RA2/YR: The Road to a 5-Star Map: General Guide to Map Planning and Designing
by CannisRabidus. Published: 1/6/02. Last update: 6/3/03.

Making a multiplayer map for any game is an art and not a science. So there are no hard and fast rules to follow. Still, it is easy for a budding mapmaker to make a failed map, as we have all seen several times. There are several reasons why this happens, but I believe the three most important reasons a map fails is:

1. Poor planning.
2. Inadequate testing.
3. Insufficient attention to detail.

So, what I'll do here is describe in very general terms the process I follow when making a multiplayer map for Red Alert2 and for Yuri's Revenge. Maybe this rambling will help someone, or maybe not... in any case here goes.

The Basic Layout

First begin with a design concept... a general idea of what the map will be like when it is done. It's OK for the design concept to be flexible - several of my maps turned out differently than originally intended. Design concept includes the layout, the theme, and whatever else creates the character of the map.

Map size is kind of a slippery concept... after you have a few maps under your belt you'll pretty much know what's too big and too small and what's just right. Remember you can always change map size under Edit>Map.

The center of the map should be discovered and marked. When you first begin a new map, Final Alert2 places waypoint #0 at the map center. Waypoints can be used for markers, but I like using the "pavement detail" tiles (the ones that don't LAT). Note that a lot of area at the top of the map cannot be reached in the game by units. So after finding the center, enlarge the map height by 5 cells, giving all 5 cells to the top of the map. (Be sure to adjust the visible area of the map to compensate for the changed map size). This map enlarging will ensure that the playable area of the map is the shape you intend it to be.

If the map layout is to be symmetrical, more pavement tiles can be laid outward from the center, and used as guides for the overall design.

Setting up for the First Playtest

Continuing the layout design, make a simple version of the map - just a flat featureless map with temporary water tiles placed (with autoshore off) to create barriers where cliffs will eventually go. Slap down a bunch of temporary ore mines and a pile of temporary gems at each base, and then play against the AI in Free-For-All mode. Watch for base-building and pathfinding problems. To do this, some mechanism for revealing the map must be present - a reveal map trigger could be used, but I am fond of a SpySat=yes and Radar=yes mod to conyards. Either way it is important to be able to see the whole map from the start.

If the AI can find its way around, then player units can do it too without needing micromanaging. Make sure to study how the AI builds its bases as the game progresses. Many maps do not allow enough room for a fully-developed AI-built base. When you playtest, superweapons should be on since the superweapon buildings are large and take up a lot of base-building room. If the AI cannot build what it plans to build, it often never musters an attack either.

You may be unconcerned about the AI on your map if you intend your map to be played online only. But it is a fact that most players of any computer game never play online. If you want your map to have the widest appeal, you should support skirmish play too.

Time for Terrain

Once pathfinding and AI base building appears OK, then it's time to start making the real terrain. My advice is to never-never-never-never-never use the cliff-painting tools. Ditto the auto-shore tool. Personally I don't use auto-level either. Instead, I sort of 'sculpt' the terrain with raise/lower tools instead. Make things interesting with hills rather than just pancake-flat plateaus, and try to find innovative uses for the cliff ramp pieces, else the map looks boring. This work is best done in framework mode. Playtest occasionally during the terrain building, since new pathfinding problems can turn up. If they do, sort them out before proceeding with making more new terrain.

Final detailing of hills may require manually placing ramp tiles. The "raise/lower single field" tools are invaluable for this.

Only after the map is fully 'terrained' is it time to think about trees and buildings. Only after the map is fully detailed is it time to think about triggers and mods. And do not forget things like lighting. What I'm saying here is, get the terrain done first!

If you are creating an area of interest that requires key building placement, well of course you will have to place those particular buildings and position the terrain around them. Once this is done, remove the buildings - you can always restore them later.

Now lay out some more temporary resources and playtest again. If you find you need more room, always remember you can enlarge the map. If you have a lot of unused space, you can get rid of it by changing the map size too.

Detailing

If the map is working out well, and you are satisfied that the maps size won't need any more changing, then it's time to start detailing. Begin with the LAT tiles. Look at Westwood maps - hardly an area is untouched with extensive LAT tiling. This is what makes flat ground look less flat and uninteresting. Be careful not to overwrite cliff tiles and the like when placing LAT tile detail. Finish up with 1x1 LAT detailing so your detailing looks less like square blotches. Use the "show/hide" tools to hide all cliff, slope, shore, etc. tiles, and you will be able to place 1x1 LAT tile detail much faster.

Now it is time for resource placement. Ore and gems are best placed while in framework mode, this way you can see where you are actually putting them. No overlay (and this includes rocks!) should be placed on slopes. Trees can be placed anywhere that looks good. (However a one-cell buffer of clear space should exist between all ore and trees). Buildings should never be placed on cliffs, and they should never overlap each other (especially garrisonable buildings).

Objects of interest like vehicles and people can be placed if appropriate for your map's setting. A city without them looks empty. A bunch of people standing around in the middle of a heavily wooded area looks silly.

Just keep in mind that lots of objects will slow your map down. Don't go overboard with faithfully creating a world with every little detail, when just enough details to suggest the world are sufficient.

Attention to detail leads to quality in all areas. A map with poor quality in one area (especially terrain) is a crap map even if everything else is great. If your map has a flaw, players will take notice of the flaw and ignore everything that's good. So make sure there are no flaws.

Learning from Other Maps

The way I got started making maps was to look at other maps. After downloading a map, the first thing I do is look things over in Wordpad (I can read mods and triggers in text very easily by now). Then I play the map, and I look - really look - at the way the terrain is constructed. If I see something I do not like, I resolve to never make the same mistake in my own maps. If I see something I do like, I look at it really carefully to see how it was done.

Notice I didn't talk about triggers and mods at all. They are the icing on the cake - first you have to make a good cake.