What's your process for clearing away the fog of an abstract idea and drilling down to a realistic implementation? How long does this take you? What stands in your way? Do you ever find that you can get down to specifics too early, discarding a good idea before it's had time to properly seat itself in your mind? I've noticed a bunch of manifestos popping up of late about what games should be. One key element they seem to share is that they're all abstract and fuzzy, maybe sounding good in principle but lacking enough specifics to really test their worthiness. So I thought I'd take DogCity up on a suggestion in another thread to maybe start a discussion on how we get from abstract to the detailed, and especially what traps lie in the way. Here's a bit of my process: Personally, game design is for me a strange mix of warm, feel good creativity and hard, uncompromising analysis. When I first get a concept I'm often excited by the newness of it (that is, new to me, anyway). I do a lot of pacing and walking when I have an idea, so you'll see me going back and forth visualizing what it would be like to play it, what the experience would be like. My chief pitfall is that I can get so psyched by the idea that I don't think anything at all about the interface, the implementation, or asset requirements. Fluid movie-like scenes roll through my head, I see dramatic events, and I often even hear music that would go with the scene. I have a background sense of the player being surprised enough to make emotional outbursts, being emotionally engaged. But there's really no hard form. It really ticks me off because while it might serve to raise my morale a bit, it gets in the way of what the actual player experience would be. So I catch myself, often now only after a few minutes (I'm slowly getting better at this), and drain color from the imagined world. I visualize flattening things out, framing things in terms of a screen rather than the livid sensation of imagination, and start picturing interface controls. I introduce errors or defects that sort of "chill the romance," so to speak. I then see keyboard and mouse, and hands making movement. I'll try to visualize wireframe meshes and textures waiting to be applied. Making the visualization grittier serves to ground me. I notice that we humans seem to often fantasize in perfection-- our dream job will never have any downs, only ups; our perfect mate will have no blemishes, etc-- I think it's the nature of human imagination. So developing a discipline of intruding the real into the dream I think extinguishes what can be a compelling temptation to forever dream about it and never make your design ideas real. So I then spend a much longer time methodically (though still abstractly) playing through the feature I've imagined. Often I'll sketch images or write down notes, but I'm careful not to kill the idea early. It may still be good even if I don't yet see a way of implementing it, or if (more likely it seems these days) I know there's no way in hell I can afford to build what I want. Once the idea has cooled a bit, I start trying to mine it's essential aesthetic. THIS is the core of the idea, one that can be captured in something as cheap as a card game. So I try to discover why it's good and what keeps it hanging around. I ask whether cutting this or that would still carry the same flavor. This process unfortunately isn't as smooth as I'd like, especially because I'm forever hunting for a way to replace scripted gameplay with self-contained, replayable processes and I often complicate everything I touch. Emotionally it sometimes feels like jumping down a rabbit hole, burying myself in my own complexity and falling into the trap of thinking the idea is good because it is complex. I suppose that's why I post here so frequently, because I think without people to bounce the ideas off of I'd get lost in my own head.[rolleyes] When I've finally got something that's cooled from the white hot formlessness of imagination into something concrete and testable, I have now taken to trying to build small pen & paper or card game prototypes that capture the essential essence. The weakness here so far is that I don't have many victims play testers to try the ideas out on yet. My best buddy unfortunately moved to Nevada, and other best buddy is so taken with raising kids that he can barely spare a moment to look at games. So I've got to widen my gamer connections to really make this last part worthwhile. This last part, btw, works constantly in tandem with technical limits. I find that when I start having to specify rules or moves or resources or animations, I really get real about an idea. But a weakness here I admit is that you can get so focused on technology and resources that you become creatively conservative, only willing to risk on ideas that you know for a fact you can build. I still don't know what the right balance is, but admit after working in gaming shops where I've seen theoretical technology (especially AI) crash and burn badly I err pretty strongly on the side of off the shelf approaches. How does this process work for you?

Quote:Original post by Trapper Zoid
One aphorism I'm trying to incorporate into my design strategy is "Good game designers know what to put into their designs, great designers know what to leave out" (does anyone know who orignally said that? If not, attribute it to me [smile]).

-Trapper Zoid, circa 2005, once said that good game designers... [wink]

Quote:
I find that with my designs that at first I try to cram everything I've ever wanted in a game into it; fabulous graphics, innovative music system, unprecedented technology etc. It quickly becomes much to much for a single lone hobbyist developer to handle.

