Three Pillars of Environment Design
Events
Subscribe:  iCal  |  Google Calendar
1, Jul — 1, Aug
Cambridge GB   13, Jul — 17, Jul
San Diego US   19, Jul — 23, Jul
Torino IT   25, Jul — 29, Jul
Shanghai CN   3, Aug — 7, Aug
Latest comments
by Don\'tLookAtMeI\'mBusy
3 hours ago

I would love to run Mobile Photoshop with Sculptris and Moi3D on my Surface Pro

by Don\'tLookAtMeI\'mBusy
3 hours ago

I'm sure it's spelled iTubs

by sipon
6 hours ago

Too bad I work 95% of texturing on SP now...

Three Pillars of Environment Design
21 September, 2017
Interview

Environment artist and a teacher Dan John Cox discussed the most important things about environment design that aspiring game creators should take into consideration. Don’t forget to check out his lectures and his website!

Introduction

I’m Dan John Cox, I’ve worked in games for 10 years on lots of very different projects. Most recently I’ve been working on Below and was on the small team that conceptualized and created Starlink: Battle for Atlas. I’ve also worked on Splinter Cell Blacklist, and a lot of small cancelled games!

Three things about environment art

My three big points for my students would be 1. Patience 2. Balance 3. Observation

1. Patience: It takes a long time, and a lot of failed work to get good at art, and even when you get good you don’t think you are. Patience to me is more important than practice. Cause anyone can practice, but how long can you practice before you expect specific results (a job, great skill, external validation of some sort). Patience means being kind to yourself and keeping at the art form over time. I’ve had so many talented students who didn’t get a break despite their amazing skill, but they stuck with it and it’s paying off.

2. Balance: Be good to yourself and your mind and body. When I taught at a school our course was short so we drove the students HARD, they often had sleepless nights. This can and should only be done for a short period and then the artist needs time to recover. I always pushed my students to fully rest when break time came and to not create unless they desire it. One can create art when their mind and body are in shreds, but learning doesn’t take place there. Plus it can create long term problems.

3. Observation: This is the most cliche one but it’s too important to leave out. Basically you have to look around. As you learn your craft you need to compare you results with what exists in the world and attempt to replicate it or at least pick it apart to learn more about it. I never looked at wood surfaces until I was learning 3D and discovered how super compelling they are and what makes certain wood forms interesting to me and what makes some 3D ones fall flat, you can’t gain that knowledge without constantly looking at the world around you.

My interior design talk fundamentally changed how I teach environment design and construction. I had a vague idea about how these things worked visually but never had to sit down and more completely break it down as I did with that talk. I’ve often used those same terms to discuss what’s working in a scene and what isn’t, it’s been very helpful for me to articulate whats right and wrong, I’ve often used it in studio work as well.

Focus

Two games came out recently that are both narratively focused, environment focused and handle order with different levels of success. That’s Tacoma and Hellblade. Before I start any critique here let me be clear, these are both great games and my points don’t make either bad or good its just simply how they’ve handled one small aspect of their world structure.

In my opinion Tacoma has much better level design, which is effectively what Order is when it relates to interior design and environment design. In Tacoma you often can just walk and find you way to the place you want to be, each path is unique and its pretty easy to understand the structure of a set of rooms very quickly. There are nearly zero repeat rooms or hallways, the structure is always just different enough to help you memorize the structure of the room and adjacent hallways. Visually the walls are simple and non distracting and the set dressing is what stands out making it again easy to remember each room in your mind with little effort, which means you can visualize a set of spaces very easily.

Whereas Hellblade’s scenes can be full of visual information which can sometimes to the game’s benefit as it’s meant to be disorienting. However there are times in the game where you get lost because the area is so large and objects and materials are so similar the whole way through that it’s hard to remember the structure of a space you were meant to puzzle through. In this case I went through one area where there was pouring rain, bright lights hanging from trees, intense chomatic aboration, thin spiky trees, thin spiky ground blocking objects, and more taking up my visual field. I got lost multiple times trying to solve one puzzle and only really solved it through stumbling into it which was my experience with a number of puzzles. Chaos in the visual field fits the game’s narrative 100% but it starts to hurt a player’s ability to reliably solve a puzzle. Keep in mind too that Order and visual chaos can still exist hand in hand but that’s an incredibly tight rope to walk.

