Contribute .
#Challenge on ETG.
#Challenge on Quakenet.

[an error occurred while processing this directive]

News Archive
Help Wanted

Powered by:
Powered by

Challenge Player Index
Challenge ProMode



Killer Instincts

CPL Europe

Link to Challenge World
feel free to use this

How to make Levels for Q3A Tournament Play $comment_count ?>

Recently I've spent several months working on making competition oriented maps 1v1 maps for Quake 3. I've learned a lot during that time, and now that I have a column I thought writing this might be a good way to share some of that knowledge.
When I first started working on competition oriented maps it was because I'd been following the pre-Razor CPL coverage. Everyone was talking about what maps they should use. To my knowledge that event was a the first big quake 3 tournament, or at least the first big one in the US. The maps they would choose would become the maps used for most of the events during the period that people would play q3a.
And looking at the four maps they choose, and the selection of maps they had to choose from, no one was completely happy. Each map had some major flaw that was complained about over and over, and two of the four maps hadn't even been originally designed for tourney and had to be modified.
So an idea started forming, I would make a map, and everyone would love it and use it in the major tournaments, and cry out unanimously "thank you twoAM, now we don't have to use those crappy id maps we've been complaining about". I had no idea what I was getting into.
So anyways when I first started I had a pretty hard time figuring out what people actually wanted for a competition map. I know more now than I did before, but I don't think I know everything. In fact I hope that players will read this and post replies telling me where they think I went wrong. By actually getting these issues written down, people will have an easier time talking about them.

The power points around the RA in dm13. (A) is the ledge near the RL which has good lines of sight to the RA and the entrance to the teleporter. (B) has a good line of sight to the entire RA floor, and its a easy jump from here to the RA. (C) The drop down hole to the RA. (D) The RA location.

The area around (C) on the RA floor allows players to camp and wait for the RA, invisible to almost every location except (B). Once a player gets the RA the narrow doorway leading to the teleporter exposes them again to enemy fire. Also a player at (A) or (B), seeing a player near (D) or dropping through (C) has enough time to get to the teleporter destination and wait for their opponent there. But instead of teleporting, players can rocket jump out of the hole at (C).

The Q3A sound system
If an event occurs within your "visibility area" you can hear it. There are two good cheat protected console commands for finding out your visibility area, "r_showtris 1" and "r_lockpvs 1". Levels with good "vis blocking" (levels where not much is drawn that isn't directly visible) like dm13 are better for stealth (if someone is camping the RL/RA in the middle you can bunny hop through much of the level without fear of being heard) while in levels like t4 stealth is much more difficult if not impossible.
Allow players to choose weither or not to be stealthy by putting moderately valuable items (like LG ammo) near the entrances to important rooms, or in mostly concealed parts of major rooms. Make sure that there is enough space that the player can choose whether or not to pickup the item.

Space levels
No one likes space levels other than newbies and level designers. Don't even bother.

Jump pads
Jump pads are a complicated issue. In q3tourney4 jump pads are the best thing about that level. Every neat trick involves misusing a jump pad. Players can take the pad by the mega to the mega itself, or to the door above the mega. Players can rocket pad from the short jump pad by the rl to the ra. The list of tricks goes on and on. On the other hand jump pads also almost ruin that level. The limited air control in q3a means that it extremely easy to predict the movement of someone flying through the air after hitting a jump pad and then railing them. It sucks having to take pads in that level to get to the top level, and it sucks having to take the pads in dm6 to get to the rl and the top of the ammo columns. Both of those areas wouldn't be so bad if there was a good alternative, like a teleporter or a elevator, but there isn't any.
Jude, in cpm1 and cpm3, hides the path of the jump pads. Generally this means you cant go anywhere unintended on them, but also your only vulnerable at the top and the bottom of the pad path. I think this is generally the best way to handle jump pads, but in maps without a railgun, or in maps where there is plenty of good alternative routes, having a jump pad which is exposed like in dm6 or t4 is okay also.
One common annoyance I've found with jump pads is making the trigger too big. This can result in a player hitting his head on a ledge prematurely and then dropping back to the pad and then hitting the ledge again, boing boing boing, until finally air controlling out of there (this happens in cpm1 going from the low rl to the top, and q3ctf4 near the low 50 healths). Oversized jump pad triggers can also make it annoyingly difficult to go down a jump pad path without hitting the pad again at the bottom, this happens in cpm1 and q3tourney2 near the rl.
Quake 3 level designers, especially ID's level designers tend to over use jump pads. There are lots of good alternatives to jump pads which should be used more often. Teleporters, elevators and stairs are the most common and I talk about them all later.

ID didn't use any func_plats (generally called elevators) in their levels for some reason. Probably they didn't want to confuse any of their anticipated "hoards of newbie players". I was surprised to find that not only did func_plats work great, but bots also knew how to use them.
Platforms have lots of game play possibilities. They can be placed so that they block a passageway when they are in the raised position. Then a player can jump over it, leaving the passage way unblocked, or run over it and get off, leaving the passage way blocked behind him. Platforms are great ways to escape for fleeing players, since usually once they start going up there isn't enough time for the chaser to also get on. Plats can make great deathtraps as in q2dm1, if someone tries to take a plat and there opponent is on the top they are completely trapped. Also if they are wide enough, players can dodge rails while riding them.

Many of ids lower numbered maps are painfully flat. The most obvious importance of the vertical element is in RL fights, where the person on lower ground is at big disadvantage. But there are also other ways that vertical variety is adds complexity. In dm6 if one player is in the quad pit and the other player is on the second or third floor around the quad pit, the higher player has many more choices. The higher player controls when he is visible or not by approaching and backing away from the edge, the higher player also can break off the fight at any time and run away without much fear of being chased.

