My names Jim and I've been building maps for FPS games for about 10 years.
Currently I'm mapping for Left 4 Dead. Why only the first game?

Well mainly because my hardware isn't up to scratch for the second one and also there's still a huge player base who are in the same position and still lap up new Left 4 Dead content.

This blog is my effort to document the mapping process in hopes of educating others who are starting out or looking for some inspiration.

So where to begin?
I could tell you my history of mapping or what inspires me ( rather arrogantly expecting you to be interested) but I wont . I'll start with the single basic key to everything involved in mapping....

An idea!

If you're reading this I'll assume you have an interest in mapping, which means at some point you've played a game, been inspired by the content and felt a need to remake that content into something you'd like to see or experience.

Here's where your first challenge appears and it's fundamental to the whole ethos of mapping.

Who are you mapping for?

If you are mapping for yourself, this can be a bit of a non-starter I find. The fact that you've slaved over the editor and put together each element of the map and built it up from scratch, tends to leave you in the rather akward position of knowing what will occur round every corner. It's no fun because there no surprises. You may be able to build in a little randomness and certainly the A.I. of enemies in games like Half Life 2 help a great deal but at the end of the day it's just not going to deliver the kind of player experience you'll get from unknown content made by someone else. Ces la vie my friends... tough titties!

So having dived into some tutorials you discover the active forums and suddenly there's a whole bunch of folks whom you can map for. Ahh but they're a fickle bunch and highly critical, expecting nothing less than professional game quality content from a lone mapper with big ideas.

It's the eternal challenge of mapping. Are you happy to create content for yourself and live with no surprises or suffer the slings and arrows of you fellow gamers?
I'd suggest there's a middle ground. You can create fun, playable maps that others will enjoy but you wanted to create. You may have to make some sacrifices along the way but always remember... its the IDEA that will see you through. Never compromise the original idea to please your audience!

Here's a little secret I've learned on the way.

People don't know what they want until it's given to them!

It's the one universal truth in life. People, for the most part, cannot concieve of an idea that is outside their history of experience. Go read, The Black Swan by Nassim Nicholas Taleb and or check out the overview here... you'll get the point.

Let's take, the still amazingly playable, Half Life 1 as an example.
It used a first person narrative the entire way through. This is still pretty unique to the Half Life games actually. No cut scenes, at all. Prior to the point of release gamers we're used to being ripped out of their own bodies, sailed around in a free floating camera to be told the story of what they were about to play through. We were spoon fed the story and our perspective was jammed into each story scene at just the right angles.

Not so with Half Life.
Valve decided they would run the narrative around you as you played. If you were looking in the wrong direction or not paying attention, tough luck buddy. It was a wonderful idea and allowed the player to never be broken from their own reality. As a result the experience was so involving that you're every nerve was shredded by the end of a playing session.

Before Half Life's release many gamers would have been appalled at the idea of only seeing important content if you cared to look. Any surveyed customer focus group would have given a massive negative response to this.

Half Life was voted PC game of the year and set the standard for FPS games for the following 10 years. I still don't think it's been beaten even by Half Life 2 as a player experience.

So make your map!
Ram it down their throats!

To hell with the focus groups and player feedback (within reason!).

If the idea is original and fun you'll always come out a winner and who knows... you might even change some perceptions while you're at it!
One thing that appears to be a bit of a struggle for new mappers is the concept of the game world being mostly just an illusion. This is a bit of a concept jump for some people.

Some folks are happy to be absorbed into the game world and never question how it's put together. So when it comes to mapping for the first time and they want to build a cliff edge with a horrific drop they build a 2000 foot cliff actually within the map.
Its an easy mistake to make.

That's another thing us mappers have spoiled for ourselves. That was amazing... how did they do that? We look for the edges of the map so to speak so that the reality intended for us to enjoy never really kicks in unless its totally engrossing.

I don't remember looking for the map edges when playing Theif 2. This was mainly because I was hiding in a cupboard praying the mechanical creature with the creepy arsed voice didn't find me. (If u havnt played theif 2: the metal age then you're a disgrace to the gamer name, go play it, love it and buy new underwear when your finished).

Theif 2: Amazingly pant fillingly scary but equally funny as hell

So... the bottom line is, the better at illusion you become, the more fantastical maps and amazing game events you will be able to create.

3D skyboxes are a godsend for this kind of work. You can create a world beyond the edge of your map that has a realistic look and feel to it and can also include lights, fire, fog... pretty much anything you need.

That 2000 ft cliff we mentioned earlier becomes far easier to create with a 3d skybox.
Or how about adding a huge floating space craft hanging over your map? Done in two minutes... well maybe three if you add wing mirrors.

The most useful tool in you're arsenal of illusion is sound. What the ear hears the brain forms an image of.

