Daniel Hinckley talked about his education, work with vegetation and references for vegetation, and the process of creating A Cycle of Violence, a UE4 scene inspired by The Last of Us: Part 2.
I’ve completed a couple of environments and several materials over the course of this year during both university and my own personal time, allowing me to sharpen my skills and techniques in preparation for the scenes I’m currently working on for my final year of university. I’ll be making an update post about these shortly on my Twitter.
I’ve loved playing games ever since my parents bought me a PlayStation 3 as a gift during the holidays when I was around ten years old. I specifically remember playing the game LittleBigPlanet which was my first exposure to creating art for games. I used to build my own levels, and among my friends, I was often the artist.
During high school, I wouldn’t have considered myself to be the artistic type, but I became progressively more interested in applications like Photoshop and Premiere Pro. I jumped into the YouTube Gaming or ‘Lets Play’ haze, which was popular at the time, and also fell in love with games like Journey, Shadow of the Colossus, Dark Souls, and Silent Hill. This would have been my first exposure with these tools while following YouTube tutorials to learn the software and techniques.
After high school had ended, I had no idea what I wanted to do. I loved art and video games but felt like pursuing this as a career would never be feasible or accessible to me. Despite this, I would spend all my free time programming games and creating art for them in Photoshop.
In my second year of college, I volunteered for a two-week QA work placement at Codemasters and I became fascinated by the 3D Art workflow after learning more about it since it tied all my interests together into one single job. Since then, I’ve never looked back and I’ve been working towards the goal of becoming a 3D Environment Artist every day since.
At Staffordshire University
I applied for Staffordshire University after hearing many stories of graduates who got into the games industry upon finishing the course. The vast number of courses and modules available and the framework which the course follows was definitely more tuned for industry than any other university I visited that was local to me at the time.
Starting my course at Staffordshire University has been by far one of the best decisions I’ve ever made. I had no prior 3D or fundamental art knowledge before starting my first year, and I’m now entering my final year. I’ve learned so much in the past two years, and I feel more confident and excited than I’ve ever been before.
My motivation comes from the pursuit that one day I’ll be able to work on AAA open-world titles as an environment artist and play a role in building some of the same worlds I felt so inspired by. Having the pleasure of being able to contribute to something that can positively affect people all over the world would honestly be my dream.
Procedural Approach to Environment Art
I enjoy using procedural workflows, and I’m always looking at improving how efficient and reusable my assets are, especially with my materials. Creating materials in Substance Designer is probably one of my favourite pipelines whenever I’m working on game-ready environments due to its versatility, and how quick it is to create something with great results.
The environment pipeline as a whole can be very iterative, and I find myself reopening my Designer files and tweaking my materials quite regularly. The ability to procedurally create the results you want in a program like Designer feels very intuitive and easy to use once you begin to pick up the fundamentals for it. I often set up custom utility nodes and parameters within my material graphs, allowing me to create quick variants of materials I can later use to blend together in Unreal.
Reference for Foliage
I also look at scanned foliage atlases quite regularly to use as a reference to bring the best out of my albedo maps. Quixel Megascans are great for providing accurate real-world references, which you can rotate around and examine closely in a 3D viewport.
For the fern plants in my scene A Cycle of Violence, I simply used a plant in my garden and took quite a few close-up shots with my camera to use as my main means of reference. I enjoy photography as a side-hobby, so whenever I get the chance to take photos I will, and I’ll usually refer back to them for my scenes if I feel they’ll come in useful. Understanding the asset you’re creating whether it’s a prop, a material or foliage, etc. it’s incredibly important that you’re creating something that’s accurate to its real-life source.
Using Megascans to Your Advantage
I have still yet to work on a scene that uses Megascans assets, but this is definitely something I plan on doing shortly for one of my next scenes. There’s certainly a place for Megascans assets, especially when you’re aiming for photorealism. When you’re working as an individual artist or within a small team or studio, the ability to create large-scale photorealistic scenes that are sheer in scope is now something that’s much easier to achieve with Megascans, allowing small teams to create large gameworlds. I look forward to seeing more of that in the future, and with Unreal Engine 5 around the corner, I can only imagine this being pushed even further with the capabilities of the engine.
I remember watching the Quixel Rebirth video around the same time I started game art and feeling completely blown away by it. The things we’re able to achieve now on an artistic and technical level have accelerated so fast over the last couple of years.
