ROOM OF THE GRAIL - single shot cinematic, focus on crystals and storytelling

Elias Tsirides from Splash Damage talks to us about his submission for the Artstation Challenge and how he crafted the single shot cinematic. He also details his unique process of sculpting the geode, as well as his approach on trim sheets, foliage and artistic fundamentals.

Introduction

Hi guys, my name is Elias Tsirides and I am a 3D environment/level artist at Splash Damage in London, working on the Gears Tactics game. This is the first job I got in the industry after building my portfolio by teaching myself the craft based on the amazing resources that other artists put out there for us. I come from Greece, where I graduated from the school of history and archaeology but decided to get into game art because of the original Tomb Raider level editor released all the way back in 2000. I would spend countless hours learning how to and eventually building my own levels. It took a very long time since that point and a very long journey but I eventually managed to become what I knew then that I wanted to be.

Artstation Challenge - The Room of The Grail

Artstation challenges always seemed fun and exciting and I always wanted to participate in one of them. They also seemed daunting though because of the time management factor but that is what makes a challenge a challenge. So when the Legend of King Arthur came along and even if that period was not a super chill one, I decide to go all the way in and try out to push myself with time management. I browsed both the environment design concepts but also the keyframe ones. Daniela De Anda's concept just clicked with me and I decided to go with that.

Colour wise, the original concept reminds more of a desert type of location, at least that's how it felt in my head and based on Daniela's inspiration by locations such as Petra in Jordan. While doing some reading on the legend I decided that the image could also fit the story of Galahad entering the room of the Grail, and since there are a few versions of where that could be, I decided to go with the UK and drew inspiration from the cathedral quarry in the lake district. Another aspect with this concept was that since it was a keyframe challenge submission it did not come with a breakdown of the scene and assets, so I would have to adapt and figure out solutions myself. So at the end, I tried to stay true to the main composition and skeleton of the concept, which was that cave like environment, the stairs leading up to the "treasure" through pillar like shapes and most importantly the light. So essentially I kept that core and applied a different skin to it.

Building the Scene

Everybody will always tell you to start with a blockout and that is extremely useful. I will admit that when I started out I wanted to skip that part. Who likes to see grey cubes besides the level designers right? But in my latest projects I applied it and saw how key it is. That is what I did in my submission as well. As I mentioned the keyframe concepts don't come with a breakdown so it was essential to set up a blockout with the simplest shapes possible to see how I could solve the lighting and atmosphere, what arrangement of the structures and light would result to it. So while building the blockout I was also setting up the light adjusting back and forth as necessary.

Once I was happy, I started with figuring out the rock modules, since they are one the most important ingredients in this scene. In the past I might have created a whole set directly to game ready models and use it to dress the scene and this was not a good choice. So, this time I didn't, because many times what looks nice as a game ready model might not easily work in your scene. Hence, I sculpted some initial shapes in Zbrush, trying to keep their form in a way that they could work placed next to each other and brought them decimated and textureless in the editor. I used them to replace the blockout cubes and when I saw that they could indeed work together I revisited Zbrush, refining the details and producing low poly models to uv and texture.

While texturing I made sure to not overdetail the maps since they would need to be used to cover large surfaces and too much detail could result in too much noise. After that setup was finished I went on with the architectural parts which also had rough blockouts in during the blockout stage and made them using a trim texture I built for that purpose. Trim textures are extremely useful since they can guarantee consistency and ease in tweaks and changes down the line, plus they significantly drop production time and give you more space for adding new assets which you might not have accounted for from the start. The statues are assets that have unique bakes though.

Finally I made some foliage pieces based on the references from the cathedral quarry and keeping them all in one atlas. Atlases are also extremely useful the same way trim textures are. Initially I wasn't planning on having the scene quite as overgrown, but while testing them out with the foliage brush tool in unreal, I saw that overgrown works for this scene. I felt it gives an extra touch of fairy tale and hence I went with it.