Take this scenario as an example. The player walks up to a set of lift doors and presses the call button. They hear an electrical explosion above them, a wrenching sound of scaping metal and a screeching sound heading down towards them. A huge bang then occurs behind the lift doors and their point of view shakes.
Then you hear the familiar ding and the lift doors slide open revealing a twisted mass of mashed metal and debis.

In my map the lift didn't go anywhere. The mashed up lift was always sitting behind those doors but through a combination of sound and effects the illusion is that the lift plummeted down the shaft. Sweet!

Sound is a fantastic way to flesh out the world because almost subconsiously your brain fills in the rest of the unseen world.

A map without sound will always feel empty regardless of the detailed mapping work.

Just remember that when it comes to the physical world you create, what the player can't see, you don't need to build and what the player hears becomes the reality of what's around the unseen corner.

Pretty soon we'll all be a bunch of budding David Copperfields! (Even if he is a bit of a twat!).
Anyone whose watched a good horror movie will tell you the best scares are the ones that are timed to perfection.

One of my faves occurs in Jaws. As Richard Dreyfuss's Hooper dives to inspect the wrecked hull of a fishing boat he finds a huge tooth stuck in the hull. As the audience is intreaged by the mystery and wondering if he's going to be gobbled up at any second they don't see the manky de-occulard head of Ben Gardner coming as is looms out of the darkness for a perfect scare.

Its a brilliant moment and one that terrified me as a kid. It's the set up that makes it work so well. The audience know the scares are coming but they're taken completely off guard as the timing is just slightly off. The shot lingers on the tooth a little too long and our brains are programmed to think that the scene is drawing to an end.

In most game environments it's fairly easy to control the timing of events. The mapper has control of when and where bad guys spawn and has full control over the environment where the events will take place.

Left 4 Dead's director makes this a little more tricky. Infected spawn at the directors command and while threat markers can denote areas where witches and tanks spawn its hard to be precise. This is where some cheeky forced spawning can come in handy.

It's possible to spawn any special infected including witches and tanks via a trigger. Here's the link if you want to know how.

So utilising this forced spawning I'm striving to create a full on perfectly timed horror moment.

The team are at the end of a long corridor at a power box. As they hit the power switch the lights come on one at a time. First above them then a little further down the corridor and eventually lighting the whole length or the corridor.

It's a the timing of the lights coming on that will create the tension here. Accompanied with each light coming on will be a heavy metallic thud. Each light is timed about 2 seconds apart. So the player experiences an ominous build up.

Boom... boom... boom... boom... just before the last light comes on I force spawn a tank in the final darkened area. BOOM! TAAAANK!

I have this working in a proof of concept map at the moment and have been tinkering with the timing to get the feeling just right. The gaps between each light coming on are the key to the tension.

Oh and I'm aware that lights don't generally go boom when they come on but watch any movie with a dramatic reveal and I bet you $100 its accompanied with a big sound.

I'll post the test map here when I'm happy with it. I may even add a panic event just after the reveal to really pile on the pressure!

Cheers for now!
The story so far...
Looking for a mapping challenge I decided on creating a map centered around a US style projects tower block.

I tend to expand ideas as I go. My original idea was that the team would need to climb the tower block in order to reach a makeshift bridge across to another building which was outside the quarantine zone.

This seemed like a fairly fun idea and had the potential for lots of rooftop drama. Then it occured to me that we've all seen rooftop drama with No Mercy and that actually I'd just be re-building a crappier version. Time for a rethink!

What if we play on the towerblock theme a little? What if we begin the campaign with a botched helecopter pickup on the roof and the team then need to make their way down the towerblock. What if the reason the rescue was botched was due to some kind of terrifying event. What if there was an earthquake?

Now we've got a team of people on the roof of a very unstable building full of infected and they've got to get off the damn thing before it collapses.

That's drama I'd pay to see!

Not only that but we have the added beauty of being able to experience the same environment twice only under two fairly different scenarios! (Once on the way up, and again on the way down). Descending a collapsing building offers lots of opportunities for scary jumps, collapsing walls, floors and ceilings and also lots more places where infected can spawn and appear from.

So where to begin?

I designed the basic floor layout for my tower and created a modular shell of a floor complete with stairs and lift shaft. Once I had this I could just copy and paste in a new floor when i was ready to build.

I like modular mapping. By creating prefab sections of maps I can just copy and paste then position each new element and build up as I go. Don't like the way something looks? It's easy to delete just those small sections and build them again.
The downside is that you're brushcount will be higher. I doubt I'll be reaching the limit anytime soon though.

So once the basic structure of the first few floors were in place it was simply a matter of adding walls to create the rooms and areas on each floor.

From Aazell's Left 4 Dead Mapping Blog

Here's a shot of the current layout of the first few floors. You can see how each floor has the same basic shape in the upper part of the map!