Q3DM7 has an example of less extreme vertical difference. In the lower RL room (QUAD and Plasma in TDM) an upper area and a lower area lay alongside each other with a small set of stairs connecting them. Players in the upper area have easy access to the lower areas, allowing them to easily give chase to lower players. Lower players by contrast only have access to the upper areas by rocket jumping or taking the stairs, meaning that they can't easily chase fleeing players in the upper areas.
For those reasons areas of even small height variation are more interesting and complex to fight in than flat areas. For small height variation stairs are the best. Stairs are quieter than jump pads, teleporters and platforms and usually easier to dodge on. Stairwells also make good alternate routes to major areas like the stairs in dm6.

Q3dm13 has a great teleporter. It's destination is the perfect place for an ambush. It also really fun to bypass the teleporter by rocket jumping up from the RA, then you can ambush the ambusher. Its a good idea to make sure that a teleport destination is not right against the wall, so that an ambushing player can wait right behind it without getting telefragged.
In general I think one way teleporters are better than two way teleporters. Attacks through a one way teleporter are all or nothing, with a two way teleporter you can come through, fire and then teleport back before your opponent can react. Two way teleporters are also more likely to result in telefragging which is generally considered a stupid and cheap frag. Two way teleporters encourage slower more sneaky gameplay, while one way teleporters encourage faster bloodier combat.

CPM3 has a good example of a two way teleporter (near the RA) where the destination is far enough from the trigger that each player has enough time to attack before teleporting back again. The area marked in red is the extra fighting space in between the destination and the teleporter. This generally plays better than most two way teleporters although having one destination face the same way as the teleporter and the other destination face a different way can be disorienting.

Teleporters are great for connecting two areas that can not see or hear each other, and for making ways to get up from a low area to a high area. I like to place my teleporters and destinations in places where the teleporting player is sometimes at a disadvantage relative to the non-teleporting player and vice versa. The teleporter by the SG in CPM3 is a good example of this. If both players are down below, the teleporting player is at an advantage, he has the upper ground, and can camp the teleport destination. But if the non-teleporting player already has the higher ground then the teleporting player is at a disadvantage by arriving in a predicable place.

Q3tourney3 is a good example of a bad rail map. It has large areas of no cover in the middle and unnecessary jump pads (they rise so little that you can only go to one place on them, and they could easily be stairs instead). These two factors mean that whoever rails the best on this level will win. That makes this is a one skill level, the other skills of tourney like using the rl, using the lg, timing the armors, predicting your opponent and keeping track of where your opponent is are not important for this level.
Good rail maps are maps where there is enough cover and enough confined spaces to make the other guns and the other skills as important or more important than rail accuracy. The railgun is a controversial gun, whether or not it is in a map is a important (political!) decision. Ideally you should test the map with good players both with and without the railgun when you think you might include it, and decide which is more fun. Maps like CPM1 and dm6 are good rail maps, the rail is important, but no more so than the lg or the rl.

Imagine if Q3tourney2 had no large pillar in the middle of the rl/ya room, that room would be much more boring to fight in. The best fights in that map involve fighting around the pillar with the rl and lg. Having pieces of cover to fight around is critical to every good tourney map. In t4, cover includes the corners of the "outside area" by the RA, RL and MEGA, the catwalks themselves, and the support columns near the rg. In dm13 cover is mostly in the form of narrow doorways and twisting corridors. In CPM3 cover takes the form of rectangular walls that hide a area of passageway, giving a player a chance to regroup, or double back.

Grenade Launcher
Often ignored, the grenade launcher is a useful weapon in the right hands. I believe it has a place in every tourney level. New uses for the grenade launcher is an area where players can continually improve their game and add depth and excitement. Also its important to have both powerful weapons and weak weapons in a level. That way weak weapons can be placed in unimportant "out of the way" parts of the map where they will be available for newly spawned players, so that they can use them to try and get a more important weapon. This adds more variety to the types of fights (SG vs RL, rail vs LG, Plasma vs rail, etc...) between players, as in the case where a newly spawned player grabs a grenade launcher and tries to use it to hold off a pumped up player with a rocket launcher.
The placement of the shotguns in q3dm13, q3dm6 and t2 are good examples of weaker weapons that are placed in out of the way places especially for newly spawned players.

Armors and Megas
Two or three armors/megas (referred to from here on as "defensive powerups") is generally a good amount for a tourney level. Having only one defensive powerup is a bad thing, it makes it extremely easy for the controlling player to collect all the defensive powerups in the level and concentrates all the fighting to one part of the level.
The number of items must depend on the size of the level. Dm6 seems to me to be a level that doesn't have enough armors. The level is too large for the controlling player to completely search for his opponent in between picking up armors, so generally all the important fights occur around the 2 armors, despite how large the map is. Since it only takes a second rate player (aka myself) 10 seconds to get from one armor to the other and it takes 25 second for the armor to respawn, the controlling player spends a lot of time camping the armor.
The ideal situation for defensive powerups is one where it is difficult (but not impossible) to run all the armors (collect all of them as they spawn). This can be achieved in a couple of ways. In CPM3 and dm13 there are just a whole lot of defensive powerups and its really hard to fight your opponent and collect them all. In CPM1 the 2 YAs are in exposed (dangerous) places to be. In q1dm4 the RA is in a pit, the only exit being a teleporter (other than rocket jumping) which takes you to the dangerous catwalk over the lava in the middle of the main room.
Since fights for the defensive powerups are the most important and the most exciting fights, it important to give them the best possible locations for fights. Good locations for fights are places with a lot of vertical elements, and enough space that each player can get a couple of hits in before one of them reaches the armor.

The YA pictured here in dm6 is a excellent location. It is easily visible and accessable from all three levels. Players approaching from the areas marked with blue are visible and vulnerable to players perched on the ledge marked with red.

This YA in CPM1 is also in a good location. Players trying to take the armor must either jump from the third floor (red line), follow along the ledge (blue lines) or rocket jump from below (red line). Timing it is critical because camping on its spawn spot is so exposed.

