Post news RSS Devlog: How we start a new project

We talk about how we started our project, what inspirations we used from other devs and finally what approach we took for the pre-production

Posted by on

As we were getting preprepared to start our next project, there was one thought in my mind louder than the others:

"The most risky thing you can do is not to take a risk"

Grossly simplifying, most AAA games are sold by brute force. First, lots of money is pumped into the production. They choose a known formula and master it with better graphics, better performance, more content, good balance. The game is quite expensive to make but it competes with lots of other fairly similar games and there's no guaranty that it will reach enough audience to turn a profit. This is why up to even half of the overall budget is planned for the marketing and it's basically spaming the potential buyers. It's all done to minimise the risk of loosing money and it all requires lots of it.

Indies obviously don't have that kind of money at hand. The production itself already eats up most of their funds and they cannot afford big marketing campaigns. They need to rely mostly on their communities or hope to go viral to reach a wider audience. So their game cannot simply be good, there are planty of good games out there and lots of them will have better graphics, more content and more polish than the indie can ever dream to muster. The indie game needs to be interesting on its own terms. Indies cannot play it safe, so they need to take the risk of going into the unknown.

The idea

So we came up with an idea we think might just work. The main premiss is something we've been talking about in our team for a long time. We want the game to be a narrative experience that takes place in a model with a highly reactive social environment that responds clearly to the players actions. The player is a trader and can buy and sell all different kinds of goods on a large scale. The players primary concern will be profit, but the type of transactions they are making will have a visible influence on the societies, usually clearly constructive or deteriorating. This is how narrative part of the game is supposed to emerge without any big blocks of text or heavy story (we don't have time to write that). It's a choice and consequence based primarily on the gameplay and mechanics.

We think it's a cool idea but there's planty of ways it could be turned into a dull, flat game. Ides don't make players want to play more and share the game with others, experiences do that. So we want to turn our idea into a unique experience and we can't just leave it up to a pure luck. We need some good tools.

Creative direction

Darkest Dungeon is the game that delivers its experience unambiguously. The music, the art style, the core mechanics, the narrator, all of it is very consistent unquestionably directing towards a specific experience. Thanks to that people don't really notice some of the problems of the game and even if they do they mostly accept it. Chris Bourassa explained how they did it in a GDC speech, and they did it with a strong creative direction. That means they very diligently discussed, defined and wrote down what the game should actually be to the point where the vision itself allowed them to resolve dilemmas in any field of the development like aesthetics, mechanics, marketing etc. Then they just stuck to it.

I really cannot think of a game that is more coherent and apparently they did not achieved it trough luck they did it consciously using an adequate tool. So this is how we'll try to plan the design of our own game.

Poetic design

Sunless Sea is another game, where you can't be wrong about the experience it presents. Though it's very slow and subtle every single minute sinks you a precisely crafted experience. I had a chance to participate in Alexis Kennedy's workshop on Poetic Design where he explained how he makes his games (The only video I could find about that is here, though the workshop I've been to was much more detailed. You can laso find some interesting thoughts on Alexis' blog).

Poetic Design is a technique of delivering game narrative with quantitative game assets such as mechanics and UI (independently from any text descriptions or illustrations). It is about making sure that every piece of the game using its specific function and meaning points toward the same theme. As a result the player gets a consistent experience and in consequence a story or at least some kind of narration told independently from the text. Sounds like exactly what we need with our desire to tell a story trough the gameplay.

I like how both of the approaches seem to frame the same topic coherency from different angles and can be complementary to each other. And so, while creative direction is our plan, poetic design shall be our implementation tool.

Our framework

Having those two inspirations we still struggled to combine them and workout a version that would work for us. It wasn't easy also because some key terms such as "theme" or "experience" are being used by devs around the world quite loosely and often in almost contradictory ways. Eventually this is the structure we worked out:

Creative vision statement: is the shortest possible description of all aspects of the game (either trough explicit description or implied consequences). Its main role is to ensure the game as well as any supporting materials remain consistent throughout the whole development process and beyond. At minimum it should imply the game's genre, experience, story, mechanic and actions the player takes.

Experience: is what the games seems to be to a player. We created a description of what we want the player to feel like during the game. It doesn't need to be a single thing. In our case it was a whole set of things and relations between then. In some cases we even branch the experience into alternatives. E.g. among other things we want to give the player the possibility of feeling like a cunning space smuggler or a cold and powerful businessman.

Meaning: is what the game is actually about, what kind of message it brings to the player. You can easily imagine a game without or at least with only residuary meaning (what would it be for Tetris?) but we specifically want our game to say something about the world. It doesn't have to be a closed statement, questiotns work fine as well.

Themes: are the motives used in the game. They should be leading to a single main theme. This is how the mechanics of the game can be put into a context and trough that the assumed meaning can be achieved. This is were we open to the poetic design principles.

Story, Player's actions, Mechanics: these terms are fairly unproblematic, no need to define them I think. The only thing worth mentioning is that we believe ultimately anything that happens in the game is most of all shaped by the mechanics and this is a principle we want ot embrace.

Populating the scheme

At first we wanted to take a top-down approach where we define things in order from the general level to more detailed ones and don't come back to a once concluded level. We started with defining the experience which is supposed to be the foundation of what the game is. Then we planned to list a number of scenes and sentences which would constitute the meaning of our game and parallel to that write down the vision statement. Once that was done we were supposed to start working on the production aspects such as mechanics, art, story etc.

We've quickly however realised we are constantly moving on the generalisation axis both ways and amending more general stuff to align it to some of our specific ides. The key example of that was one of the core mechanics: flying with an aircraft on a 2D circular planet. We wanted to make it this way to reuse some of the code, balance and experience we already had from the previous game (to save time), so it had to find its way in the experience. On the other hand we couldn't think of interesting and coherent ways of putting in the game some of the experiences originally planed, so we got rid of them. Yet another time we had an idea for a cool mechanic of getting older and so we extended the list ot themes.

At first I perceived that as undesirable exceptions to the rule, but after all, we're not trying to prove anything or meet some client's specifications, we just try to minimise the risk we take to make an interesting game. The key value here is the coherency of the experience not the direction of the work. Ultimately I think this is simply about finding the balance between all the elements and not giving to one of them the rule over the others.

Moving to production

The nice thing about the framework we worked out is that when we completed it, we basically had game design document ready and in a format that actually had a chance of being useful in the development process. It's in a kind of a final draft state now and we are still looking to get a feedback on that from people smarter then us.

Now we set out for production and while there's no chance of eliminating the risk, we at least can clearly and definitely say we tried our best to be able to assume that possibly we do have a chance of actually knowing what we are doing... to an extent.

Post a comment
Sign in or join with:

Only registered members can share their thoughts. So come on! Join the community today (totally free - or sign in with your social account on the right) and join in the conversation.