What makes a good teamplay map?

This topic ties in with two things, the first of which is the Source Mapping 101 topic that Valar started. That being how to make a good map from a more technical standpoint, I thought why not make a topic about what makes a good map from more of a map layout standpoint. The second thing is a teamplay guide I have written which might soon make it into the official curriculum here. However that may be, a part of it discusses different aspects of map quality. Since this is not the finished text, comments, additions and whatever criticism you might come up with is greatly appreciated.

Types of maps
To make things easier, you can put maps on a scale depending on how readily they have something both worth and possible to control. I will call this the map's control index, ranging from 1-5, which is somewhat dependent on the number of players (Tab. 1). In one end of the extreme you will find maps that end up in a campfest, like dm_powerhouse (index value 5), while the opposite is nonstop action with little to no thought involved such as a typical killbox (index value 1). However, for our purposes, a better example of a low value would be dm_lostvillage if played without the rpg (index value 2). In general, for any map, the control index will increase with the number of players, since fewer resources per player means greater potential for successful control.
A good map should have a value close to 3, encouraging all of thought, communication and technical skill. In a way, players can also be generalized to fit onto this scale. Depending on how their technical skills compare to their tactical skills they tend to choose a map accordingly. As the balanced player the pro is, a map with a value of 3 suits him the best since he can then exploit the weaknesses of any other kind of player.



Key map qualities
Balance (of resources): by a balanced map we do not mean one that is symmetric in design, but one where weapon and item placement promotes a balance between technical and tactical skills. This quickly becomes apparent if you consider a map with an rpg as the single item compared to one where all sorts of weapons are to be found everywhere. The former degrades into spawn luck followed by a campfest, whereas the latter would not exactly be much more of an encouraging experience. Balance is attained if you give teams the opportunity to control items at the same time as giving them an equally good reason to move around. You also have to realize that any given map is best balanced for a specific number of players.
Flow: a map in which you can move around quickly between key areas has a high flow, whereas a map cluttered with debris, narrow passages, ladders, etc., while lacking interconnecting passages and teleports, lacks in flow. Enemies are easier to control in a map with low flow since they generally move slower.
Size: appropriate size is of course important, but this should be self-evident enough to not really warrant any further discussion. Just keep in mind that appropriate size depends on the number of players.
Control points: these are built around items of high value, e.g. the rpg platform on dm_powerhouse or the items under the stairs on dm_lockdown (2 smg nades, 2 health packs, 1 orb & 1 shotgun).
Choke points: points of a map which you almost certainly have to pass in order to get to another area, especially important if that area contains a control point.
Crossroads: areas which can be accessed from three or more paths, making anyone there unusually susceptible to crossfire.
Spawn point placement: In general, I think everyone will agree with me when I say that a good map is one where the element of chance is minimized. In any map, we all know that spawning is random (except that you cannot spawn at a point if someone else is already occupying that spawn space already). Things that can remedy this are: a) trying to keep down the control index of the map; b) seeing to it that no particular spawn point is closer to a control point than any other; c) avoiding placing spawn points so that they can be camped from a position that has other reasons for guarding (such as a control point, a choke point or a crossroads). In addition, look out for areas with a high concentration of spawn points, potentially making them valuable enough on their own to justify guarding.
There are of course other map qualities like eye candy, light conditions and frame rate, but they are more general and of secondary importance to what we look for in a team deathmatch map.

What do you think makes a good teamplay map?

Team Deathmatch
I'm feeling a bit lazy now so I'll just shout out short sentences instead of paragraphs:
- Items and weapons are able to be selectively grabbed. This is to ensure that a player doesn't accidentally grab a weapon he intended for his teammate to get. Easy to implement with the usage of shelves, gg-grabbed places, etc.
- Room for more. Having narrow hallways promotes a one-way one-man passage and is better suited for stealthier, riskier 1v1 matches. You want the narrowest hallways in your map to still be comfortable for at least two guys to bunny hop in opposite directions unless specifically intended as a sneak-around ambush path. But generally, vents, long-cool downed elevators, narrow halls, and ladders promote more one-man play.
- Dynamics. You can have certain "arenas" where one team is fighting from west and the other from east. Perhaps on another area, you're having one team in control fighting from above and the attacking team from below. For a 2v2, maybe you'll want places where one guy can take cover and distract 2 guys long enough while his team mate goes around another path to flank them from behind. You can create such dynamics by including "lesser" paths, height variations, and even jump pads. Use things carefully though. Jump pads may emphasize more "1v1" skills than teamwork if used too liberally and carelessly.
- A sense of control. Maps really need a sense of control for non-DM teamwork to exist. Watching entrances, guarding each other's backs, sharing resources, working together to flank, etc. For example, Lockdown has some interesting spots for teamwork. One guy could perhaps be sniping the bottom way and watch the hall while the other guy orbs the hallway if his teammate asks for him to do so, or SMG nade guys on the bottom with his own discretion. Another scenario would be one guy guarding the fan and RPG vent while the other guy goes and gets it.

Very useful info and tips. This is so important to map design(at least for competitive play) but all you can find tutorials and help on is for the technical stuff.

Well, I thought, what with all the advertised interest in mapping and whatnot, it might be a good idea to see to it that if some of these people are thinking about making teamplay maps, why not give them some useful tips on the way.

I guess that apart from DOs and DON'Ts, you could try to find out what it really is that makes the good maps good.

Here's a classic set of articles on this topic: http://www.lunaran.com/page.php?id=9

In the end, the only way to really learn this and get a feel for it is making maps - and making mistakes. It all comes down to experience, I guess. I love reading those guides, though. It's like the Zen meditation of level design. And it really helps with large-scale decisions.