A Cycle of Violence: Goals
A few weeks before I began working on my Last of Us scene, I worked on a smaller test scene to teach myself the fundamentals of creating a small organic game-ready scene. So my main objective with this next Last of Us environment was to take the best aspects from what I’d learned previously, push them further, and improve upon some of my weaker areas at the same time.
While working on the scene, from the blockout up until the final renders, I referred back to my references and compared my workflows with industry examples, thinking about each step I was taking and implementing any feedback or suggestions from peers.
I particularly wanted to push my lighting, materials, and foliage skills further, which led to exploring some new workflows, such as opting to create my foliage in Substance Designer rather than sculpting it in ZBrush. I had also been watching some of Naughty Dog’s Gnomon talks, where they discussed their workflow for approaching the environments featured in Uncharted, which I strongly recommend to any aspiring environment artist as they’re some of the best material out there.
The Last of Us: Part 2 is a stunningly beautiful game, both artistically and technologically. Many of the techniques used within the game definitely inspired me to push myself the extra mile. I was fascinated by the amount of visual storytelling that was applied to every part of the game. Every single asset had a story and a purpose for being there. I particularly loved the ‘Seraphites’ locations in the game, where during one segment the city’s park had overgrown and evolved into a jungle in the middle of a city. I based my environment heavily on this scene.
I didn’t have a specific story in mind for my scene, but I wanted to further emphasize the idea of this “cycle of violence” which is a prominent theme within the game’s story, which is also where I took the title for the scene from.
Creating the Environment
I started working on this scene after the artists at Naughty Dog posted their work to ArtStation, where they uploaded concept art, environments, and some extensive breakdowns. After looking at the Seraphites locations I wanted to create my own environment inspired by that.
I started by blocking out my landscape in Unreal with the main primitive forms of the scene. I added in some of the main light sources and exponential height fog to set the mood and theme. This allows me to control the intensity of the fog using Unreal’s volumetric fog, driving the intensity of that with the lights used in the scene.
I then created a landscape material shader and blended some materials together which I created in Substance Designer. The main materials I used were a dirt ground material, a grassy ground material, and an eroded, broken road material I created specifically for this scene.
I created a tree bark material and pine tree branch atlas and took them into SpeedTree to create the pine trees. SpeedTree is an amazing program for tree creation. Being able to procedurally build your trees to the specifications you want, with the ability to randomise the seed which allows you to create as many variants as you need, certainly saves a lot of production time. It would take much longer to sculpt them individually.
I framed many of the building positions for this scene with my main shot in mind and decided to extend some of the buildings inwards. I did this to create a more isolated feel with the scene, rotating some of the buildings quite far in the background to create more of a triangular silhouette to lead the eye. I wanted the buildings in the distance to feel overpowering to the player, creating that contrast between the natural elements in the foreground and some of the more brutal, concrete structures in the back, reminding the player that the scene takes place in the middle of the city.
I began my asset production by planning out the assets I needed for my scene and gathering reference images to support each asset. In this case, it was the streetlight, car, building front, lobby entrance, and flag. I then searched for images I could use for reference online, and I created a Miro board to save them all in.
Before I knew I wanted to be an environment artist, I specialised heavily in hard-surface modeling in 3ds Max, doing particularly for props, but also a few cars and weapons. Because of this, I’m able to model props for my environments quite quickly, which definitely helps to boost my overall workflow speed whenever I’m blocking out or creating props for my scenes.
I modeled the hard-surface assets using 3ds Max and textured the props using Substance Painter. Some of the most visible examples of this within the scene would be the streetlight and the car which are in the centre of the shot. Due to their close proximity to the focal point, it was important that the textures for these assets would receive an extra pass of detail when it came to texturing them since most of the attention would be focused on those assets.
Adding Substance Designer to the Workflow
I actually only began learning Substance Designer at the start of this year. I began by watching Daniel Thiger’s Fundamentals tutorial series, which taught me how to use the nodes as well as the fundamental techniques and workflows used for material creation within Designer. There are some other great tutorials out there too, on websites like Gumroad and YouTube which will provide you with a strong foundation for material creation.
I later progressed onto creating rock materials, tree bark, and various ground materials which all pushed my skills in different areas. I think when you’re just starting out, it’s best to work on the main different types of materials that will get you the techniques and workflows down, i.e. dirt, rock cliffs, roof tiles, to name a few, and you’ll then begin to build your skills from there.
I’ve wanted to explore creating foliage in Designer for a while due to how easy it would be to create quick variants that are based on parameters, which would save time in the long-run as opposed to sculpting variants individually. After watching Bogodar Havrylyuk’s Creating Foliage In Substance Designer tutorial I wanted to try something similar myself for the leaf plants and fern in my scene.