Guiding the Viewers Eye

This is a scene about the grail and is the object that requires the most attention, therefore I tried to keep things more balanced and toned down colourwise, trying not to make them attract too much attention and also made the grail the only asset that uses metal and a strong colour like red. This was important because it is also a very small asset in this cave. If for example the architecture parts on the background stood out more, they could take away from it. I also made sure it is lit as much as possible and also set up the light in a way that it creates a path towards it, leading to its area. In retrospect, I believe I could have even enhanced this further, but there are always lessons learned along the way.

Single Shot Cinematic

To showcase the level I decide to set up a fly-through cutscene. I really like showcases where the final video is composed of smaller cutscenes but I wanted to make a continuous single shot to convey the experience of a player entering this space and walking towards the point of interest while looking around as well. I feel it also helps establish the sense of the space of the level a bit more clearly. Of course it exits somewhere where the player would not really be able to go but I believe you also need a smooth in and a smooth out in these types of cutscenes.

It can be a little bit taxing to try and set shots like that because the camera has to follow a long path and bugs can easily happen, like weird spins of the camera trying to compensate for the turns and create the curves of the path in between the keys that you set. Since I am not a cinematics artist I adjusted the keys and transforms of my camera or sometimes the time it takes to reach from one point to other accordingly so I can avoid that. Adjusting the time also helped because you probably want to allow enough time in certain points to help people read the scene.

Important Skills for an Environment Artist
When creating environments, I would say there are many important skills an artist should have or explore. There is always of course the technical side of things, which means the knowledge and command of software. Learning these is always an ongoing situation but the more you explore the limits and applications of software you can come up with different combinations and approaches to provide solutions for yourself.

There is not just one way to do things and by exploring you find what fits you more and is more fun for you. An example in this scene for me was to use the Unreal's foliage brush tool in order to create the crystals inside the grail / meteorite stone. I could have done it manually or I could have used tools in Zbrush for example, but I was happy to find out that I could use unreal for that. I used the foliage brush to populate the interior of a proxy of the meteorite shell with the crystals and then export that as a mesh to use as a part of my high poly for the grail.

Besides the technical parts though I would say that the most important thing is to develop our understanding of composition and how to use the arrangements of shapes and the relationship between colours to create readable and appealing scenes. Knowing how to create focus, establish leading lines and clear paths. This is something I constantly try to learn, improve and figure out how to use myself. It is important to be observant and take the time to analyse the work of other people that do this successfully in games or their own projects. Take the time to figure out what makes things work and educate ourselves in the elements of aesthetics that result in that, like composition, proportions, colour and such.

Also it is important to take moments and treat our environments are real places and hence always ask ourselves throughout the process of creating them questions like could this really happen in the universe of my environment, could this type of object exist in this specific corner of my map, would someone be able to use this, walk there, built there, would plants grow here and many other things of course along those lines. Because it is very easy to just treat your assets as 3D models to be placed and forget about what they really represent in the "real" world.

Challenges & Advice

The biggest challenge I encountered during building this scene was of course time management. More specifically this means that I had to reduce the scope of things along the way and that happened mostly by pushing myself to say something is good enough and move on to the next thing that needs attention. Sometimes it feels like we can never be one hundred percent happy with what we make and we try to improve and improve upon it. But it is important to see the bigger picture and not dwell on just one thing and exhaust or disappoint ourselves. What was also very challenging was to create things out of my comfort zone.

For example I have never done statues before and they felt intimidating to do. But I tried to elaborate on the basic understanding of Marvelous Designer that I had just started exploring before the challenge and use it to help me make them, and even if I was not fully happy with them I also had to say they are good enough at some point and move on because otherwise you loose your focus.

I believe that at the end of the day it all comes down to reestablishing your focus on what you want to achieve. By that I mean it is easy to get lost in notions like wining or loosing, being good or bad, but at the end with everything new you build, everything new you explore and learn you always win and you always become better, even if you don't see it right away.