Powerups are bad for tourney play. The smart thing to do when your opponent has a powerup is hide and wait for it to run out, which slows down gameplay.

Doors are good sound cues. In q1dm6 the shoot-able door in the floor can be heard from every part of the level, going through is dangerous and made for some good fights, also people sometimes shoot it to fake there opponent out, or try and lure them in. In Q3a you can make a door that is hearable from every part of the level by making the door target a target_speaker entity with the global flag. Q1dm2 uses a button to make access to one of the RAs one-way, some other buttons to open a lava trap, and another button to control access to the teleporter. Doors are an interesting area to experiment with.
T4 is a lot better because of doors, especially since they are shoot-able, players can shoot them to fake their opponent out. When you arrive in the middle room just in time to see a door close, you know which way your opponent went.

Symmetrical Architecture
Symmetrical architecture is boring at best. If in T2 the two side hallways didn't have the items they have and had one of the YAs each that would be worse than just a boring design. Because in that case each side would be equally valuable and the map would have no center (see Shard/plus Fives).

Shards/plus Fives
Shards and plus fives are best as sound cues. For that reason its a good idea to vary the number of each of the shard groups. T4 has a couple of good examples of placement of shards and plus fives. If your at the RL or above the RL and you hear your opponent pick up some plus fives you know which side of the support divider he is on. If your by the SG or under the MEGA and you hear your opponent get the mega and then the shards you know which way he is going. Generally speaking the best locations for these items are in short hallways outside of the line of sight of major control areas.
T2 is a good example of another use for shards and plus fives. T2 only has two armors, both yellows. If it wasn't for the 10 extra shards in the RL/YA room, both YA rooms would be equally valuable to camp. But the presence of the shards makes this room more valuable, and the one most important to control.
Having a center point, or a room that is more valuable than any of the other rooms is very important for 1v1 game play. Imagine a level made up of six identical empty square rooms, each room has five teleporters one for each other room. In each room in the exact middle is a RL and a 50 health. Each room would be of exactly the same value and your chances of finding your opponent in any given room would be completely random. There would be absolutely no strategy to use to try and find your opponent with. In a normal 1v1 level, by contrast, there are areas with good items and areas with bad items. Aggressive play involves being in the areas with good items and defensive play involves being where the bad items are. Trying to predict when your opponent will be aggressive and when he will be defensive is one of the skills of 1v1 with the most depth and strategy. Good maps are maps that emphasize this skill.

T2 is a good simple example (although its simplicity is also its major fault). All the important items are around the teleporters in the two major rooms (red lines). The two long hallways that connect the two main rooms are almost entirely outside of the hearing range of the main rooms and have no important items in them (blue lines). Aggressive play involves going back and forth through the teleporters between the two main rooms collecting the armors, defensive play involves mostly staying in the hallways and sometimes entering the main rooms either by the RL or the LG.

There is no one way to place health, which style you choose seems largely a matter of personal preference, but its important to understand the effect of each. Most 1v1 maps have between 100-250 points of health total.
DM13 has almost 400 points of health. As result of having so much health, health denial is almost impossible in dm13 and all the fighting revolves around denying your opponent guns and armor.
In T4 the health is in only 2 places, both of which are dangerous, also there is little health total (175 points). As a result of this, is it common for controlling players to run around the level with a lot of armor and not a lot of health, its also more common for controlling players to use the mega just to get back to 100 health. Constantly forcing the controlling player to decide whether or not to leave his strong position just to get health can sometimes be fun and sometimes be annoying.
CPM1 has 4 distinct and strong (50 points each) health spots, one of these is along a major pathway and is usually not there when you really need it, the others are in out of the way places. An aggressive player in this map can predict that his opponent is going to run to one of these out of the way spots and make them pay.
CPM3 has three strong health spots which are along major pathways, the controlling player has little trouble collecting these but the down player will often miss out on picking these up.
DM6 has lots of little packets of health and one important 50 in the middle. Its easy for a player who covers a lot of ground to get one or two of the 25 healths, a severely hurt player will need to collect the 50 in the middle. As a result its usually worth the time to "health control" (hurt yourself on purpose) the 50 in the middle, just to deny it to your opponent.
T2 has less health than any other map, 125 points, of which 75 points are in the major rooms. As a result its not worth running to the defensive areas of this map when your hurt, instead its better to use the teleporters for your escape routes. The player who can collect the 50 health on the bridge during a fight usually wins.

Ammo is usually not important. Most maps don't need any SG, or rail ammo. RL and LG ammo is the most important kind of ammo. CPM1 and CPM3 have lots of RL ammo (and 2 RLs), which goes a long way towards encouraging rocket spamming. DM13 is a large map with only 2 important guns, the RL and LG, each of these guns have only 1 ammo box each and as a result these ammo boxes are unusually important. T4 also has RL ammo scarcity, making its one RL ammo box unusually valuable. Although RL ammo scarcity adds another dimension to the game, it also reduces another dimension, the spamming possibilities in a map.

Level designers can use shader files to remove footstep sounds from their level. Heres an example texture (a new type of clip brush) with no footsteps:

surfaceparm nosteps
qer_trans 0.40
surfaceparm nolightmap
surfaceparm nomarks
surfaceparm nodraw
surfaceparm nonsolid
//surfaceparm nolightmap //proto_addition 11/08/99
surfaceparm playerclip
surfaceparm noimpact

To my knowledge no one has ever set out a list of rules for where to put spawn points and where not to put them. Generally spawns should be out of plain sight of the major control areas and have their back to the wall. In T2 there are 7 spawn points, 6 of them are in the side halls near the entrance to one of the major rooms, the 7th is right by the YA in the YA/RL room. That seems to work well. Early control of the map in T4 almost entirely depends on who can sweep the RA and MEGA and to that extent the spawns are quite unbalanced. But otherwise they are placed in areas where there is a wide variety of escape routes. The spawns in DM6 and DM13 are very unbalanced, most of them are in out of the way safe places, but some of them are right near the major power points, which adds a lot of luck the frag count on these maps.