I’m personally a big fan of centre framed wide angled shots, and how you can create depth with them. I designed a lot of the scene while considering that composition in mind, and then I further exaggerated this with value and colour contrasts.
I feel that composition can be something that’s preferential to the artist, as it’s one of the main ways you can communicate with the player, and get them to focus on the area you want. In my case, I wanted to demonstrate how something can feel isolated, but also large and sheer in size at the same time, because that was the vibe I felt while playing through the game.
The Last of Us: Part 2 is a very dark game, and the world is quite violent, but among that, you have these glimpses of hope, natural beauty, and relatability within certain aspects of it, and this contrasts heavily with the actual theme and the story, and I think that’s what draws people’s attention to the game, by creating those small glimpses which people can connect to and feel familiar with.
In my scene, I used trees and buildings to frame most of the scene towards a narrow path in the centre, which leads the eye to the main focal point of the shot, where the flag and lamp post would be, and then I used fog for the parts of the scene which carried on further than that, to add more depth, allowing me to focus most of the viewer’s attention to towards the midground. However, I still wanted to make sure I left the background of the scene visible to show that the scene continues onwards beyond where I want the player to focus on.
I think as an environment artist, you have to consider each of the foreground, midground, and background elements to your scene, and ensure that nothing is seeping through that’s breaking the immersion, which is critically important for games.
There are certain rules I tend to follow with lighting, and then there’s the aspect of what mood or theme you’re looking to present to the player. A lot of different elements can also tie into this, such as the climate, location, time period, or condition of the scene. I tend to put a lot of detail into my lighting as early as the blockout, to plan out some of those primitive values and set the mood for the environment I’m looking to create.
Exterior lighting can be quite intimidating if you’re unfamiliar with it. There are definitely some important considerations to make when lighting an outdoor scene, such as the time of day, direction of the sun, weather conditions, so it’s best play around with these first by adjusting the sky colour, skylight and directional light values.
I originally planned for my scene to be much smaller, so I added some volumetric fog, and in my earlier shots this seeps way further into the midground than it does in my final version of the scene. I personally love playing around with fog and lighting values to set different moods. It was definitely one of my favourite parts while creating this scene.
Once the main mood is set, I’d then highlight the parts of the environment I want the player to focus on. You can do this with spotlights typically. In this case, I’d have my main focal point which is the centre which would have more light, and I’d push this even further by increasing the contrast and vignette within my post-process volume. It’s best to give these some extra subtle attention, either by increasing the lighting values or contrasting the colour in a way that draws more attention to it.
One lighting challenge I faced was the fire within the scene. I wanted it to stand out to the player, but it was also important that it didn’t overpower the scene and draw attention away from the centre at the same time. This took a few iterations to get right, but with some squint tests and feedback from peers, I was able to get there and find a result which felt right.
The biggest challenge I had with this art piece was trying to emulate one of my personal favourite games, and trying to live up to it the best I could. I went through many iterations for the composition and lighting especially to push the depth and atmosphere as far as I could. The most important rule I set for myself was that the environment I created needed to feel recognisable as a Last of Us scene, and it needed to feel believable and carry enough weight in terms of visual storytelling.
This environment was also the first time I tried out many new workflows I now actively use. It was the first time I created foliage using Substance Designer and trees with SpeedTree.
The experience I gained from using these workflows is already helping me with the scenes that I’m currently working on now, where I’m faced with both familiar and new challenges for each of these too. I enjoy the problem-solving process that comes with environment art and figuring out the most effective method to achieve the result I’m looking for.
I generally enjoy exposing myself to new workflows and techniques I wouldn’t normally use. I love learning new methods and finding solutions, allowing me to expand my knowledge of the pipeline and skills as a result.
I definitely plan to continue making more realistic and organic environments. This is something I’m really passionate about and want to keep exploring further.
I’m currently beginning to work on a fantasy RPG inspired scene, along with a Bioshock inspired scene, as well as an abandoned building too, all part of my final year assignments at Staffordshire University. I’ll be sharing some more on these very shortly!
I’d like to add that the environment in its current position certainly wouldn’t be where it is now without implementing the feedback from my peers and iterating the scene many times over. I’d also like to thank one of my friends Ben McDonald for giving me countless feedback on this project, as well as our game dev Discord server ran by students.
If you’d like to check out more of my work or connect with me on any social platform, feel free to follow me or drop me a message!