Designing a Noir Scene in UE4 at Ringling College
Events
Subscribe:  iCal  |  Google Calendar
Milwaukee US   17, Jun — 22, Jun
New York US   17, Jun — 20, Jun
St. Petersburg RU   19, Jun — 21, Jun
TBA BR   22, Jun — 24, Jun
Amsterdam NL   25, Jun — 28, Jun
Latest comments
by Yes
5 hours ago

Gabe is amazing and im happy to have him basicaly be the face of gaming. If you have something against him reply to this, i would love to discus

by Kai Marshall
20 hours ago

Awesome! So proud of you!!! <3

Positive site, where did u come up with the information on this posting?I have read a few of the articles on your website now, and I really like your style. Thanks a million and please keep up the effective work. rv altitude

Designing a Noir Scene in UE4 at Ringling College
15 April, 2019
Environment Art
Environment Design
Interview

Devin Dixon talked about the development of the large interactive UE4 project Ain’t No Mountain, Baby made at Ringling College and her studies.

Introduction

My name is Devin Dixon and I am currently a senior at Ringling College, studying Game Art. I enjoy working in environment art while specializing in asset creation and lighting. In the past year, I have also experimented with UI design and implementation of my senior thesis, Ain’t No Mountain, Baby.

Originally from Virginia, due to my parents being in the military, I have been able to travel around the world and spent 3 years in Japan. My travels brought me to Suncoast Polytechnical High School in Sarasota, Florida, where I was exposed to 3D programs such as UDK 3 and Maya. This was my first experience with 3D as I previously wanted to be a cartoonist. I enjoyed experimenting with 3D art during this period in high school a lot. One of my teachers at the time, Cathie Janssen, constantly pushed me to practice my 2D and 3D art. I have to thank her for helping me consolidate my ideas throughout high school and ultimately steering me to Ringling.

At Ringling

At Ringling, our curriculum entailed learning the basics of environment art. Lessons included VFX, material creation, hard-surface modeling, and programming using UE4’s visual scripting. Everything prepared us to create our own game for the senior thesis. During my sophomore year, we created a lot of blockout levels including a battle arena and racing map and studied levels in games to get a good grasp of game design. In retrospect, I remember being frightened at the thought of doing a grandiose thesis. It was an ominous concept at the time. Throughout my time here, though, I learned to really appreciate what is involved in game development and manage asset creation pipelines, alone and within groups. Overall, the sophomore year focused on acquiring programming skills, therefore, the visual fidelity of our levels looked more like early blockout stages. I really enjoyed this period, because it was challenging but a pure expression of our programming ability.

During the first semester of junior year, we worked on biomes, interactive environments, and first-person shooters. During my time here, our curriculum has mainly focused on the visual aspect of environment art. The programming aspect is taught as a supplementary skill to push the boundaries of our environments and enhance the player experience.

Idea Behind Ain’t No Mountain, Baby & Worldbuilding

During the second semester of junior year, we went through pre-production for the thesis. During it, my goal was to solidify the relationship between my care for intimate environments and storytelling. Ain’t No Mountain, Baby began because I wanted to explore why cults are so prevalent, specifically in 1970s America. I have always been fascinated by the abundance of crime, serial killers, and the cult-like mentalities that seemed to flourish in the 70s, and how the Vietnam War’s lasting ripple influenced society at the time. The juxtaposition of the seemingly flowery, free period of the 70s with its more violent, cruel aspects was so dynamic and appealing to me, and I wished to amplify this dissonant relationship in Ain’t No Mountain, Baby. Due to the nature of the subject matter, I wanted my environment to feel close, intimate, and be filled with nuanced details.

Therefore I had to build my own cult, from the ground up. The cult in Ain’t No Mountain, Baby is called The Holy Mountain and they own a radio station referred to as The Mountain. The game occurs in an alternate universe two months after the mass suicide of the People’s Temple cult in Guyana of 1978. This event deeply influenced The Holy Mountain’s leader, Bailey, and pushed her further towards extremism. I know what you’re thinking, but these key names are not just based off the 1966 Marvin Gaye hit! While further exploring the lore of The Holy Mountain cult, I wanted to provide a greater commentary about religion and its interaction with racially challenging ideals as this is a racist cult that takes advantage of its participants. With the name Ain’t No Mountain, Baby, I wanted to imply that the values and riches that Bailey promises her followers do not exist and that Bailey is not the savior her followers wish her to be.

Cults, in particular, are formed from a powerful leader taking advantage of weaker-minded individuals’ desire to belong to something greater than themselves. I pushed this overarching theme throughout the gameplay, as it is a dialogue game and the player learns more about the cult as he or she speaks to one of the followers at the station.

I am talking about worldbuilding a lot because I value research and reference in my work. Because of the time period, subject matter, and location, I found I had to do a lot of research. I studied and scanned 70s photography & ad books to better understand how beauty was idealized 40 years ago. Through these avenues, I was able to better inform my modeling decisions. Through my education and attendance of GDC, I have learned that sometimes we as an industry, are preoccupied with looking at only game content and are incapable of referencing other forms of media (which contain so much value) leading us to create in a vacuum.