One more thing I'd like to add in the name of most mapping communities I've ever been to: Looks don't contradict gameplay. There are good looking maps that play bad. But if you make a really great layout and it looks like a huge block of concrete, you're just a lazy mapper. By making a map look good you show respect for your own work.

awesome thread. its interesting on what goes into building a decent tdm map.
http://www.lunaran.com/page.php?id=192
funny how q3dm6 --> http://www.youtube.com/watch?v=FqxY-WHaYxE
is such an epic q3 map, but when brought over to hl2dm, same layout just more half-life 2 fitting, it made me sad.

DEFme wrote:awesome thread. its interesting on what goes into building a decent tdm map.
http://www.lunaran.com/page.php?id=192
funny how q3dm6 --> http://www.youtube.com/watch?v=FqxY-WHaYxE
is such an epic q3 map, but when brought over to hl2dm, same layout just more half-life 2 fitting, it made me sad.

While I haven't really played that dm_campgrounds_pro (right?) you do realize that dm_octagon is pretty much the same thing? I know ppl have different opinions about it, but I really like it (apart from it being a resource hog).

Since Quake2 gameplay doesn't employ much strategy, weapon choice is not a heavy issue, and the weapons don't demand much in the way of it anyway. The more specific weapons in HL bring on more strategic gameplay.

LOL, I can't believe I just read that. Of course q2 can be just as strategic as HL.

Awesome. Thanks, reaper, that link looks full of win.

Very interesting site with useful tips. However, only by trying different kinds of maps and making a few of your own can you understand how to make a good team map. I learned very much about mapping simply just by playing maps I normally wouldn't like and through creating two CTF maps, then actually spending time pubbing or even speccing them and listening to people's reactions.

Connectors was a semi-popular map, perhaps mainly because of the jump pads. They allowed for a more "aerial combat" dynamic unseen in most HL2DM players, and fortunately, aerial combat turned out to be a great amount of fun to players. Since then, I've added jump pads to every one of my maps. Perhaps the biggest complain with connectors is its gigantic size. Especially without proper movement skills, players have complained that it takes way too long to get out of your base and to enter the enemy's. The giant size also led to some "akward" moments in combat, rendering all but the Magnum and RPG useless as if it were a giant killbox map. Another problem with connectors was its general connectivity and flow. Due to the use of "literally" a single-digit amount of textures, the map looked like a maze rather than 2Forts with its bland textures. Players would get lost in the giant, dark or bright (depending on Rebel or Combine, due to poor lighting on the Combine's base) corridors, humongous and dim rooms, and the "useless" ball room. I tried to make weapons help balance out the connectivity of this map, but the placement backfired a bit. The RPG on the bridge was intended to support chaos in the middle for players to struggle to grab, but the paths to it were too simple and boring (long walk or consecutive jump pad bounces), the bridge-roofs provided too much cover, and it was a tad bit tough to control it given the architecture. The AR2 in the ball room promoted spamming. The narrow hall from the flag room to ladder promoted crossbow/orb spamming and a one-way entrance, which is a BIG no-no for team maps. The giant ladder slowed down the pace and also promoted xbow/orb spam, which led players to hate the long hall and long ladder immensely, sometimes even leaving the game due to frustration. On the bright side, people really digged the jump pads and speed ramps, which offered a unique and fun gaming experience. Learning from my mistakes and achievements, I went on to make Revolution.

Revolution wasn't nearly as popular, although I do believe it is significantly better designed for smaller pubs and scrims (sorry, Revolution isn't intended to play 4v4s on a Los Angeles listen server that gives you 160 ping). Rather than posing as just gimmicky features, the jump pads of revolution were immensely helpful tools that took you to vertical routes that normally can't be accessed through foot, something unseen in many HL2DM maps except through slow and boring ladders. Being a beginner jump-padder allowed you to take basic aerial routes, but mastering it was satisfying and allowed for some intense, rewarding jumps. Jump pads also gave players a very effective way to evade enemies, which turned out to be incredibly fun when flag running (half the time in Revolution, I was just going for flags and no kills at all). The jump pads of Revolution were largely popular among DMers who learned how to use them, for both in combat, flag running/evading, and in simple route taking. The speed pipes, which were a new addition that may or may not see some action in my future maps, were also praised due to them allowing players to trek through large distances in short times. Speed ramps were not present, but speed lines (a new addition, those things on the roofs) allowed for insane velocities that exceeded over 900, allowing for some quick flag running (which was rather imbalanced, especially when you played against people who couldn't even bunny hop or find their way around the map). Revolution was a giant map too, and the pacing was either too fast or too slow, depending on whether you knew how to use the speed tools or not. Not much thought was put into weapon placement, except for the RPG, which is in a "cage" that you can only acquire by dropping down and fishing it out with your gravity gun or shooting it out. Under the bridge, there were 2 200-suit chargers as well. This made the middle a very important "stop" for players planning on camping, maybe a bit too powerful. They were easy to get access to with the only threat being snipers. Weapon/Item balance was rather poor in Revolution due to this, but the speediness somewhat balanced it out. All in all, Revolution was praised for its jump pads and other speed tools, but was hated for its size, hide-n-seek spots, and RPG/200 in the middle.

Now. King_air.

Just thought I should point out that my first post in this topic included a passage from a draft of my since then published teamplay guide. This passage in its current form in the guide is much improved, with a clearer logic behind sections. I strongly suggest everyone who is about to make a teamplay map (or any kind for that matter) to go through each section and see how it could apply to their map. If only the creator of dm_overwatch had done this with the spawn point paragraph, the world would be a happier place

yes yes and yes!thank you