seir reacted to leplubodeslapin for an article, Source Lighting Technical Analysis: Part Two
This is the second part of a technical analysis about Source Lighting, if you haven’t read the first part yet, you can find it here.
Last time, we studied the lightmaps, how they are baked and how VRAD handles the light travel through space. We ended the part 1 with an explanation of what the Constant-Linear-Quadratic Falloff system is, with a website that allows you to play with these variables and see how lighting falloff reacts to them. We will now continue with basic examples of things you can do with these variables.
Examples of application
The simplest type of falloff is the 100% constant one. Whatever the distance is, the lighting has theoretically the same intensity. This is the kind of (non-)falloff used for the sun lighting, it is so far away from the map area, that light rays are supposed to be parallel and light keep its intensity. Constant falloff is also useful for fake lights, lights with a very low brightness but that are here to brighten up the area.
Another type of falloff is the 100% linear one. With this configuration, light seems to be a bit artificial: it loses its intensity but goes way further than the 100% quadratic falloff. It can be very useful on spots, the lighting is smooth and powerful. Here is an example:
This is the default configuration for any light entity in Hammer, following as we said before the classic Inverse-Square law (100% Quadratic Falloff). It is considered to be the most natural and realistic falloff configuration. The biggest issue is that it boosts the brightness so much on short distances, that you can easily obtain a big white spot. Here is an example, with a light distant of 16 units from a grey wall:
This can also happen with linear falloff but it is worse with quadratic. Simple solutions exist for that, the most common is not to use a light entity but a light_spot entity that is oriented to the opposite direction from the wall/ceiling the light is fixed to. You can make the opening angle of your light_spot wider, with the inner and outer angle parameters (by default the outer one is 45°, increase that to a value of 85° for example). If needed, you can also add a light with low brightness to light the ceiling/wall a bit.
50% & 0% FallOff
A second light falloff system exists, overriding the constant-linear-quadratic system if used. The concept is much simpler, you have to configure only 2 distances:
50 percent falloff distance: Distance at which light should fall off to 50% from its original intensity 0 percent fall off distance: Distance at which light should end. Well ... almost, it actually fall off to 1/256% from its original intensity, which is negligible. The good thing with this falloff system is that you can see the 2 spheres according to the 2 distances you have configured in Hammer. Just make sure to have this option activated:
An appropriate section for models lighting is needed, because it differs from brush lighting (but the falloff stays the same). In any current game engine, lightmaps can be used on models, a specific UV unwrap is even made specifically for lightmaps. But on Source Engine 1 (except for Team Fortress 2) you cannot use lightmaps on models.
The standard lighting method for models is named Per-Vertex Lighting. This time, light won’t be lighting faces but vertices, all of the model’s vertices. For each one of them, VRAD will compute a color and brightness to apply. Finally, Source Engine will make a gradient between the vertices, for each triangle. For example:
If we take a simple example of a sphere mesh with 2 different light entities next to it, we can see it working.
With this lighting method, models will therefore be integrated in the environment with an appropriate lighting. The good thing is that, if a part of the model is in a dark area, and another part is in a bright area, the situation will be handled properly. The only requirement for this is that the mesh must have a sufficient level of detail in it; if there is a big plane area without additional vertices on it, the lighting details could be insufficient.
Here is an example of a simple square mesh with few triangles on the left and a lot on the right. With the complex mesh, the lighting is better, but more expensive.
If you need a complex mesh for your lighting, you don’t want your model to be too expensive, you have to find a balance.
Two VRAD commands are needed to make the Per-Vertex Lighting work:
StaticPropLighting StaticPropPolys You have to add them here. You can find more information here.
Another system exists, that is much cheaper and simpler. Instead of focusing on the lighting of all the vertices, the engine will only deal with the model’s origin. The result obtained in-game will be displayed on the whole model, using only what has been computed at the model’s origin location. This can be an issue if the model is big or supposed to be present in an area with lots of contrast in lighting. The best example for that is at the beginning of Half-Life 2 with trains entering and exiting tunnels. We can see the issue: the model is illuminated at the beginning, but when it enters the tunnel it suddenly turns dark. And this moment is when the train’s origin gets in the shadow.