Research & Planning of a Large Project

As mentioned, I did a lot of research for this project. I started by creating a massive asset list based off of objects I saw in a fantastic ad book that cataloged 70s ads. I scanned a lot of pages from this book to use as reference and worked from there. After I made this list, I started filling in the gaps because I knew I needed to create audio props and papers like propaganda images, but obviously, those are not well-documented in an ad book. I watched shows and movies set in the 70s and actual 70s media for more perspective on the era to help inform my design decisions. Additionally, I became inspired by some modern props and wondered what their 70s equivalent looked like. Thereforeб in order not to be anachronistic, I tried to find 70s versions on eBay if they even existed. Thankfully, eBay sellers often take pictures of their products from multiple views. There are 3D artists everywhere who quietly appreciate it! Sometimes, seemingly unimportant prop ideas popped into my head such as a Speak & Spell but I included them in the list anyway as I felt they enhanced the legitimacy of my environment. I also gave myself an extra challenge to approach my thesis in a stylized way. When I created my art guide, I aimed for textures between Overwatch and Sunset Overdrive. I used the deformation tools in Maya to either bend or squash my models after modeling them “straight”. I also had some rules during modeling: mainly, I chose to exaggerate small details and push the chunkier aspects of the real items in 3D to make them feel more oversized. Generally, objects have beaten edges, a high center of gravity, and tapered silhouettes.

Texturing

I primarily used Substance Painter and Photoshop to texture each asset. Certain objects have an assigned world-based gradient made in UE4’s material editor and a couple of choice materials were made in Substance Designer like the ceiling tiles. But primarily, I used Substance Painter to bake my props onto themselves, regardless of if they had a high poly, in order to get a curvature map that I could then use as a mask to get the lightened edge effect seen on most of the assets. I used this tutorial by 3dex as a starting point. I then overlaid painterly brushstrokes – it is a fingerprint roughness mask that I like – on many of the props if there are large, empty spaces in order to avoid leaving just flat color. For metal objects, I would simply choose a base color and edge outline with metallic properties. For other objects, like the radio or reel-to-reel, I made several layers following the tutorial and took out what I felt was unnecessary for my uses, such as the light masks, but then added more detail using Photoshop. Wooden props were quickly banged up in ZBrush using the popular Orb brush pack by Michael Vicente and later baked in Substance Painter to preserve and emphasize the sculpted dents and cuts.

Once I established this process, it made most of the assets relatively easy to texture. Granted, it was time-consuming, but texturing objects in Substance Painter is a method that makes the most sense to my workflow. I prefer custom texture generation over procedural, however, they both have their uses.

Lighting

During the blockout phase, I established my lighting very early and improved upon it throughout the months like tweaking the saturation or intensity values of the lights. One main rule that was established early on was the concept of simplistic lighting. With the exception of Bailey’s DJ desk, which is lit with two lights, the scenes you see here are lit with one light each. I wanted to keep the lighting simple and isolated, something I observed while playing Dishonored 2 and watching Mindhunter. By choosing to approach my lighting in this way, I mainly wrestled with dark corners and blurry shadow shapes as assets started to get added in the game. There is a fill light for the entire level just to boost darker corners, an issue I repeatedly ran into with the side office image. I also used a custom light function for each of my 3 blind models to create strong noir-esque shadow shapes. Prior to doing this, the streaks through the blinds were not as striking as I wanted them to be. Before using the light function, I played with each object’s lightmap values that were affected by the light. The light function was a relatively simple solution to this inefficient approach, with a big aesthetic pay-off.  Perhaps, more of a secondary concept, but a lesson I learned is that it is important to pay attention to how lighting interacts with the addition of assets and their materials. I was particularly occupied with the muted color palettes in 60s photographs of astronauts. I wanted to maintain the saturated orange light through the windows to bring out the dull cool blues of the tech props and muted tones of the other assets.

Optimization

Because of the number of objects in my scenes, I credit most of my optimization efforts to lower poly models! I never feel the need to use more polys than necessary or to smooth everything that goes in-game. Sometimes, a bevel is good enough. As for in-game, I made decals to be used on a variety of objects, so that textures could be a little less custom for certain assets. Those marker writings on the cardboard boxes are 512 decals. The wallpaper design is also a decal, started in Illustrator. I like decals, they saved me a lot of texture space and are easy to implement. And, as mentioned earlier, the simple world-based gradient I made was useful for simple props that did not need special textures with lots of detail.

Achieving Quality

I would like to believe that the development process, pre-pro, is the most valuable stage for any project. Providing optimal time to research and actually develop your ideas is endlessly beneficial to the end result, often in unnoticed ways until an obstacle is encountered. This means fewer decisions will need to be actively made during the execution process. Basically, the more thinking I did early on in the process before I was responsible for making textures, models, interactive gameplay, etc., the less likely I would fall behind and trip over my own project’s work-load. As for the high-quality visuals, I can really only owe that to Ringling having a high bar for the technology and programs that the students have access to and the tough curriculum meant to prepare us for the industry.

Time & Challenges

I started this project in summer 2018, so it took me around 10 months. This excludes the pre-pro time, however, meaning the entire project took a year. Materials were my biggest challenge as I was unsure of how I wanted to create my textures until month 3. Also, programming a dialogue system with conditional aspects was incredibly hard, too! I got some help from quite a few of my technical art friends, shoutout to Izel Moctezuma and Arty Bulgakov!

Devin Dixon, Environment & UI Artist

Interview conducted by Kirill Tokarev

The goal of the ClearCut courses is to teach you a solid workflow that is used in the AAA game industry. The first episode covers the process of creating an AAA fire hydrant from start to finish.

Any future updates are included and will be available for download in case they are released. Next episodes are not included.

Check the full description

Contact Emiel Sleegers

Leave a Reply

avatar
Related
Education
Character Art Program
Education
Environment Art Program
Education
Character Art Program
Education
Electives