T4: Q3tourney4
T2: Q3tourney2
CPM1: originally released as Q3jdm8a, it was renamed to CPM1 when it was included in the promode map pack, this map was created by jude.
CPM3: originally released as Q3jdm9, it was renamed to CPM3 when it was included in the promode map pack, tthis map was also created by jude.
dm6: Q3dm6(tmp), this map was modified by Clan Abuse for tourney play,
dm13: Q3dm13(tmp), this map was also modified by Clan Abuse and is available in the same file, however the only changes were to remove the holdable health kit and the quad.
RL: Rocket Launcher
RG: Railgun
SG: Shotgun
GL: Grenade Launcher

Nicely done
Comment #1 by on 04:54, Thursday, 24 August 2000
I think I pretty much agree with you. :-)

I do have a lot of comments on the various sections of your article, and some other stuff too -- since you asked for `em, here's the dump. More later if anything else occurs to me. :-)

* Sound system -- Great info! You know, I believe that an author could deliberately broadcast item pickups (if they wanted to) by linking the items to speakers, like you mention later for doors. Haven't tried this yet though.

With any drastic change like this, a map author should of course mention it in the readme.

* Space levels -- Qualify that with "for 1-on-1" (just to be clear), and you'd improve that sentence a bit. Still not entirely true, depending on what you call a "space level", but close enough for government work I s'pose.

* Jump pads -- make sure that there is a "default destination" where players will arrive someplace interesting without using any air control. Test this to make sure it works no matter which direction you enter the jump pad from.

In maps without the rail, jump pads that throw you in a high arc can make for some fun fights and give people more options for picking a dest using air control. If you feel that you have to place a jump pad in a map that has a dominant rail, then give it a low arc instead.

* Platforms -- platforms are a sticky issue for map designers because there is no sound for them provided in the id paks. If you want to use platforms, you have to provide a sound in your map pk3 file. However the problem with that is that resources in pk3 files are NOT map-specific; they apply to all maps. If there are several map pk3 files installed on a client, all of which contain a platform sound, then the sound in the pk3 with the alphabetically last name will "win" and be the sound that is used for platforms in ALL MAPS for that client.

This is just an aesthetic issue, but realize that if you use platforms in your map, then the platform sound that you provide could end up being used for platforms in some other map, so don't make the sound too annoying or highly theme-specific. Also realize that _your_ plat sound might get overridden by somebody else's plat sound. In many cases you can use vertically-moving doors instead of plats, which you probably should do if that works out for you.

Also keep in mind that plats can be blocked by players. There's no way to make "crusher" plats like there is for doors. So if you use a plat, look out for the various tricky things that a player could do if they block its path.

* Stairs -- The id guys recommend making really tiny stairs, 8 units high I believe, but personally I think that looks odd and makes stairways take up way too much space. I'd make stairs at least 12 units high, even higher (up to 16) if in a grandly scaled place, like an Aztec temple or something.

If you're designing for CPM, the "damage through floors" will probably be disabled when people play your map. One minor but still noteworthy effect of this is that rocket fights on a stair become harder, especially for the person on the lower elevation, since all those little stair edges can now completely block splash damage.

* Teleporters -- I generally agree on the badness of reciprocal teleporters where the dests are right up against the teles. However that setup can serve one purpose, especially in CPM where running for your life can be more important. It discourages pursuit. When someone runs through a reciprocal teleporter, the person chasing them will often back off even if the pursue-ee doesn't have a good ambush weapon, because of the danger of telefragging. (Anyone who played a lot on DM4 or DM6 in Q1 is real familiar with this whole concept.)

So a reciprocal teleporter pair could be interesting. But to prevent someone being able to camp at it and ping-pong back and forth out of danger, make them vulnerable to someone who is at some other portion of the map (i.e. not chasing them). Classic example, from DM4: a person can position themselves to have rocket coverage of one teleporter dest and chuck grenades at the other dest.

* Grenade Launcher -- I agree that the author should have a VERY good reason to leave this out of a map. But do limit the ammo. Probably only 1 ammo pack is needed.

* Armors -- One CPM-specific comment; when a player has full RA, they would have to significantly damage themselves and downgrade their armor to pick up a YA (as a denial move). So mixing armor "flavors" in a CPM map can make it more difficult for the controlling player to deny all armors to his opponent.

* Powerups -- One of the bad things about item placement in Q3 is that there is no differentiation between FFA mode and tourney mode. Powerups can be fun in FFA; if an author wants his/her map to support FFA play, I think they should feel free to stick a powerup in there. CPM, OSP, q3comp, every comp mod under the sun will allow people to disable that powerup for 1-on-1 play, so that's not a big deal. Just don't design the 1-on-1 gameplay of the map with the powerup in mind.

* Symmetry -- Amen. Global map symmetry makes a map half as interesting. Certain areas can be symmetrical if you think that serves a purpose, but I see almost no reason to make an entire map symmetrical for 1-on-1, FFA, or team DM play.

* Shards -- Shards are a great resource for the player not in control. They allow him to survive a rail hit or near-rocket hit. Put them in the areas of the map that are not in primary combat areas, and off the main juice loop. If the other player is completely occupied running the armors, the down player can just hang out getting shards (if there are enough on the map). In that case, the controlling player will eventually have to come after them, or they'll be able to exceed his armor value. This can help break up the armor-relay style of play.

* Health -- Health placement probably needs a good amount of playtesting. You want to avoid the problem of the test version of T2, where a player could instantly fill back up to full health after winning a fight. If the losing player put up a great fight before going down, the winner should be hurting a bit afterward. But if the fight was pretty one-sided, the winner should be able to top off afterward with little problem. It's an interesting balancing act.

* Footsteps -- Just to be clear: it's the "surfaceparm nosteps" there that does the trick. All the rest of those parameters are doing unrelated stuff.