This cheap lighting method will replace the per-vertex lighting for 3 types of models:
For prop_dynamic or any kind of dynamic models used in the game (NPCs, weapon models in hand, any animated models...) For prop_physics For ANY MODEL USING A NORMAL MAP (vertex lighting causes issues with normal maps apparently), EVEN IF USED AS A PROP_STATIC
The big problem with these models is their integration in the map, they won’t show any shadow and their lighting will be very flat and boring (because it’s the same used for the whole model). But hopefully there are 2 good things with this cheap lighting method.
First, the orientation from which comes light is taken into account, if blue light comes from one direction, therefore all the faces oriented toward this direction will be colored in blue. And if you have different lighting colorations/intensities coming from different sides of your model, they should appear in game.
Here is an example of a train model using a normal map with 2 lights on both side. If you look closely, you’ll see some blue lighting on the left, on faces that are supposed to be in the shadow of the blue light but are oriented toward the blue light.
The second good thing is that there is still some kind of dynamic per-vertex lighting, but much simpler: it only works with light and light_spot entities (NOT with light_environment), and it just adds some light to the prop, it cannot cast any shadow (it only takes into account dynamically the distance between the light and the vertex). If we use again the high-poly plane mesh we had before as a prop_dynamic, being parented to a func_rotating that ... rotates. Light is dynamically lighting the vertices of the props. There is a limit of 3 dynamic lights per prop, it can’t handle more at the same time.
And if you add a normal-map in your model’s texture, this cheap dynamic lighting works on it:
Projected texture and Cascaded Shadows
Few words to finish the study with dynamic lighting. Projected textures is a technology that appeared with Half-Life 2: Episode Two in 2007, it consists of a point-entity projecting a texture in the chosen direction, with a chosen opening angle (fov). The texture is projected with emissive properties (it can only increase the brightness, not lowering it) and it can generate shadows or not. The great thing with this technology is that it’s fully dynamic, the env_projectedtexture can move and/or aim at moving targets. This technology is used for example on flashlights in Source games. But as usual, there is also a drawback: most of the time you can only use only 1 projected texture at a time, modders can change this value quite easily but on Valve games it is always locked on 1.
The cascaded shadows system is only used on CS:GO. The concept is quite similar from a projected texture but it doesn’t increase the brightness, it only adds finer shadows. It is used for environment lighting, using much smaller luxels than for the lightmaps and it is fully dynamic. It starts from the tools/toolsskybox textures of the map and cast shadows if it meets any obstacle. Shadows from the lightmap are most of the time low resolution and the transition between a bright and a dark area is blurry and wide. Therefore, the cascaded shadow will be able to draw a clear shadow around the one from the lightmaps.
When an object is too small to get a shadow in the lightmap, it will be visible thanks to the cascaded shadows. There are 3 levels of detail for cascaded shadows on Counter-Strike, you can configure the max distance at which the cascaded shadows will work in the env_cascade_light entity at the parameter Max Shadow Distance (by default it’s 400 units). The levels of detail will be distributed within this range, for example:
Since cascaded shadows and projected textures share some technology, you can’t use them both at the same time.
I really hope you have found this article interesting and learned at least few things from it. I believe most of these informations are not the easiest to find and it’s always good to know how your tools work, to understand their behavior. Source Engine 1 is old and its technologies might not be used anymore in the future, more powerful and credible technologies are released frequently but it’s always good to know your classics, right?
I would like to thank Thrik and ’RZL for supporting me to write this article, and long live the Core!
// Written by Sylvain "Leplubodeslapin" Menguy
Additional commands for fun
Mat_luxels 1 // Allows you to see the lightmaps grids Mat_fullbright 1 // Disables all the lighting (= fullbright). On CS:GO, cascaded shadows stay and you should delete them as well (cf next command) Ent_fire env_cascade_light kill // KILL WITH FIRE the cascade shadows entity Mat_drawgray 1 // Replace all the textures with a monochrome grey texture, useful to work on your lighting Mat_fullbright 2 // Alternative to Mat_drawgray 1 Bonus:
Mat_showlowresimage 1 // Minecraft mode
seir reacted to Furyo for an article, Level Design in The Last of Us: Part One
This is the first article in a three-part series. Part One / Part Two / Part Three
Intro Level 1st scene In typical movie fashion, the game starts with an exposition scene which establishes the bond between Joel and his daughter Sarah. Here the watch plays a type of backward MacGuffin (http://en.wikipedia.org/wiki/MacGuffin), which movie fans will be familiar with. A term made popular by Alfred Hitchcock, this initial narrative element will keep showing up in multiple scenes in the game to move the scenario forward and link back to the initial bond Joel had with his daughter. Many other items in this game, which we’ll see exposed in Sarah’s room play the same role. It’s important to note that this game doesn’t use forward MacGuffins, instead relying on the early experience players have with Joel in the intro levels to help them relate to Joel in the later parts of the game, when he faces adversity from the other main characters in the game. This type of catalyst is sometimes linked to gameplay in some games, but not here. For instance, in a quest, looking to get somewhere or obtain something but never managing to. In BioShock Infinite, Elizabeth keeps trying to go to Paris, but never quite goes there in the main game.