The lower part of the map is an underground garage structure with a few side rooms. I may expand this out to a detour through the bottom of the lift shaft to activate the power for the building.

Inside or Out?
One key question is should the players be able to go outside and look up at the building. This obviously presents a challenge as I will have to create a complete outside face for the structure and also an outlying area with a huge box covered in skybox texture reaching to the top of the building. Not very engine friendly I'm afraid.

If I do want to do this I will have to split the tower in half as two maps. this is due to the skybox needing to change at higher levels. Here's the outside view from a higher balchony:

From Aazell's Left 4 Dead Mapping Blog

By using downward facing fog I'm able to mask the ground therefore saving me from having to map it

From Aazell's Left 4 Dead Mapping Blog

From Aazell's Left 4 Dead Mapping Blog

I may take the easy way out and have the players arrive in the building through the basement, then for the second map use my existing skybox. This is mainly a time saving exercise. By the time I'd finished the entire ground floor exterior people will have stopped playing Left 4 Dead 1 completely!

Natural Lighting

One of the concepts I'm trying to stick to is to only have lighting from the environment objects within the map.
These car headlights in the garage are a good example:

From Aazell's Left 4 Dead Mapping Blog

The players will activate the power for the building at some point within the basement though this will be sketchy at best throughout due to damaged electrics.

From Aazell's Left 4 Dead Mapping Blog

A partially lit reception area

Some of the lights will flicker. Some areas will be lit purely through sparking electrics with strobe effects (better stick a warning on the box).

Connecting floors
As you can see in the noclip screenshot above, the stairwell to the building is on the right hand side. This is realistic but it doesn't exactly create a complex playing area. Ideally I want the players to have to fight and move through each level then move up through each floor.
In order to do this we need to block the stairs off and create new and interesting ways for players to reach the floor above.

From Aazell's Left 4 Dead Mapping Blog

A broken floor with ladder reaching up to the floor above

The Way Down
The main characteristic of the return journey back down the building will be speed. I want the players to feel pressured to pile through the infected at lightning speed. This pressure will be increased by the feeling that the building is coming apart around them.
As a test map I re-built the lift shaft and added an action sequence where the players move across the roof of the lifts. As they jump onto the first lift it creaks loudly and drops a few feet. As the last player jumps from the lift it drops down the shaft and crashes to the bottom with a huge explosion.

From Aazell's Left 4 Dead Mapping Blog

Bill, Louis and Zoe jump for it as the lift finally gives out!

Wrapping up

I'm currently working on my basement panic event trying to make it as big and brutal as possible. I'll update soon on some more elements including the rather comical helicopter action rooftop scene and how the hell I can fit some tank battles into such a small space.

Till then....stay frosty!
Dum dum dum da da dum dum...

Horror based games are all about putting the player under pressure. No, scratch that... all games should be about putting the player under pressure. A game where the player doesn't feel the heat is no game at all in my humble opinion.

The trick is to ensure there's enough gaps in the pressure to make it an enjoyable experience.

One of my favorite moments from Half-Life 1 was the "surface tension" level with the Apache helecopter where the player has no choice but to run away. Sure the first time you come across it you may chance a few shots with your pistol but you soon get the message. RUN BITCH!

I love the reversal in circumstance. Here's our bold player churning his way through every foe the game throws at him and then suddenly you're in some serious trouble.

It leads to some real action moments. Who can forget the amazing "frying pan or fire" decision with the chopper closing in behind you and you make the decision to jump off the dam. Only then do you discover that the water below is infested with big bad fishes!

Trying to replicate this kind of experience in other games is really quite difficult. The huge array of enemies are all designed to be killed. Hell, in Left 4 Dead even the tank can be brought down with pistols if you're team is coordinated enough.

I'm tinkering with this idea at the moment. A great way to ramp up the pressure is to show the player what they are going to have to go through before they have to go through it.

For example, the players have to descend into a maze of underground corridors which are for some reason infected free. They traverse several difficult areas and finally reach their goal, a power switch needed to activate the door they came across 5 mins ago. It's about 3 levels down in a dank, dark utility room.

They then hit a switch and the lights come on which, in turn calls a panic event.
They'll then have to fight their way back through all the difficult areas in order to proceed.

The player knows what's coming, they know it's going to be a nightmare but push on anyway because the areas are familiar.

Watching the clown with a knife between it's teeth grinning up at you then entering the building below you has always been a recurring nightmare of mine!

That's horror. The knowledge that a terrifying foe is on it's way and there's nothing you can do but face it.

Another idea I've always wanted to try out but never been able to establish a decent map layout for is having a team deathmatch or CTF map where both teams spawn next to each other, seperated by glass or forcefield. Both teams get a good look at each other and know what's coming. Plus it's fun to make rude gestures through the window before a match kicks off!