Yesh, I know what you mean. If this were a crime I'd be doing triple life sentences.

Quote:
So my new design principle that I'm trying to stick by I call the "One Great Idea".

Wow, this is a great way of expressing it. At GDC I ran into a similar concept called the game's central aesthetic, but not expressed in terms of all other ideas supporting it. I've also heard some designers (like Will Wright) describe a central gameplay loop, but that touches on the repeat process of gameplay.

I like it! Thanks for the insight.

I've played a MMORPG called puzzle pirates for a while, whose developers post often to their forums and are otherwise generally accessable. Some things they've done is have a few 'rules'. "all acts on land are consentual", "stuff that makes people play together is better than stuff that isn't". They also have 'case law'. Essentially things that have come up in game, and were ruled upon, and are thus similar to the rules but usually more applicable to stuff that's hard to codify. Things like "blockades should be short", "buying up all the cannonballs on the island is not okay".

It seems like these rules are very good about keeping the game pointing in one direction. It seems far too often in these sort of things, games have some good ideas but the ideas pull in different directions. You end up with an uneven playerbase, or something that's hard to balance, or something that's just not as fun as it looks on paper.

As for me, I try to do design the same way I code. I try to abstract out different areas into their own isolated system. Then I give one at a time a try [actually making numbers, and unabstracting the idea into codified rules]. The first few suck. Sometimes I just find the idea is untenable. Sometimes something good comes out. I take the good, and save it. Then it's just a matter of piecing the bits together and playtesting.

Though most of my designs are either steaming piles of crap, or shiny abstract ideas with perfect imaginary implimentation... so umm, that procedure is yet to be validated :]

Well, when I try to define my abstract ideas, I start out with thinking about the reasons that they appeal to me, in their own "warm, fuzzy" way. I think about why I'm turned off by a particular concept, or why it seems that this idea won't work. For me, it's about understanding how my thought processes work and why I like certain things. For instance, I hate turn based combat and random miss calculations, independently. My reasoning is that turn based combat doesn't cater to my desire to control dynamic and complicated systems on a more subjective level - having time to think doesn't satisfy the simpler, less conscious urges to whack on things. Having random hit calculations leaves me frustrated - conceptually, it seems to me that I'm being cheated of a few good hits without reason.

The next step, of course, is to sort out what is just obsessive compulsive irritation (All of us have some streak of obsessive compulsive 'disorder') and what actually is and isn't a problem. I find myself frequently thinking back to things like the human psyche and the two basic desires of all humans: To survive and to feel pleasure. On top of that, there are things like concepts of power, self, and logic. People don't like things that threaten survival - imaginary or not. People don't like things that take power from them, hurt their concept of self (either the individual or the larger self; family, country, community etc.), or break out of what they see as an established mold. This is why people find change frustrating, and why things that "don't seem to make sense" no matter how trivial, are irritating to some people.

Naturally, a good game caters to the needs of the human pscyche in the best possible ways. This doesn't neccesarily mean just giving people power, self, logic, pleasure, and garunteed survival. It's much more complex than that, and there are infinitely more dimensions. But, I try to think of why I feel certain abstract things, based on these principles, to figure out how they affect the player's experience of the game.

Most of this is subjective and subconscious, however. I generally try to 'feel' out an idea, test it, and see how it works. I tend to think in ends/results/states of being "The game should be fun," "The player should feel safe," "The viewpoint should be isometric," so my challenge is then to figure out why I think that this is a worthwhile end, and how to get to that point.

Yet, I'm too lazy to adequately document the goings on in my mind - I'm more directly oriented and prefer to design my ideas than translate them into written words. I don't like to dream up amazing things without the context of plausibility. Well, I love to dream, but more important to me is knowing how I'm going to make that dream work. I'm not naturally this way, of course; I've simply grown tired of unrealized ideas and prefer to imagine how I will get my concept to work instead of what it's going to look like. I easily know what I want it to look like - I get bored of worrying how I'm going to get it there.

I have in my mind a sort of skeleton model of a gamer playing a game. This includes a lot of theory about why gamers enjoy playing games, and an anatomical model of the 'complete game' i.e. one which has slots for every possible type of game content and gameplay. Then there are three places I can start from - a story idea, a gameplay idea, or a gamplay genre. Then I lop off any bits of the skeleton that don't work with that. The goal is to end up with a game concept, a desired features list with general strategies for implementing these features, and a rationale of why gamers will want to play this game, why these features will fit together well, and why the game will be possible/practical to implement. That's my requirements document, and from that I fill out the skeleton and implementation methods to get a game design document.