Gameplay Sarah goes to bed, only to be woken up by her Uncle Tommy’s phone call. The initial frame of Sarah getting up is a textbook example of player exposition. Using the mirror on the wall, Naughty Dog adds depth to the room, and presents Sarah as a playable character from both front and back. The color of the lamp shade also leaves nothing to chance. Using a warm color near her beloved items reinforces the comfort of her nest (bed) and contrasts heavily with the cold blue of the night in the mirror. These items on the wall, much like the watch earlier, play a narrative role in the rest of the game as tools to drive the scenario forward between Ellie and Joel.
Sound wise, Naughty Dog made a classic choice not to add any music and simply build tension with environment sounds in the background. It plays great and helps focus the attention on the initial reveals. In the room, the placement of the birthday card ties the opening scene with this scene the following morning and introduces the “triangle” button as the “hand interaction” one in the game (pick up item, open door, etc).
At this point in the game, the player still does not quite know what to do, despite having played for a few minutes already. This guidance is given very clearly in a single word, right outside Sarah’s bedroom, as she yells “Dad?”. Perfect example of narrative design coupled with level design, that tells the player his/her immediate goal, and invites him/her to check out every single door, increasing the chances for maximum exposition of every theme in this intro level.
With the goal now firmly established, the placement of the door straight ahead, too obvious, makes that room a natural space for continued exploration of the game’s themes, as opposed to a significant room. It comes as no surprise to find inside that room just one piece of content. The placement of the only light source in this room, obvious as it is, reminds us that lighting is one of the most straightforward tools to guide players.
Joel’s bedroom After leaving this bathroom, players are naturally directed towards Joel’s bedroom thanks to the window, extra light placed in front of it, and the barely open bedroom door. The sound of the TV is an ideal guidance tool, suggested instead of shown. This bedroom also offers the chance to see the gameplay loop closed for the first time. The Last of Us’ gameplay loop is always a variant on four themes: Exploration, Tension, Challenge, Reward, Return to Exploration. With the initial exploration started in Sarah’s room and the tension of having to find her dad, this room introduces the simplest of challenges – opening the door and following the gameplay instructions (L3) – which ends with an immediate reward of seeing the explosion in the distance. Finally, Sarah’s “Daaad?” closes the loop and makes players return to their exploration phase. Gameplay loops can express themselves over varying lengths. Second to second, minute to minute, hour to hour.
Returning to the hallway, a few classic LD rules can be seen in the same frame. The placement of the door allows players to face the staircase in the right direction, and the placement of the lamp at the bottom automatically invites players to go down. Notice its actual placement. While most times placing a light source away from a player’s immediate field of view increases its attraction, this one is placed to the right in plain sight of the player.
Once at the bottom of the stairs, the window introduces a new narrative element, in the form of police cars. Notice that the cars are driving in the same direction as the player needs to move. This is yet another good tool to guide players subconsciously. If the lamp at the bottom of the stairs had been placed to the left, and away from the player, he/she would have likely turned the camera towards it and away from the windows, negating the introduction of the police cars. That’s why this light source was better left in the player’s field of view.
Players naturally progress and now face this door to the garden, this time with a light source hidden beyond the playable space which naturally increases the mystery and tension in the scene. Just like upstairs in Joel’s bedroom with the TV, the dog barking here increases tension and attracts players forward.
Once players reach the window, Naughty Dog doesn’t fall for the cliché of having an infected appear in the garden, and plays with players expectations instead interrupting the sequence with Joel’s phone. Doing so, “L3” is introduced again this time on the player’s critical path to make sure this mechanic is understood by all players.