Repetition

Since it’s fresh in my mind I’ll use Tacoma again as an example here. Each area of Tacoma is fairly similar visually which makes sense, it’s a space station. However the game does a good job of simply changing colour and lighting for each area just enough to make them feel of the same place but visually distinct so that a tiny team can get great results.

Moreover, each area has LOTS of small, interesting objects to pick up and interact with and this is where the player will start looking because there is lots to dig into with little knickknacks that are full of world information and detail. These are repeated sometimes as well but the Tacoma team did a good job at creating impactful elements that they don’t repeat while heavily reusing much of the general environment structure which ensures areas are going to feel like they go together while also being manageable from a production standpoint.

Enrichment

You are correct when you say it’s easy to go overboard here. Your best bet is to pick only a few techniques for enriching a space and try and stick with them instead of throwing everything you can at the wall to make it beautiful. Often times in games we add lots of grime and objects to make a space feel detailed and compelling, we do it because it works and it gives the eye lots to do but it’s easy to go over board. Many new developers that have started using Unreal have created games that have compelling visuals but get very noisy because they aren’t using enough restraint in their use of enrichment.

Symbolism

Hellblade is one of the few modern examples of using heavy symbolism in their environment design and visuals. They frequently use symbolism of light and dark, fire, descent, disorientation and more to express their story. It’s easier in this case for them to do this as it’s kind of the entire point  of the narrative so everything is built to support that but regardless it’s still pulled off very effectively in the visuals and story in a way that feels cohesive.

Do you think interior design can help gameplay?

Oh it helps for sure. Interior design is all about supporting a person’s experience in a space, which is the exact same thing we want to do with game environment design, so their goals are linked. Just like in the real world you may want to make a place that convey’s simple information to people and gets them to where they are going (like an airport). So some airports are not visually compelling, however bigger international airports pay close attention to aesthetics and ensure you can always find a sign telling you where to go that is surrounded by interesting shapes and designs that don’t distract from the signage (which is functionally the most important thing in an airport).

We do the same thing in games but we usually can’t put signs down to tell people where to go. So we kind of need to be more like Ikea. Where you simply walk, see interesting things. To me, walking through an Ikea is what it feels like to walk through most game worlds. You enter, see interesting things at a very prescribed rate and then just walk and walk until you’re suddenly at the exit with no clear sense of how you got there because the rest of the route was pretty seamless unless you go into the branching paths where it becomes very easy to get lost. For linear games this is good, cause you don’t think about “where you’re going” in Ikea instead you just move forward and focus on the objects. For a more open game this would be terrible because as soon as you break from the main path it’s easy to get totally turned around in Ikea where that’s less likely to happen in an airport because you’re meant to move back and forth.

Tacoma handles this work really well because you never notice that you’re being lead around visually and the art supports the gameplay a ton.

Learning

I’m not going to pretend that I read environment art books all day like some sort of Renaissance man, so I’ll be honest and say I don’t have many book recommendations 😛 but Shaping Interior Space and The Design of Everyday things are great places to start for thinking about the basic structure of environment art and design. The next thing you can do is start to listen to concept artists like Shaddy who breaks down the elements of visual design and structure. The more you do this the better you can start to understand what it is you’re actually working with as an artist and what you can remove and take away. These techniques are hard when it comes to environment art because this work is so very technical in games that it can be tricky to focus on the big picture when there are so many small elements to take into account like tricounts, texel units, overdraw, culling, overlapping shadow casters, memory budgets, and on and on. Just try and take time to stop, maybe paint over your 3d scenes and think about what you really need, learn some of the basic tools you can use to dress a scene and use those as jump off points for figuring out what could improve the scene by simply using those rules, what happens when you do?

Dan John Cox, Artist at Capybara Games

Leave a Reply

Be the First to Comment!

avatar
wpDiscuz