If I was someone who knows how to draw, I could look at something and put it on paper. Ok, I know it’s not that easy. It requires points of reference, models, a lot of practice and so on. Even so, I am not that person.
This makes it really difficult to draw the hard stuff, like the human body. I wish I could get one of those little wooden dolls, position it the way I want it and copy it. But I need something more exact to start from.
Then there are programmers, and when programmers find a roadblock we sit down and code a solution. Once again, not that simple. But it helps.
That’s how this little tool came to be – and believe me, it was tricky. Say hello to the virtual wooden doll:
So, what does it do? Well, mainly the same as that other tool. It creates models for people, using the same perspective as the scenes. After all the variables are set up, it prints a page like this one:
It is intended to generate the main positions for a character, but I’m not doing it right now – I’m mainly creating some random poses to some random characters so I can test it. So, after this is printed, we can start some concept art. For instance, here’s David Green:
No, I don’t like it either. I don’t know why, but this guy was created with a common face that’s simply bland. If (some day) we get to work on Trapped again, this will have to be solved. But right now, it’s not a concern. Let’s move on.
And here’s a ghost/zombie:
And, finally, this is Betsy:
Next steps are to “ink” these drawings, much in the same way as the scenes. This is what the next post will be about.
Anyway, as a side note – before deciding to create the tool I researched a lot for free, simple ways to get the same kind of results in the Internet. No luck – there are some good references, but I really needed a tool. So when this one is over, after we enable animations, we’ll upload it here and leave it free for use to anyone who has the same issues.
Hi again. Where were we? Oh, yes. I showed you a small tool that generated a 3D wireframe, as the one below.
Now, this needs to become a cartoonish drawing to use in a game. It didn’t look promising. So let’s see what happens next. First off, we go analog. Which means the wireframe gets printed and I pen a quick draft on top of it.
Next, the draft is scanned and the colors and shadows are added, layer by layer.
After some time and a few tries, this is what I’ve got:
And that’s pretty close to what we expected. It took more time than it should, but that’s only a matter of practice. The more I draw the scenes, the quicker and better they become.
I wanted to test that theory. So I tried a second scene, one that could be compared with previous games. And sure thing, it took half the time.
And I gotta say, I’m happy about it. It’s almost there on what we’re trying to follow, and it’s sustainable – easy to reproduce over and over.
Of course, none of this means anything unless we can follow the same style for the characters and items. So that’s the new challenge. Next: characters. I’ll see if I can finish some concept art tonight.
So, we’re not going back to isometrics for the next games. There are a few good reasons for that (the fact that we don’t like it as much is not the worst of them), but the details are not important right now. The thing is, we want to keep the scenes based on perspective and cameras.
This poses a series of problems. Isometrics are quite easy to draw, but real cartoonish scenes demand you to actually know what you’re doing. Or that you cheat.
I like cheating.
For The Labyrinth, what we did was to generate all scenes within Google (now Trimble) SketchUp, and use textures everywhere. It worked, but it wasn’t close to perfect. There wasn’t much control on lighting and colors, and all lines and angles were too straight and hard. And integrating the scenes with the game engine was like tripping and falling face-first on an ant colony while running away from furious african bees. Not really pleasant. We needed a better way to cheat.
That’s what I’ve been working on in the past couple of weeks. A useful little tool to be able to create scenes for the game in a way that’s easy, looks good (maybe not great, but good) and works with the game engine. This last part is important.
The result is what you see above. It has a number of functions, all aimed for the creation of rooms and environments using a simple grid. After you put all the walls, doors and objects in place, it generates a simple wireframe.
“If all you needed was a wireframe” you may ask, “why not keep using a 3D tool and then doing whatever you’ll do with it?” Good question, actually. Well, an actual 3D tool would be a bit overkill for this. But this is not a good reason. However, our tool does more than just the wireframe. First of all, it keeps the perspective fixed and optimized for the game engine. Second, and more important, it generates maps along with the wireframe. These maps will be used within, with no need for edits, to regulate the walking behavior for the characters and clicking responses for the player. This is the whole integration part that we had issues in The Labyrinth – and not anymore.
“But how is that wireframe supposed to, you know, look good?” might be your second question. The answer for that will be in the second part of this post, some time in this week.
* Bad pun intended. I’m sorry.
Blog powered by WordPress