Blog Archives

« Height Changes Designing on paper »

Making a map: first steps (Jul 20, 2005)

One of the hardest bits about creating a map is actually starting one. There seems to be the perception that a designer just sits down at the PC, and half an hour later they've got 5% of the map done. That's not how it works.

Creating a map requires a different process for everyone. Some designers can just sit down and start designing what will turn into the final map. Others try something, scrap it, try again, scrap it and so on until they're happy. Others design many parts of the map separately then pull only the best bits together. These are mixed and combined; the process is rarely predictable.

Generally I think the second approach is most common: you try a few things and if it isn't working, you scrap it and try again. This is essentially an iterative process whereby you try the most obvious things first, then work on what you learnt to improve or try different approaches the second and third times round until you have a fair idea of what is good and what is bad.

Sometimes things are more drawn out. Dust PCG and one of my Nightfire maps, Romania, followed this pattern.

Dust PCG started with several small maps, each featuring just a very small portion of Dust geometry, but none of them went anywhere. One I remember featured just a section of underpass and a circular staircase, but even that small section felt wrong, just too 'busy', so I opened up a new map and just copied the useful brushes of geometry across and tried another approach. Often a design I've got in my head or on paper feels entirely different in 3D and highlights all the shortcomings.

The first versions of Romania went well, with 2Fort-style bridges and buildings, and let me experiment with a very vertical of play. The second versions used a flatter, more TFC-like design, with two small town areas connected via the perimeter of a third town area. The third version built on the lessons of both, adopting the looming buildings and long drops of my first attempts with the connected-town areas idea of the second attempts.

Of course, these aren't really the 'first steps' towards making a serious, original map. A proper map requires first obtaining some assets, identifying architectural styles and designs (e.g. reference photos) and using these to establish some geometry. However, casual map design follows the basic principles outlined above.

« Height Changes Designing on paper »

user comments

Ray at 15:32 on Jul 20, 2005

What do programs do you use to make maps and models etc?

Simon at 16:08 on Jul 20, 2005

I'm guessing he uses hammer.

Anyway, How is google adsense working for you john?

Anthony at 16:38 on Jul 20, 2005

i am a map maker and I know for a fact he is talking about hammer. Anyway, this should help new people.

dave at 16:58 on Jul 20, 2005

Ray: I use Hammer for maps, Blender for models.

Simon: it's alright, nothing too amazing. Just enough to fund the hosting, which is exactly what it's for.

Magikus at 12:30 on Jul 21, 2005

You described the typical iterative process in finding style and architecture for your map. But what about the original maplayout? I find it most useful to create the maplayout first, outside of hammer, just to know where I want to go with the gameplay, before I'm losing myself in art. Do you do this or are you creating the layout in the progress of creating the whole map?

dave at 16:47 on Jul 21, 2005

Magikus: I find it easiest to get reference material, build a few things in Hammer to get an idea of what I can and can't do with the theme, then work from there (both inside and outside Hammer) to establish a layout. The aim is to establish some general patterns and design principles that can then be adopted for the rest of the map. Often really cool features of the theme/architecture can contribute to the gameplay in a very unique way.

Graham at 00:52 on Jul 22, 2005

When I map it's not really "make something, scrap it, try again", but that's close. I guess it depends on what you mean by "scrap it". Do you mean delete it and start over? Because I don't do that. Mostly it's a case of, if I don't like something, simply re-working it over and over. As opposed to scrapping it and re-doing it from scratch, I just move the walls around a lot until I get something that works well.