The placement of the phone to the right of the door is also no accident. It makes players move further away and to the other side of the living room from the office, where players are now supposed to go. This forces players to go by the garden window a second time and the use of the phone has forced the player’s immediate attention of the window to be erased. This perfect set up was necessary to create an element of surprise for the second traversal of the living room.
Narrative scene and transition A sign of truly great narrative games is the justification of every single camera takeover by placing the played character in a situation that allows for a smooth transition. Here, that’s the only reason this half open door is placed here. Since it’s already half open, like Joel’s bedroom door, there is no “triangle” interaction required. The goal of this door is to justify the exact position of Sarah in her animated entrance into the room, therefore the position of the camera, to allow for a smooth take over for the rest of the cinematic and the entrance and reveal of Joel, the very first in the game.
While the entire scene could have played out inside Joel’s office, Naughty Dog chose to have Sarah and Joel leave the office half way through, so as to cut back the length of the walk to the exit door and outside to the car. This allowed them to shave off a few seconds of otherwise boring content and make the action denser. Uncle Tommy’s introduction is pretty hectic with the rapid movements and a lot of information handed to the player, so to reinforce his presentation, Naughty Dog chose to have him turn towards the player once inside the car.
The Last of Us, as many narrative games, presents choices to the player. They come in two types, active and inactive. Active choices are generally self-explanatory and are not even presented as such in games. For instance, play style (stealth or action) influences the outcome of fights, but they can also come in the form of a single button press, which usually destroys immersion in a world, and breaks the fourth wall. Inactive choices on the other hand play with morals and psychology between players and the characters they play. Here we see the first glimpse of the troublesome relationship to come between Joel and Ellie, in the eyes of her daughter Sarah who wishes that the car had stopped to help the stranded survivors. This makes players ask themselves where they would stand, and take sides with the future protagonists.
The accident serves here as a transition and justifies leaving the car behind. The timing of that sequence is probably not purely by chance. First Naughty Dog had illustrated all the themes they could from inside the car, and second the attention span of test players was probably waning beyond that point. Notice however the subtle and progressive introduction of infected in the world. The first one, inside Joel’s office comes in when they may yet be taken for humans, and as an element of a cinematic, it offers no challenge to the player. The second, in this car sequence, attacked other survivors and when posing an immediate threat to Sarah, was stopped by Tommy driving away. The third, much closer view comes in as a direct consequence of the car crash. These three progressive introductions of the game’s main antagonist are a textbook example of exposition that allows players to fully grasp the concept of the infected before having to deal with them. Many games fail to present enemies that way, instead relying on one-shot cinematics with poorly explained antics and backstory. The fact these enemies can’t talk and explain themselves pushed Naughty Dog to work them this way into the game, to great effect.
Combat Tutorial First introduction of the « Square » button for combat. As a rule, tutorials are better introduced when players attention span can be solely directed at the mechanic in question. Here, the tension resulting from the car accident and the sight of the infected nearby focuses the player’s attention on the tutorial for maximum effectiveness. Instead of breaking the immersion of the scene, this tutorial actually flows inside it, and players never question it. Naughty Dog’s choice for minimal UI invasion on the screen really pays off here.
Chase sequence Introducing Joel for the first time as a playable character (and the main protagonist) in the middle of a stressful situation could have been a risky proposal, which Naughty Dog managed to make work by taking away all combat (automatic death if caught by infected and gun given away to Tommy). Sarah, who would have trailed behind and become a center of attention for players is reunited with her dad after getting hurt in the accident. All of these moves act as perfect justification to have players only ever care about their own character. The only action required of the player is to dodge pedestrians, which is exactly the best activity to learn to control a brand new character. The use of a car that crashes into the gas station is a bit of a cliché by now, but remains a very strong justification for sending the player to the right, and have a couple invisible walls across the other streets. It remains better than a simple static object across the street and infinitely better than nothing at all.
The firemen truck across the street serves both as justification for the fire raging in the building next door, and as a way to direct the player to the car crash about to block the street off. This second instance however occurs much closer to the player, increasing tension further, and culminating with stopping in front of the theater beyond. This “three strikes” approach to game design is rampant through most games these days, and certainly The Last of Us, for instance during the combat tutorial (three square button presses to leave the car) and is a sort of golden number first generalized by Nintendo.