* Spawns -- Because of Q3's simplistic not-nearest respawn system, you don't want to put too many spawns in a map. Otherwise a spawn point could be selected as "not nearest", but be only a few feet away from the actual "nearest" spawn point, making the distinction moot. (What we really need of course is a smarter respawn system.)

Spawnpoints should also not be right on top of items/weapons. It's a tiny bit of fun for a player to be able to go pick up a good item, even if it's just a matter of moving a few feet. More importantly, if a player spawns right on top of an item, it may not be clear to them that they have received that item.

* Glossary -- Actually cpm3 was q3jdm10, not q3jdm9. I don't believe that the final version of q3jdm10 was ever released, except as cpm3.

**** General comments:

* Layout -- This is potentially a big topic, but the basics: a linear map is bad. You should almost always have a circular flow on the macro level in your map. Not that any path needs to be _shaped_ like a circle -- but a player should be able to move forward, making turns that are not too big, and eventually get back to their starting point, meanwhile having covered most interesting areas of the map, at least on the map's main level. You need at least one "crosscut" path too, because just a basic unmodified ring layout is a track meet, not a deathmatch.

It's a fuzzy rule, but it generally applies to every great, fun, successful DM map. Dead-ends can have their uses, if you treat them carefully, but generally "turning around and going back the other way" should not be a big part of the gameplay on the map.

That's horizontal layout. As for vertical elements, there's a lot of freedom to what "works" with those. But there does need to be some vertical element.

* Map proportions -- Make your architecture nicely "chunky" and reasonably spacious. I suppose an author that has a chance of making a good tourney map already has a good feel for this sort of thing, but anyway: tight spaces, bad. Thin fiddly brushwork, bad. You want architectural features that are big enough to affect combat, but still on the "human scale" so that they can be used in interesting ways. For example, a simple scattered field of player-high cubes could produce some interesting fights. If the cubes were 1/4 that size, or 4 times that size, their affect on combat would be much less interesting.

High ceilings are a great way of adding the feeling of space without actually mucking up the playing surface.

Low arches and door lintels are bad. Keep in mind that good players strafejump just about everywhere. If a door lintel is low enough for a player to hit their forehead on at the peak of a jump, it is too low.

* Cover and elevation -- Most walkways and platforms are just solid blocks, and any multi-elevation fight in that case involves one player peeking over the edge at the other. Maybe you should try different variations that create different combat situations. In the process, you can make the high player less dominant, or even try to switch the situation around and give the low player the upper hand. This is not something that all maps should do all the time -- high ground = good is a great general rule of thumb -- but mixing it up occasionally could be interesting.

For example:

- Make the walkway/plat a grate that can be seen through but not fired through (except for splash damage). This can actually make the higher player MORE vulnerable than the player below; an interesting reversal.

- Make the walkway/plat a grate that can be seen through and shot through. This gives the higher player more opportunity for shots, but it makes them more vulnerable to several weapons.

- Make the floor area a grate that can be shot through (another dominance reversal trick). No more splash damage against the low player.

- Keep an eye to place ceilings or walls that could be used as targets for a low player to splash-damage a higher one.

* Killzones -- Slime, lava, and deathfog should be used with care. Don't make it the central defining feature of your map. Also, slime and lava should usually be shallow. Having a danger area to avoid can make combat more interesting, but you want to be really careful about crossing the line over into frustration. Pay attention to what your playtesters say. A map author always thinks such things are much cooler than they actually are. :-)

As for trap mechanisms, either automatic or player-activated: these are almost never a good idea.

* Jump distances -- The jump to the MH on DM13, and the jump out of the tele on T2, are both examples of where a map author was trying for a certain gameplay effect (can't jump and get the MH from below; have to fall into the RL pit when coming through the tele), but since they "cut it close" on the measurements, their intentions were defeated. I realize that sometimes when this happens, a cool trick can be the result, and maybe the map's gameplay could even be improved. But there's also the chance of a player circumventing the map design in a way that really hurts the gameplay -- and as for framerate-dependent jumps, those are just all-around bad. I'd say that map authors should always be generous in their measurements; if you want someone to not be able to make a certain jump, then make it so that they would miss by 16 units, not by 2 units.

There will still be chances for trick jumps in your maps; some that players will find, and some that you can put there explicitly. But if you definitely do NOT want a player to be able to make a certain jump, then deny them with authority. :-)

* Trick jumps -- First I want to comment on the "clip brushes are bad" mentality. I think this has been taken a bit too far in some cases. OK, if you want to provide little horizontal surfaces that players can jump on to be tricky, that's cool. But vertical surfaces that can snag players as they move around the map (door posts, beams, etc.) ... clip `em! Such obstructions serve no purpose, they only frustrate people.

As for the intentional placement of trick jumps in the map, I think this is cool if they can be used as shortcuts, but not if they are the only way to get to some location or item. See my post about q3nbdm1 ( ) and the original version of wdw3team1 ( ) for more details about why.

* Aesthetics -- The way a map looks is of course a whole 'nother topic. Briefly: neglecting a map's looks "for the sake of gameplay" is a copout. Make a map that looks good and plays good. It is quite possible to do this while still keeping framerates reasonable and without making architecture that gets in people's way. If a map looks bad, people won't play it long enough to find out that it plays good. q3jdm8a is the common counterexample to this principle, but keep in mind that 1) most maps won't "strike it rich" with that amount of pimpage, and 2) q3jdm8a, although it could look better, doesn't look _bad_.

My post about q3nbdm1 also goes on a little about aesthetics. I had a really expansive ramble on that topic in a Q3World editing forum post but it seems to have expired away -- I can't even find it with a search now. Oh well.

One last note on that topic: Q3 neo-gothic is very passe for now. :-) Try out some other themes, especially with custom textures if you can hack that. If you really crave making a neo-gothic style map, just hang onto that impulse, I'm sure they'll eventually come back into fashion.