Anyway... I digress...

Forewarned is forearmed. Give the player a hint of what's in store and you can be sure they'll be pooping themselves before it arrives. You can even play with this a little and have the arrival of the foe be something completely different to what was expected.

Now lets all go and make someone defecate!
In Left 4 Dead you play one of group of survivors of a zombie holocaust.

What the hell happened here!!!

That's our scenario and ya can't get away from it. So let's have a good ole think about what this means.

Well, first off it means that you won't probably won't encounter many other non-infected people on your travels. In the official game there was only one other character in the game who wasn't controlling an escape vehicle. The dude in the church.
What did he look like?

We'll never know because you never see him until he's fully turned (a neat bit of work exclusion from valve in my opinion).

So the only evidence of human existance is what they left behind. Every level in the official game had a story to tell though even without anyone there to tell it.

The graffiti on the walls is about as close as we got to a narrative and yet we still had a pretty good idea of what had happened and how. This was done through the placing of props and entities through the map. Think about the cars on the bridge in Crash Course. Here we see a scene of mass panic with cars and buses abandoned by their inhabitants. No one got out alive!

By imagining the events that came before we can create scenes of realistic utter devestation.

Take a look through the models available to us in the SDK and you'll find loads of items through which we can tell our story. A childs teddy, a downed traffic light, a wrecked family saloon car, an ambulance overturned on its side, blood on the floor, and corpses nearby. These elements together tell a story of just one inciddent. You'll need to think of quite a few in order to get your map looking great.

The hardest part is looking at an empty room and asking, "what happened here?".
Is the window broken?
Is there a hole in the roof?
Is the funiture overturned?
Did people try and survive in here?
Did they try to get out?

Are they the infected motherfu**ers trying to break down the door right now?


Heh heh!

I find it easiest to play it as a movie in my head.
Mom and dad put the kids in the closet and try to beat the infected over the head with pots and pans to stop them clawing their way through the door. They didn't see the hunter as he leered through the skylight above until it was too late!

The hunter finishes off mom as dad cowers in the corner frozen in horror. Dad looking over at the closet tries to think of any possible way to protect the children. In a final act of terror he hurls himself through the 4th storey window and plummets to his death figuring the hunter will follow him down. Mr hunter considers this option for a moment when he hears a faint muffled cry from the closet.
He approaches and sniffs around the door.


And so on...

I've just created a trashed room full of corpses that the player will see for two seconds.

Is it worth it?

In my mind it is. It expands my little map into a rich world with its own history and makes it a shed load easier to place props in a way that makes sense.

Comment with your own mini stories if you like! Would be great to hear some other ideas!
As you plan out your level you should be careful to ensure that the player is never in any doubt as to what they are supposed to be doing.

Its fine to have puzzles as long as the method to solving the puzzle is clear. You can make puzzles as tough as you like though.

A good example of bad signposting can be seen in the early Tomb Raider games. You entered an area only to discover there's no obvious way out. No hints are given as to the possible exit so its left up to the player to explore every nook and cranny trying every action possible. This leads to the player running along walls pressing the action button a million times yelling in frustration and eventually just deciding to drown the big titted bitch to teach the game makers a lesson. (I eventually discovered a hidden zip line at the highest point of that area which led to another area of equally frustrating design).

It should be clear to the player at all times what they are attempting to do and how it should be done. You can allow the player a little discovery every now and again just make sure the solution is pretty obvious when the find it.

So signposting can take many forms. A shaft of volumous light beaming down onto a key on a desk. Using a lit window to guide the player through the darkness. The safehouse symbols in L4D (these are useful but don't over do them. They don't have to be by every doorway).

Another thing the filthy Tomb Raider games were so guilty of was the instant player death syndrome.

The player walks down a dark corridor, suddenly they fall down a hole filled with spikes. YOU ARE DEAD appears on the screen.

The player is given no chance to avoid this death, no notice at all. This death isn't a punishment for doing something stupid. Its a game element that is supposed to make the game more fun when in fact it robs all fun from the experience.

A player death should only occur when they fail to complete a task correctly or just do something plain dumb. Jump off a cliff and instant player death should occur. Fail to run to the end of the tunnel before the train enters it... fine... you saw the damn thing coming its all your fault.

Instant player deaths without warning are evil and should be avoided at all costs.

The best way to ensure that players arnt going to get lost in your map is to get it playtested. What may seem obvious in your warped and twisted imagination may be completely confusing to someone else.
Don't explain anything to the testers. Don't tell them anything other than the general theme of the map and let them go into it blind. If a particular route or puzzle is confusing then more signposting is required.

Also, try to avoid over obvious dialog in the game. Its really rather patronising when you see "OMG look at this wall, how will we ever get past this?" "All walls have weak points! We should look for that!".

Try and be a little more subtle if possible please!