The use of audio after the third street blockage is mandatory given the exit is not readily apparent within a 45 degree field of view on either side of the theater. In third person games, presenting exits within this 90 degree angle is generally considered a safe choice to have players flow, on consoles at least where the FOV is usually limited more than on PC games. For instance, this was the rule we followed when designing Prince of Persia in 2008.
In the alley and through to the other side, great use if not subtle of lighting to direct players to the left, and first use of the yellow color on all directional and interactive hints, which we’ll see later in the first level. We’ll find the same use for the ambulance down below, which strong red lights contrast easily with the night, and even the wind indicating the sense of direction. The headlights serve as guidance too of course, much like the hurt survivor on the ground.
This is the first article in a three-part series. Part One / Part Two / Part Three
seir reacted to Thrik for an article, Interview with Jean-Paul Jarreau, Black Mesa level designer
Tell us a little about yourself. Where do you live, and what's your day job? Oh man. I unfortunately live in Rochester NY (no its not driving distance from NYC) and am currently unemployed. I graduated college with a degree in Photography and a minor in Philosophy in 2011 and worked for my step father's company until july of this last summer. So I currently spend all my time working on levels and photography, I honestly cannot complain right now.
When and why did you decide you wanted to bring worlds to life in 3D? When I was 14 I had a passion — playing games. I loved it, it sparked something inside me and I had no idea what it was, but I liked it. My personality can be pretty analytical at times so naturally I tried to figure out how to make them. I love architecture, buildings and 3D space, so I just went for it. I spent hours and hours in every level editor I could find and I, admittedly, stuck with hammer because it was the easiest to pick up, had the largest support and the biggest community. I've been with it since. That's what, 9 years? 9 years of blue screens and move_rope causing crashes straight to the desktop — it's a love–hate relationship.
How did you get involved with Black Mesa? This is funny considering the level of amateur crap I have made. I applied to Black Mesa back in 2004/2005 by sending a VMF to the founder of Black Mesa (Jon) of my recreation of the first Power Up level. It was a mess and ugly as sin. I had quite literally no idea what I was doing, but it was satisfying to create, so I kept with it. I clearly remember Jon's words: "It looks, eh, amateurish". So, I spent the next few weeks refining my work and I applied a few other times. I eventually got on the team and I started work from there. I remember a good friend of mine, Daniel Junek, was already on the team and when I loaded up the private developer forums and saw his orange map layout of the blast pit silo, immediately knew I was going to be on the team for a long ass time — I saw what COULD be Black Mesa and was instantly inspired.
Funny story: I always considered Daniel's work (he did Blast Pit) to be godlike, it was just incredible, I had never seen anything of such quality and ridiculous planning and care. I have this habit of comparing anything I do to the current best work in the world so I know how to improve myself and I considered Daniel's work to be the best in the world. So, naturally, I compared my work to Daniel's and for years it never even compared, but I was slowly getting there. So, a few months ago I, slightly buzzed, messaged Daniel on steam and came to realize he did the same thing with my work this entire time.
I joined the team and looked at what was available to work on and I, of course, had to pick Lambda Core. I had no idea what I was doing and some of the early parts of Lambda Core show that.
What was it like getting to reimagine the world that kicked off many MapCore members' interest in level design? Oh wow I never thought of it this way. Want to hear something hilarious? Before getting on the Black Mesa team I had never played Half-Life before aside from one or two killbox matches on a friends computer.
But as for re-imagining a world that so many hold dear? Its a difficult process, at least for me. I wanted to strike people on an emotional level with my work, I wanted people to see certain things I made and say "Holy shit, this is amazing", and while I don't know if I have accomplished that or not, I do know that getting there was an intense process.
I remember loading up a Lambda Core map and thinking to myself, "What makes this interesting? What makes this bad? What makes this fun? What makes this area instantly recognizable? How do I create something in the present that completely awed those in the past? It was honestly tough in certain parts and I can remember re-doing sooooo many parts of the game because I personally thought it didn't hold up to any creative scrutiny.
