__ ___ _ _ __ _ _ __ __ ____ _ _ _ _ __ __ ____

Game Development as Picture Development
 

If there is one 'most important secret' that I wish this article to convey, this section is it.

Think of those Polaroid instant pictures (If you have never seen one of those picutres, think of a .gif image that starts out fuzzy and fades in to clarity as you load it)... the kind that start out grey and gradually fade in through blurriness to clarity. First you see some big patches of color, and you realized that that is someone's shirt. Then you see a thing that looks like a human head. Then you can see that there are eyes and a mouth. Then you recognized that that person is you.

That is what the creation of a game should be like. It starts out as an outline where you can recognize things in general but can't see details, and gradually you add details, until the game is fully developed. You shouldn't start making a game at the beginning of the game and then add to it chapter by chapter until you reach the end, making things in the order the player of the game will see them in. This may be the intuitive way to make games for the person just starting out, but it's now how the best game designers make games. Think of a painting. When you paint a painting, you don't start at the top and go down, painting down inch by inch, until you reach the bottom corner and the last thing. When you draw a sketch, you don't start with the top of the head, to the rest of the head, to the neck, to the body and arms, to the legs, finishing with the shoes. You paint and sketch in broad strokes at first, getting a general vague outline on the canvas or paper, and fill in the details later, and finally color it.

Here is the way I envision most people making a large scale Ohrrpgce game: They begin their game. They think 'what is the first thing that a game needs? a title screen!' so they make a title screen and import it. Then they want to 'make the place where you start the game', so they decide to make that. They start with the heroes room. The draw the hero graphic. They draw his bed graphic. The draw the rest of the tilemap for his room. They add detail upon detail to his room. They put a sword next to his bed for him to pick up. Then they draw the tilemap for the rest of his house and make his house. Then they decide that they should put someone in that house, so they draw another walkabout graphic for the mother. They put the mother in that house. They give her a textbox, that says something like 'here is 100$, spend it well, blah blah blah, go get this and that'. Then they make the town, and add more houses, house by house. The make more walkabout graphics for heroes as they need them, and put them in. Each time they put in a new npc, they then go and make the textbox for that npc. They keep doing this until the town is done. And then they start on the overworld. By this time, a month has passed and they only have a town and the beginning of an overworld. Eventually they may finish the first dungeon, painful detail by painful detail. When they want to make the boss, they draw the boss graphic and make the boss battle, decide he needs attacks, so they draw those attacks, adjust the attributes of that attack, and give that attack to the boss. They release a demo.

This is a ludicrous way to make a game. This is akin to painting a portrait from top to bottom. It's akin to writing a symphony from first note that the symphony plays to the last note that they play, instead of starting with the most important melody and going instrument by instrument. If this is the way most people make games, no wonder so many of them are abandoned.

The way the best game designers make games (including the professional game designers) is to make the game in broad paint strokes, and then gradually fill in the details, one aspect of the game at a time. When I made the game And& (which i made in a little more than 24 hours), I didn't go detail by detail from start to finish. What I did was do everything in large chunks. I made the two maps. Completely. I drew the walkabouts/npcs. All one of them. Then I drew every enemy graphic, from first enemy to final boss, all in a short span of time. Then I imported all of the music. I didn't import the music as I neded it, I knew how much I would need and imported that amount of music. I then made all of the items and all of the shops. After that, I made the battles/enemys/stats. All of them at once. I didn't draw one enemy and then give him his stats (that would be an insane waste of time, and I wouldn't have finished the game). I drew all the enemies at once, and then later gave them all stats at once. I put in the battles and tested them until they were finished. I didn't make the beginning battles, test them, make the middle battles, and test those, etc., I made all the battles at once, and then tested them all at once. I then put in the story. All at once. By this time, 16 hours had elapsed, and I was basically done with the core of the game. So I then playtested through the game over and over, changing anything that needed to be changed, fixed spelling mistakes, adding little details like making Ampersand's eyes transparent, etc. Then I shipped off the game to Charbile, w/ instructions on making the battle graphic for the hero (or heroine if you use gendered words). He did that, sent the game back, I put in the ending (which he was supposed to do but didn't have time due to an unforseen eventuality of some kind). And that was all, and all was well.

The moral: work on one aspect of the game at a time, and finish that aspect before moving on to another. You can always add to it if you didn't foresee something, but try to foresee everything your game will need in regards to that one aspect, and complete that aspect as best you can. Don't try to make the game linearly, map by map, from beginning to end, as the beginners do. This only leads to frustration. And it'll be very apparant in your game's quality that went about making your game in a short sighted fashion.

If you need it any clearer, here:

Do not do this (method 1).
1) the walkbout graphics, maptiles, npc textboxes, items, shops, and battles needed for "First Town"; 2) the walkbout graphics, maptiles, npc textboxes, items, shops, and battles needed for "First Castle"; 3) the walkbout graphics, maptiles, npc textboxes, items, shops, and battles needed for "Overworld"; 4) the walkbout graphics, maptiles, npc textboxes, items, shops, and battles needed for "First Dungeon"; 5) the walkbout graphics, maptiles, npc textboxes, items, shops, and battles needed for "Next Town".... ((repeat for every new part of the game that comes into your whimsical mind))

Do this (method 2).
1) all the walkabout graphics; 2) all the maptiles; 3) all the maps; 4) all the enemy graphics; 5) all the hero graphics; 6) all the battles; 7) all the story; 8) all the attacks....((repeat for every game aspect in an order that makes sense)).

Why exactly does this work and why is it much better and easier and simpler to do this? Afterall, you're getting the same work done either way, right? No, wrong. In the first way, you get distracted too easily, you can't really focus an anything, since you're constantly jumping all over the place. In the second way, you are able to concentrate on one type of creative ability at once. The larger your game is, the more time you'll save by doing it the second way.

The only possible opposition to this method is that people will say "But won't this delay the game's release? You can't really release demo after demo in this way." That is an understandable objection, but it rests on a certain fallacy... you aren't making a game to make a collection of demos, you're making a game to make a game. Who cares if you can't release a demo? You can release one when your game is nearing completion. Making a game and making a bunch of demos are two different things. It's better to spend a bit longer and have a finished game than make 5 demos, spending as much time on each demo as you could have spent on a full game of the same length.

Is it impossible to make a good game using the first method? No, not at all. A lot of great Ohrrpgce games are made that way. Wingedmene: Part One is one example (although even in that game, most of the enemy graphics and hero graphics that will be needed in the game were made early on in the game's development. It is only the maptiles, maps, and textboxes which are made area by area). So it's not impossible to make a game by the traditional Ohr way of making it chapter by chapter (first popularized by Wandering Hamster). It'll just take a whole lot longer to do it in installments than it would otherwise.

Now, the rest of the article is long, and I don't want you to forget the main point of this section by the end of the article. When you are sure you won't forget that the point of the article is to try to get you to try out Method 2 instead of Method 1, go on to the next section.

Previous Page || Contents || Next Page