I''ve been thinking of ways to build a space game recently (It''s the genre I most commonly think "This annoys me, I''d do it differently!") but I''ve had a thought. Is it better to start by designing, say, the universe first (like the way galaxys->systems->ships stuff works in the game) and then on down, or to start by working out the ship->system->universe stuff first.; Like, Top down, or bottom up design? (I have no idea if those words are in the right context) Thanks, Nick

IMHO, design your game top-down. code it bottom-up...
designing it top down will give you a precise view of how everything should fit together, and will more or less force you to abstract things.


[edited by - sBibi on February 14, 2004 12:26:44 PM]

A game that starts out with a stong metaphysical mosel will be easier to design. Decide how your physics will work, which laws you''ll bend or break to achieve the level of sci-fi grooviness you''re looking for, and what kind of scale you''ll be using (interplanetary/interstellar/intergalactic/interdimensional). Then you can set about designing the world within that model of Being.

Knowing how technology works will help you decide how wars are fought which will in turn help you decide where to put your factional boundaries. Also, so history will help. Start your world a few hundred years before the game begins, at least, so that you can have grudges, and debts, and secrets in abundance. Then design your political systems, ships, and your characters.

However, this is by no means a linear process. Feel free to bounce back and forth through the stages. What''s more, you should remember that it''s a game, and so you might actually want to start with gameplay, so that you can keep that in mind while you work through the physics end of things.

So, here''s my suggestion: Decide what sort of game you want to make, and then start designeing the world from the top down, checking every so often to make sure that you can still make the sort of game you want to make within the universe you''re designing.

Start with the gameplay - what do you want the player to _do_ in the game. Then add those things that help achieve the desired gameplay, and throw away those that don''t help with this.

I agree with Diodor. Go bottom up. Think about what the player will be doing over the course of a few seconds or a few minutes.

I read that Nintendo''s Miyamoto starts with the control scheme and the player avatar''s basic abilities (jumping, climbing, flying, shooting, etc.). I would add to this that you should consider the player''s viewpoint (i.e., camera behavior).

Nolan Bushnell (Breakout designer, Atari founder) describes "tensioning" as the cyclic increase and decrease in tension that the player experiences (in Breakout, obviously, the tension drops when the ball "breaks out" into the top area of the screen). This might be a helpful way to analyze the minute-to-minute gameplay of your game.

Starting with the control scheme is great for innovative gameplay. But in this case, that problem has already been solved, for the most part.

After that, the next decision is how free the player will be. Will it be a Freespace style game, or something more like Freelancer? If it''s like Freelancer (player is mostly free), start with the physics, then construct the history and geopolitical landscape (i.e. who gets along, who doesn''t, who''s next to whom, who has the resources, etc.), and then fill in the ships to populate the geopolitical landscape.

If it''s more like Freespace (player is assigned missions, over which he has little or no control), start with the story arc, then flesh it out with the specific missions the player will carry out, then the ships in each mission, then the weaponry each ship has, etc.

It seems I''m a top down kinda guy. The way I think of it is to start with the hardest thing first, where by "hardest", I mean the thing which will touch the most other components, and is therefore the most difficult to make consistent.

quote:

---

Original post by Flarelocke
The way I think of it is to start with the hardest thing first, where by "hardest", I mean the thing which will touch the most other components, and is therefore the most difficult to make consistent.

---

Also the most difficult thing to change later on. Really, this "hardest" thing should serve as the foundation of your game, whether it is the history of your story, the control scheme or the graphics engine. You should decide what aspect will be the anchor of your design, and make that as solid as you can before you get into other things, because it''s hard to change the shape of the basement after you''ve put up some of the walls. For every change you make in the foundation later on, you''ll have to "ripple" that change through every other element of the game design.

I just realized that what I''ve been calling the "hardest" thing is almost what other arts call the "focus". In visual art, a focus can be an area of extraordinarily high detail, or an area of low detail, or a different color, or the only region of color, but the common factor is its disproportionate effect on the way the other parts of the graphic must be created. In movies, the focus of a scene can be either in focus or not (oftentimes, there is a character well in view, but out of focus. Think of the magnifying glass scene from Austin Powers). It can even be out of sight (the box in Pulp Fiction or the cameraman in Blair Witch Project). In music, the focus can be rhythm, melody, or lyrics (in modern music, the focus is usually on rhythm). This affects the way in which music is recorded -- the instruments that establish rhythm (bass and drums) are recorded before the instruments that establish melody and lyrics (guitar and vocals). It also establishes the best way to write lyrics (based on rhythm rather than rhyme).

Can you imagine how much the final scene of Pulp Fiction would be changed if something as simple as where the briefcase (you know, the one that glows) is changes? If he set it down somewhere else instead of taking it with him? The whole story might need to change (i.e. at the very least, we have to have him go back for it). Can you imagine if the stars in Stary Night were nearly black (i.e. the focus would cease to be the sky)?

The same thing would happen to Black and White if the control scheme changed. The same thing would happen to Master of Magic if you didn''t get to pick spellbooks and abilities at the beginning. The same thing would happen to Pacman if it were a time-competition instead of a struggle for survival (i.e. no ghosts, no big pellets).

(sorry if this was a bit off topic)

I was just wondering why it had to be one or the other, and not both? I do both, and I get the best of both approaches for the benefit of my player, and I get the benefit of having each influence the other so that a better experience is had in the interaction.

quote:

---

Original post by adventuredesign
I was just wondering why it had to be one or the other, and not both? I do both, and I get the best of both approaches for the benefit of my player, and I get the benefit of having each influence the other so that a better experience is had in the interaction.

---

how do you do both? i mean, i''d assume they were mutually exclusive?