Were you often tempted to take a whole new approach to parts of the game? How did you find the right balance of old and new? This is a good question and I bet some die hard Half-Life fans still think we screwed up parts of the game haha. For me it was mostly taking a look at the original and what it tried to accomplish and improving on that. Sometimes certain areas didn't even need improvement, it just needed refinement, an extra level of polish to really take home its intended goal. When there was a controversy over what should be done in the game, we would just look at the 'numbers'. Did a majority of Half-Life players like this certain area? If not just scrap it and try again but be sure to retain the original's emotional impact on the player.
Did it ever feel like Black Mesa might never make it? Was your motivation affected? More than you can ever imagine. We worked on it for about 9 years remember! Honestly? It probably felt like we weren't going to make a release more than we were going to make a release. There were periods of inactivity on our developer forums for months at a time. We all have personal lives to attend to and sometimes it would be a perfect storm of events where no one would work on anything for extended periods of time. Then, out of nowhere, something would happen and everyone would get sparked and jump right back in and get loads of work done for months. Rinse and repeat.
In the SUPER early stages of development there was a point where we didn't even have a coder for 2 years. You can only imagine how incentive was on the team at that point.
There's still a significant portion of Black Mesa yet to be completed. Will it be purely new material when the remainder is released, or are you revisiting and/or expanding existing material? I will answer with a simple work-in-progress screenshot of something I am working on and nothing else:
What do you think is a stand-out part of Black Mesa? I have actually never thought about this! But the FIRST thing that jumped into my head is when you first enter the turn-table room in Power Up. You witness an intense fight between a Gargantua and a few human grunts. It's incredible because of the collaborative effort that went into it without a single f**king hitch. It's like it was done by one person! Let me explain:
You enter a beautifully themed section of the facility created by Daniel Junek You witness well crafted animation by Nate of the Garg getting blasted by a human grunt and consequently falling into and breaking the concrete wall near him. The Garg was made by Chris The surrounding textures are made by Mark The dead human grunt bodies turning into scorched corpses, the flame effects, and the Garg behavior was done by Paul and Mark All the sound was done by Joel Look at that! It's the personal work and vision of a team of people all together forming a unified and impressive quality of interactive work. It wouldn't even be possible to create something that well done by one person.
Now you may say "Hey, thats how all games are made." But, its a little different when you are creating something of this size and scope and all you have to communicate with your co-workers is Skype and a forum. There were a lot of times when I would be making a gesture with my hand or something while trying to describe the way I want a prop or texture to look. That was frustrating.
You do a lot of multiplayer level design in addition to your single-player work. Which do you prefer? Single player is telling a story and using the environment to make playing the story fun and visually interesting while guiding the player through a struggle. If you can hold a player's attention and feed them the facts then you are good to go. Multiplayer is about cohesive design and fun welding itself into one entity and that can be difficult at times. Honestly I probably prefer the multiplayer work because of repetition. You know players will be running around your environments thousands of times which will eventually lead to an appreciation for the map on tiers usually only seen by the actual level designer. That is a double edge sword though.
Briefly, what's your general workflow for putting a level together? LAYOUT, LAYOUT, LAYOUT. I try to get simplistic layout for a map that is fun and cohesive before doing anything else. Why do you play a video game? Because it's fun, and as a level designer, that fun can be directly proportional to your quality of work. You are almost like a spokesperson of your entire team's work and I love it. I mean I could tell you about all my thought processes while I work but that isn't as fun. But, I usually tend to think of ideas in a 3D space, so I never really draw layouts, I usually just throw some brushes around to get my idea out of my head and see what can be made of it before I attempt any sort of real layout. A lot of my past ideas failed even before getting to the layout stage, so that is good (I think).
Which bit of the process is your favourite? That part in a map's life right as you know everything is fun and you get to unleash your thematic ideas you had been planning in your head while working on the gameplay and layout.
And your least favourite? Any sort of tedious mind numbing repetition is annoying. Or when you spend a long ass time solving a problem in your level only to have it come out like crap but its 5am and you have to go to sleep knowing what you just created is terrible. That's no fun for anyone.
What's the hardest problem you've encountered while mapping? Writers block can apply to anything really. There were times where I would just stare blankly at a map for hours not knowing what to do. Recently though? I cant stop with my ideas, it's great. I'm trying to ride that wave for a little.
Also, scrapping something you have put a massive amount of effort into can be disheartening but it has to happen sometimes. But what essentially is anything creative? It's the formation of an idea while hiding your influences. Sometimes these ideas need to go through an emotional struggle before work can continue.