This is fascinating reading, its great to see most of you have to deal with the "warm fuzzy" concept conundrum. Without my ability to vividly see situations and ideas unfold in the idealist way I don't think I would have any chance at designing good games, but equally important is applying the process of "uncompromising analysis", as Wavinator describes.

Just like you guys, I can get easily side-tracked by this ideal game vision, which actually holds no true video game function or form in my head. I specifically remember the design process that woke me up to the shortcomings of not questioning the idealist approach. I was designing and building (over two years!) a modification - or private shard, for Ultima Online. Over those two years I built more and more, moving from one idea to the other, with only a vague vision of how it would all hold together. At the end I had some stuff I was seriously proud of, individual aspects were beautiful or great fun but I was sure no one would touch it.

I also recognize the "central game play loop" as very important in my designs now. Which makes earlier projects seem even more perverse in their lack of any central game play device.

I've also found that the ideal-vision sometimes cant be 'fixed' by good analysis and careful thinking. I recently fell deeply in love with creating an RPG using pixel art (see my avatar), this idea went out the window, when i began testing the ideal vision. I realized I was in love with the artwork not the game, as a result i'm now toying with this graphic style, in an entirely different genera.

[Edited by - DogCity on July 30, 2005 7:13:52 AM]

Well the way I see it is that every good game design has one core concept, and a tree of features that connect to this concept. You should also be able to define the core concept and features with single sentence. The further away from the core concept a feature is the more you have to ask yourself whether or not it is meaningful to the core concept. Further a good design should have no feature unconnected or rivialing the core concept. The more features you have that don't contribute meaningfully to the core concept the more likely it is that you're frankenstiening, which is essentially combining lots of unconnect ideas into a frankenstien's monster equivalient of a game. It is something that I think most of us are guilty of, even when we have a good core concept there often times seems to be the desire to task on just one more "cool" feature.

So, if I was outline Pokemon which is arguably the most succesful game franchise of all time - if you disagree then feel free to name another franchise that has spawned about a dozen video games, a long running cartoon series, 4 or 5 movies, a collect card game, and thousands of toys.- it would look like this:




The high level design for pokemon look basicly like that. With the three main features being collecting, battling and training of pokemon and those main features having a a few primary features some of which are interconnect and thus reuseable such as items. In this way you can see that no feature in the design is more then one degree of seperation from the core concept.

This sort high level design is the kind approach I attempt to take when it comes to game design its useful because in both developing a good game design and determing what ideas actually belong in the design. If you have a hard time in figuring out how going on missions on behalf of aliens, through randomly generated dungeons to recover lost technology from an ancient civilization and then using the rewards from those missions to build the village you live into a thriving city state connects to pokemon then there is a good chance it doesn't belong in pokemon.

Quote:Original post by TechnoGoth
Well the way I see it is that every good game design has one core concept, and a tree of features that connect to this concept. You should also be able to define the core concept and features with single sentence. The further away from the core concept a feature is the more you have to ask yourself whether or not it is meaningful to the core concept. Further a good design should have no feature unconnected or rivialing the core concept. The more features you have that don't contribute meaningfully to the core concept the more likely it is that you're frankenstiening, which is essentially combining lots of unconnect ideas into a frankenstien's monster equivalient of a game. It is something that I think most of us are guilty of, even when we have a good core concept there often times seems to be the desire to task on just one more "cool" feature.

This is a great point. I definitely struggle a lot with this one. But one of the things that really strikes me here is the paradox of the frankenstein versus coming up with something entirely new.

Imagine, for instance, what you would think if you got this amazing brainstorming that involved competitive fighting and farming. At first blush you might say, "Bah, these are unconnected, no way do they go together!" But if you were really creative, you might be making Monster Rancher.

So I'm curious what people think about this in general, if they have similar experiences.

Quote:
If you have a hard time in figuring out how going on missions on behalf of aliens, through randomly generated dungeons to recover lost technology from an ancient civilization and then using the rewards from those missions to build the village you live into a thriving city state connects to pokemon then there is a good chance it doesn't belong in pokemon.

[lol] It depends. Are the dungeons filled with pokemon? [wink] (Good point, though)