Other opinions...
Comment #2 by on 05:02, Thursday, 24 August 2000
Didn't some player (revelation?) do a writeup about Q3 1-on-1 maps on Truegamers, or XSReality? I can't find it now...

Also, hitting the Wayback Machine, you could check out this article: Can't say that I agree with all his philosophies about skills, but IIRC there was some interesting stuff in there about map layout and architecture.

One more thing...
Comment #3 by on 05:27, Thursday, 24 August 2000
Lighting. While it's all well and good to try to be "atmospheric", it's not fun in DM to have a hard time seeing your opponent. Make sure there is a sufficient light level in your map.

Also, if you have dark patches in your map that can make players hard to see, you only penalize anyone who is using lightmap (since players using vertex lighting won't be bothered by the shadows).

And another thing..
Comment #4 by on 07:21, Thursday, 24 August 2000
hehe it's not Johnny Law this time, it's only me.

Hey JL, wld you mind if I re-package these comments as a sort of article and put them in the maps/views section? I am thinking of "republishing" twoAM's article there also, so as to keep the mapping analyses together in one place and your thoughts would also be most valuable to have there.

Congrats on writing a comment which is almost longer than the original article (not a shorty either) :)

I am a believer that all these sorts of principles are useful guides. I wonder (as an experiment) what wld happen if we collected together "20 golden rules" and then a mapper made a map, any map, which was designed only to conform to those 20 golden rules. Would the map be good or not?

Could a computer be programmed to 'design a map'?

Is there a danger of taking this route? IE that mappers could start to "follow the rules", thus stifling creativity? Cld we end up with "mapping by numbers" etc? I doubt it - the rules are too complex, and the interaction of elements has too many possible combinations.

How many mappers "work to principles and rules" anyway? I imagine a lot of them to work to "gut feel", using their own experiences (rules) and enjoy the creativity of trying new ideas out. I notice MrClean and Wiebo each have distinctive styles, that they tend to explore and develop with each new map.

Song-writers who know the theory of music do not necessarily write better songs - look at Paul McCartney - he couldn't read music and yet he could compose great songs. On the other hand, a lot of great guitarists had classical training.

If there was a mapping course, what would they teach about design? What goes through the mind of a mapper?

package away...
Comment #5 by on 07:59, Thursday, 24 August 2000
Probably best if it stays clear that it's commentary for twoAM's article tho.

BTW it just seems long because of all the whitespace. :-)


I think, as in any creative endeavor, you need to know the generally accepted rules and why they exist before you can successfully break them -- and the more rules you break, the trickier it is to come out with a successful product.

There was a phase in Q1 mapping where many mappers did just "follow the rules" laid out by example in headshot and Danimal's DaPak maps, and so you saw an upsurge in atrium-based deathmatch maps of a certain size, layout, and texturing style. Eventually that passed though. (And, actually, some of those DaPak-ish maps were pretty good.) I don't think there's any real "danger" in talking about good practice in map design -- even if there were some chance of all mappers going along with it in sheeplike submission, which there isn't. :-)

Even by following all the "rules" that people lay out for mapping, you could still put together a very creative, unique, and fun map. Most of these rules are pretty soft or vague. There's nothing here that is an analogue to real music theory -- more along the lines of rules-of-thumb like "parallel fifths suck".

But if a mapper genuinely understands why it is generally a good idea to do A, and he/she still thinks that in this particular instance they're going to do B instead, then fine. As long as the playtesters agree, I guess. :-)

As for computers making maps? Heck, I'm sure that computer-assisted map design will be a required part of game design eventually as game spaces get more and more complex, and people will look back on the days of placing and texturing each brush of a stairway by hand in the same way that programmers now look back on someone writing an entire word processor in assembly code. But I truly doubt that we'll see an automated whole-map generator that is anything more than a curiosity any time soon.

A blurb about including/excluding weapons...
Comment #6 by on 08:09, Thursday, 24 August 2000
Generally speaking, in order to include a weapon, it must be at least somewhat useful, yet it should not dominate the level. It would have been odd to see the Grenade Launcher in T4, because it would have been useless, and on the other hand, space maps are generally bad for 1v1 because the Railgun is too powerful.

Even of weapons that fit the description above, it may desirable to exclude one or two of these weapons to create a special effect. Notice that T2 doesn't include the GL or PG. Without these weapons, the level becomes an offensive struggle. q3dm10 is what it is (for FFA anyway) because of its weapon selection (RL+LG+PG).

The trap I feel PM-type levels get into is using (nearly) the same weapon set. The GL and RL are expected weapons, and the SG is always included to provide the low player a weapon. Furthermore, the LG plays a central role in Pro Mode, since it is a weapon that has no replacement in combos. The PG would be a lousy substitute, so the LG is destined to appear in all PM levels. Level designers can determine the fate of the RG and PG only.

If you don't have rules you can't break 'em
Comment #7 by on 11:40, Thursday, 24 August 2000
... and so you can't do the sort of things that make people say "that shouldn't work!". Of course just cos that works in music doesn't mean it'll work in mapping. Then again - someone might bring out a really good space map one day - maybe :)

Also there should be plenty of 'hard' moves in maps that can give an advantage if you can pull them off. I'm not talking framerate dependent stuff which just sucks but the kind of thing found in maps like ztn2dm3 or tig_den. On the former you have the jump across to the RA - at first you'll tend to miss it quite regularly but after a while it becomes second nature - but there's always a risk of falling and suddenly being vulnerable! On the latter, one example is the route up to the LG via the blocks. After a while it's not that difficult to pull off but if you do slip up then you're right out in the open and vulnerable. It's those sort of things that make maps more interesting and place the emphasis on more than just aiming skill.