Which piece of level design are you most proud of? To date, my Black Mesa magnum opus is most definitely the final Lambda Core teleporter and its prior lower reactors. Fun fact I thought some designers might appreciate: the entire Lambda Core reactor and teleporter actually lines up like an actual facility if you copy all the maps into one file and place them correctly.
More recently? It's my map cp_imbricatus for Team Fortress 2. I have a feeling that the map will fundamentally change the way a good three-control-point attack/defend map should play.
In addition to level design, you're an accomplished photographer. How did that begin? Accomplished? Oh you. Thanks, you.
Here is a story I have never really told anyone haha. I was an idiot in high school, I mean who wasn't? But when it came time to apply for colleges, I just straight up didn't make the cut. I applied to stupid high reach schools and nothing else, I think I even applied to the wrong program in a few schools. Needless to say, I didn't get in anywhere and I panicked. My mother had some connections at Rochester Institute of Technology and randomly one day called me while I was out and asked me a few questions:
She said, "I got you into RIT and their photo program doesnt require a portfolio, which would you like to do? Fine art, photojournalism, advertising or biomedical photography?" I got lucky and I knew it. I wound up graduating with a bachelors of fine art in advertising photography. Crazy.
A lot of your photos are taken in very challenging conditions, such as live music performances. What skills and equipment do you employ to get such great results? Simplicity is key. Both visually and philosophically.
You must look at every situation like you are the one viewing the content with no prior context as to what is going on. Do you want to portray a story? Do you want to create something purely for its visual content? What's your end goal? I could honestly talk about this for hours.
I currently only use a Canon 5D Mark 3 and the only lens you ever need — a prime 35mm f/1.4. A good photographer with a bad camera is going to make better pictures than a bad photographer with a good camera. I also adhere myself to a lot of personal restrictions (in level design as well). An example being I never crop any of my images, I always thought it was best to just leave it straight out of camera. I do Photoshop my images, but I don't crop them. (Note: if you are a client that hired me as a photographer, I will do anything you want to the images if you request it. These rules usually only apply to personal work.)
Keep an eye out on my Tumblr for some new photography coming up soon.
Do you find that your photography skills bleed into your level design? For example, when composing scenes? Absolutely! And it works both ways too, photography helps my level design and level design helps my photography. Lighting plays a more important role in both than most even think. I am just lucky to have experience in both.
I don't want to toot my own horn here but look at the lighting in some of the reactor areas in Lambda Core. It got to the point where I set 50% and 100% falloff distances with a hard cut-off for the yellow lights surrounding the central cylinder so the centre reactor would ominously glow yellow without casting any mood changing light onto the surrounding geometry. Does that level of detail pay off in the end? I don't know but I hope so.
What games would you say have inspired you the most? I have never thought about this before but here is a list of the first games I can think of to inspire me:
Ocarina of Time: Video game perfection. Majora's Mask: I cannot even put into words the impact this had on me Wind Waker: How do they even make this stuff? It's like they converted crack cocaine into playable form. Half-Life: Was I obligated to say this? This game changed what a first person shooter could be with its incredibly inventive layouts and story telling. Painkiller: The basic first person shooter mechanics perfected and in gory glory too. Incredible environment art to boot as well. Banjo-Tooie: Incredible game design, just incredible. Battlefield 2: The amount of serotonin released in my brain by playing this game should be illegal. Super Mario 64: An amazing sandbox game before that term ruined a lot of modern day games. Far Cry: It materialised my emotions while playing a game by me actually muttering the word 'wow' while playing in certain spots. There are loads more but these are just what I could think of off the top of my head.
If you could give a budding level designer a short piece of advice to help them succeed, what would it be? Your work is going to suck for a long time, get used to it. If someone is being a dick about your work, they probably have a good point. Listen to them without bias. This will push you beyond what you thought you were capable of. You made the level, no s**t its easy to understand by you — other people HAVE to play it before anything is considered good. Scale will make or break you. Do a lot of scaling tests in every game you want to develop for before attempting anything else.
MapCore would like to thank Jean-Paul for his time, and wish him the best of luck with his future projects. You can enjoy more of his work by visiting his portfolio and Tumblr. He's also looking for people to help out with testing his latest Team Fortress 2 map; simply download 'cp_imbricatus', take a peek at the
and join in on the tf2maps.net server.