PDA

View Full Version : ETQW Map Design Workflow


Hakuryu
08-28-2007, 08:41 PM
I've been making some terrain in Maya, and buildings in the Doom 3 editor, while trying to study QW maps and come up with some good ideas that fit the gameplay. From my notes, it seems like this to me :

1. Although maps can be large, you still end up with objectives rather close to one another. Having something long range, like an obj at the end of a long freeway wouldn't play well. For designing maps, this makes me think of all the smaller objectives as pieces of a puzzle, perhaps better off left to fit into a terrain, rather than making your terrain and then fitting buildings into it.

2. Real world isn't necessarily good. From designing an Alactraz map, I found that trying to fit routes and gameplay into something realistically sized is not a very good idea. Even if you can make something work, it would probably work better to design with objectives in mind rather than trying to stay true to a real world place. Currently I'm looking at revising my Alactraz map with the basic feel and structures, but arrayed (and sized) in a manner to fit a certain type of gameplay rather than fitting the gameplay into the map.

3. Buildings solid or not? I spent alot of time creating interiors for some of my buildings, before having a finished map in mind, and I think I wasted my time (other than learning/work value). Now I usually just make solid buildings, and if interiors are called for, I think it will be easier to make them once I have the solid buildings placed - ie I can see where an interior would be good to have better with surrounding buildings/terrain finished.

So my workflow idea is this. Instead of working on maps, work on objectives. I like the idea of a southwest river canyon, and instead of starting on a terrain, I'd start by designing objectives seperately to fit my theme. Once they are layed out, then I'd drop them onto my terrain, fit them together, and then craft the terrain around them.

Anyone care to share how they design, tips, or any critiques of this post?

Mordenkainen
08-28-2007, 09:07 PM
I think you hit the proverbial nail on the head with your second advice, I've seen a lot of people suggest using Google maps and whatnot and while you can make really nice scenery terrains with them, they aren't very good for making interactive maps. On my first Q4 map I did use real height data for my terrain because it was just supposed to be in the 3D skybox, for my second map which required interaction I pretty much built the terrain from scratch. Using real world data can still be useful as a guideline but people should understand they'll need to work at it to make it play well.

I'd expand on your advice about buildings to cover anything really. People should sketch out their maps on paper, then build mock-ups to test the gameplay and only concern themselves with texture selection, geometry detail much later on. It's not uncommon to see WiP threads on Doom3World.org where ppl post good looking screenshots and then say "ok, first room is done".

[]v[]
08-28-2007, 09:18 PM
I've started with the location and terrain as you have.
But no detail in the buildings, as you said this would come after
you have decided on the objectives.
so my order would be.
1. theme, location.
2. build a terrain
3. put in some basic buildings as place holders

At 1 i have decided where the main objective is so ...
4. add detail to terrain in main objective area.

As i start to build on the objectives i can add detail to the necessary areas.

When final obj plan is place, then create more detail for buildings.

I think the objective plan is always going to change and will depend how it feels to move around a ready built terrain and may give you additional ideas.
You may well see key points that you wouldn't have seen unless you created the terrain first. So yes i dont think its a good idea to put a lot of effort into detailing buildings unless you know it will be a busy area. then again if it doesnt add to fps or increase file size to much it may well just make another interesting place to explore for the wandering player. These i would assume be applauded as its seen as extra effort to create some detail into your level.

@Mordenkainen I hope your wrong about the real world data as thats what i am doing. But its not a case of just plonking it down and leaving it. Of course it has to be adjusted. Any terrain you create no matter how, has to be adjusted, thats just common sense.

Mordenkainen
08-28-2007, 09:24 PM
v[];118700']@Mordenkainen I hope your wrong about the real world data as thats what i am doing. But its not a case of just plonking it down and leaving it. Of course it has to be adjusted. Any terrain you create no matter how, has to be adjusted, thats just common sense.

I don't think it's a matter of being wrong or right. For me, the time it takes to find a good real world location and then change it for gameplay basis is better spent on doing the terrain from scratch. Obviously, other people have different workflows. As long as you're happy with yours it's all good.

channie
08-29-2007, 10:52 AM
So my workflow idea is this. Instead of working on maps, work on objectives.

Seconded.

1. Theme & Location, Who's Attacking / Defending
2. Primary Objectives
3. Primary Spawnpoints
3. Primary Routes to connect objectives
4. Secondary Objectives
5. Secondary Routes
6. Secondary Spawnpoints
7. Vehicles
8. Terrain comes last.
9. Building, testing, tuning, tweaking, ...

I've got some articles about designing a map for QW gameplay wise. i might post them here once i'm happy with the translation. If anyone want to give me a hand, drop me a line :)

spirit
08-29-2007, 01:00 PM
There's different approaches, but I'm doing it the way Channie suggested (not only for ET:QW): have the complete design/layout of your map in my head (or on pieces of paper, 1.-7. of his list) then start building it (8.+9.).

This means that using realworld terrain is not an option.

I haven't made large outdoor areas like they are needed for ET:QW yet, but from my work on maps for q3a and other games I noticed that having realworld stuff in there is often a bad idea.
Example: Who wants to fight in the typical real-world building that consists of long corridors that have tens of dead-end rooms connected to them?