(And i know i could have quoted other maps such as q3jdm8a ro whatever - but i just kinda like those two i mention :)

Comment #8 by on 14:08, Thursday, 24 August 2000
Great article TwoAM and thanks Johnny for adding more comments. I really enjoyed reading this and I will surely take things into account when making something new.
But... Rules and guidelines are ok, but to add to that I will always use my gut feeling and go with something that may not fit into the rules or guidelines, just to make it interesting. Sometimes I deliberatly do not add or place things in a certain way or posistion just to see what my testers say about it. And you know what? Sometimes it stays in =] loveley. So my advice, take this stuff into account but don't be afraid to experiment.

20 golden rules of mapping...
Comment #9 by on 20:43, Thursday, 24 August 2000
RE: 20 golden rules of mapping

Theres so many critical areas that I didn't talk about in my article still... for example how many atriums should a map have? cpm1, cpm3, heavy duty each have 2 atriums. So a good duel map must have 2 atriums? well no, not really..

The entire critical core of the map, what the actual layout should be and what style it should follow I left almost entirely untouched. And I think there are many different kinds of styles (q1dm6 style, dm4 style, q3dm6 style, cpm1 style, etc....) , and probably quite a few still unexplored...

like look at wiebo's tdm map... I never said in my article that you shouldn't make a room with only one exit and entrance and put a very important item in it... but a lot of people would probably say that theres a rule that says that. but hey, they are wrong, cause wiebo did it and it works...

I guess I'm just trying to say that I made a list of rules where I could, but that theres a lot of territory where
there are no rules and thats a good thing..

"it may desirable to exclude one or two of these weapons to create a special effect. Notice that T2 doesn't include the GL or PG. Without these weapons, the level becomes an offensive struggle."

I just want to say that this is a really good point that I didn't understand the first 5 times I read it.

Purposefully excluding certain weapons that would have been useful when making a map seems initially a really big gimmic idea. but if done right it could have an positive effect on the character of the of the map.

on the other hand, if i made q3tourney2 tommorow and released it (and id software hadn't already made it), no one would play it. tig of LvL would slam it, the CPM team would say its boring... but if I made q3tourney2 and included all the guns, well it might be a bit more popular... (but not necessarily better)

as for guns and CPM maps... as a level designer I felt pretty limited about what guns to include even before I knew what CPM was.. after all there are only 6 of them (I dont count the BFG, gauntlet, MG or grapple). it would have been really nice of id to include some more weak weapons...

Comment #10 by on 00:36, Friday, 25 August 2000
"No one likes space levels other than newbies and level designers. Don't even bother."
"Symmetrical architecture is boring at best"
well.....don't you think you're going a bit fast there?lol
i passed entire monthes fighting in dm4, dm6, and other levels....and i don't consider myself as a newbie.
really it hurts me to read these kinds of sentences.ouch!
i understand that "gameplay" can be considered by many, as a new science. but be careful not to become narrow-minded concerning the fun any map can give you.even space maps.
space maps cannot be considered as an entire bag of crap, and symmetrical maps can give different fun than non symmetrical maps.

level design is not only a matter of accepting all the traditionnal gameplay and's also a matter of imagination.wether it is for 1v1 or for deathmatch. maybe some designers need these general rules and weapon balance you're talking about,but once again, it is in no way an obligation(i'm not talking about making a non sense map either of course).
for exemple.....jumppads are according to me a true gameplay improvement that you find in q3(if they are correctly used). with those, verticality in maps become much more interesting. well if ID had kept the same mentality that they had in q2 or q1 concerning the gameplay, they would have never imagined these items.
criticizing the entire world of "space maps" is a true mistake. even more for an experienced mapper. you may not like it, but it remains a smart way to create new gameplay.
just don't include the rg and you'll be able to have very balanced maps.listen to your creativity instead of given rules....that's what i say.

for the rest of your article i agree with nearly all of it. it is logical and useful.
concerning func_plats, id must have removed them in q3 because of bot navigation problems.

no 20 golden rules of mapping
Comment #11 by on 00:43, Friday, 25 August 2000
20 golden rules of mapping : bad idea.
it will kill all the smart ideas and concepts any level designers could have one day in future levels.

A slight mistake in original post
Comment #12 by on 01:54, Friday, 25 August 2000
Just thought I'd better point out a mistake in the original post to new readers.

Quote: "In T4 the health is in only 2 places, both of which are dangerous, also there is little health total (175 points)."

This isn't true, there's the megahealth (100), + two 25's on the ground floor in the centre room, plus 2 25's on the second floor in the centre room (the one with the launchpad), plus two 25's next to the rocket launcher, + 5(?) lots of 5 health underneath the RL ledge = 275 total health, and at least 3 distinct health areas. Did you forget about the centre room?

Apart from that this was a very good and elightening read twoAM! Thanks a lot :)

A slight mistake in original post
Comment #13 by on 02:04, Friday, 25 August 2000
Just thought I'd better point out a mistake in the original post to new readers.

Quote: "In T4 the health is in only 2 places, both of which are dangerous, also there is little health total (175 points)."

This isn't true, there's the megahealth (100), + two 25's on the ground floor in the centre room, plus 2 25's on the second floor in the centre room (the one with the launchpad), plus two 25's next to the rocket launcher, + 5(?) lots of 5 health underneath the RL ledge = 275 total health, and at least 3 distinct health areas. Did you forget about the centre room?

