In fact I have done exactly that written several chapters on this subject in my book on level design that I have now nearly finished and should come out later this year. (PLEASE BUY IT) Ahem, that wasn't very subtle was it?

Well, to give a hint, there are several ways of keeping players immersed in their game environment, but many of them revolve around making the environment itself an important reward mechanism as well as a place you enjoy spending time in. In the case of Dyson that is easy to see: almost everything in the environment has a potentially interesting gameplay function as well as being interesting or beautiful in its own right, independent of gameplay actions.

CS: Often what gets cut and why can be just as interesting as the stuff that actually makes it into the game. To wrap this up, any other big ideas that have been tested and cut before version 1.20?

RK: I'd rather you ask us that again after we release the game.  :-)

We’d like to thank Alex and Rudolf for chatting with us.  Again, you can find the current build of
Dyson along with further info on the game at  Rudolf’s book Level Design: Concept, Theory, & Practice (Publ: AK. Peters) should be out later this year, and you can find news on his other projects, including Neopolis, over at Omni-Labs.  Check out Alex’s other work, specifically the plans for his zombie-survival game Deadrock, at his development blog.