Apart from that this was a very good and elightening read twoAM! Thanks a lot :)

from memory...
Comment #14 by on 05:08, Friday, 25 August 2000
CPM3 is q3jdm10 not q3jdm9, isnt it?
q3jdm9 is a 2on2 map.

the forge
Comment #15 by on 15:22, Friday, 25 August 2000
Didnt the forge ( ) had some kind of 'mapping university' at the worldcraft 1.x time ?
but that was more about mapping techniques, getting e neat map, rather than those (really precious & helpfull) hints

Great thread
Comment #16 by on 16:16, Friday, 25 August 2000
Wow - excellent article and commentary - really opened my mind as to what's involved in map making. hehe maybe it'll help me learn maps faster.
I agree with pretty much everything said here - and there seems to be a good consensus too.
Good example of how circular flow is key - evolution (ra3map1)

I really miss elevators - scenario - q2dm1 - you're down by 1 frag - 10 seconds left - you're stacked - opponent is weak after an intense level wide chase. He makes it to main elevator - you put on your nike airs and strafe run towards the elevator chaingun in hand. You're one and half hops away from the elevator and he triggers it and starts his upward decent.... do u make it in with him and tie up the game? - or does he escape to win the game?!
That's just one of the many intense scenarios elevators have to offer - there is incredible adrenaline on both parties in the above mentioned one - also with a rocket launcher you could use a lot of finesse to fire a rocket up into the elevator just as the platform dissapeared up the shaft - it was a threading the needle move and fun as hell to do. I'm sure quake3 could harness these same elements if elevators were introduced.

Hoony - some interesting thoughts - i believe in the power of metaphor and analogy, and the music one is a good one imo.

Even tho i know very little about map making, I'd venture to say that ideally - it'd be important to understand the true nature of 1on1 dm and build the map with quite a bit of instinct. Afterall - it is a form of art.
As for computer generated maps, it's a pretty deep question. But it brings up something interesting - the nature of an art form - sure a map is a work of art - but it has function and is a medium where humans interact in. Can the "art" part and the "functioning medium" part be separated? I have a gut feeling that the answer is no, at least when it comes to truly good maps, and because of this, I'd lean towards the viewpoint that a computer would probably not be able to create a "good" map. Then again, with some good neural net computing, and some human feedback, some interesting things could happen - I could definitely see computers being used to anaylze various aspects of a map - from mundane things like health balance spread to more interesting things like quantifying vantage (measuring the effectiveness of the rl by calculating the angles of attack it has and things like splash zones etc)

Also - i've always wondered - is the thru floor splash damage a bug - or an intended feature?

Splash damage thru solids
Comment #17 by on 17:38, Friday, 25 August 2000
Ah, one of the great mysteries of life. :-P

Since I'm sure id knew about it, and it was apparently easy to change, I guess it's not a bug.

My personal theory about why they left it in is to avoid wackiness with splash damage on stairs (like in my comments above) and around small cracks or edges in the architecture. But I'm just guessing.

splash damage
Comment #18 by on 21:12, Friday, 25 August 2000
Johnny, it might have something to do with also being able to hit someone through a fence (solid brush)
... or maybe not. =]

Comment #19 by on 23:06, Friday, 25 August 2000
> it might have something to do with also being able to
> hit someone through a fence

True, true.

I wonder though if you couldn't make a special case for that, like: zero-width surfaces are allowed to transmit splash damage, others aren't.

Comment #20 by on 15:20, Saturday, 26 August 2000
> I wonder though if you couldn't make a special case for that, like: zero-width surfaces are allowed to transmit splash damage, others aren't

Yeah, I think that if id intended to leave it in then they should have made an extra content or surface property that could be defined in shaders. Ahh well, it may not be realistic but I do like it :)

Splash damage
Comment #21 by on 06:11, Monday, 28 August 2000
splash damage thru solids is variable
it has a console command, its not a bug

plus it adds another tactical element to the gameplay
you can do a lot of damage on t4 especially, from underneath, if you couldnt get them thru those walkways the guy on top would have a huge advantage, but with this feature he cant step back out of view without expecting something hot and firey to come through the floor and singe his hairy scrotum

zero width
Comment #22 by on 12:03, Monday, 28 August 2000
> I wonder though if you couldn't make a special case for that, like: zero-width surfaces are allowed to transmit splash damage, others aren't

I use zero width path meshes for grates,etc. If you use that and turn the console variable off you might have the same effect eg. no damage through walls but through grates or fences (I don't know for sure, i haven't tested it yet)

But aren't we getting off-topic here? :)
I just read some comments about the MH having to be (or being favourable) in a tourney map... hm, i don't quite agree on that as you can create just as favourable elements with other weaponry/ammo/health. I noticed that Mr.Clean's Prophecy has one though... any thoughts on that, design wise?

mega health
Comment #23 by on 17:48, Monday, 28 August 2000
The important thing about the mega health is that it increases the maxium power a player could have, instead of being 100/200 or 100/150, a player can be 200/200 or 200/150.

I think its a fine addition to a 1v1 map... make sure you count it as a armor though.. (unlike some earlier posts here)

tourney or 1v1 ?
Comment #24 by on 11:42, Tuesday, 29 August 2000
Are we talking about tourney maps here or 1v1 maps? :)
The general idea I am getting is that tourney maps should NOT include powerups, so do you not count MH as a powerup because (like you said) you treat is as armour? Which means you compensate for the MH in another are of the map i guess. Am i confused? yes, a bit

More angled surfaces
Comment #25 by on 02:09, Wednesday, 30 August 2000
Angled surfaces such as the ones by the door into the main area in q3dm12 are great and should be used more, I do not mean ramps. If possible create surfaces which cannot be stood upon but can be jumped off, imagine rocketjump to circlejump combos. Angled surfaces that can be walked on would also be good for the height advantage role reversal. More of the less used tricks like jumppad plasma climbs or double rocket jumps. Sorry if this is too off topic. JL's problem with players blocking pads is easily sorted with a bit of lava/slime.

Comment #26 by on 17:27, Wednesday, 30 August 2000
whats the difference between 1v1 and tourney for you, wiebo?

re: ?
Comment #27 by on 13:43, Friday, 01 September 2000
Nothing actually, but the problem with VQ3 is that the game entities make no distinction between tourney and FFA, so it is not possible to make a level that has different items for FFA or 1v1 (tourney) which is a shame really. I find myself wondering about that and i guess it came out because in this thread tourney and 1v1 are both mentioned.

Add A Comment

Read our Disclaimer. Quake, Quake II, Quake ]|[ and the stylized "Q" are trademarks of id Software
All trademarks used are properties of their respective owners
2000 -