I have been working at Axis Animation as an environment artist for little over a year since I graduated and had the awesome opportunity to work on several cinematic trailers for various games, some of them being featured at E3 this year; Deathloop, Gears 5, Outriders.
It’s been an amazing experience working alongside some of the best artists in the industry, and I am constantly learning tons of new techniques, tricks, and workflows that I have never thought of before that makes what I do way more efficient and faster than I ever was in the past.
Challenges When Working on Big Projects
The challenge depends on the project itself. Sometimes it would be modeling, texturing, and shading certain assets that I haven’t really done before. There were projects, where I had to do sub modeling for cars, and it isn’t something that I have done before, and it was pretty difficult at the start since cars are one of the hardest things to get right in terms of modeling. Now I can do them quite well, practice makes you perfect.
Other times, it would be the style itself. While we do tend to gravitate towards more realistic stuff, the majority of the time there are cases, where we have to do stylized environments and assets, and adapting to it can be difficult at times since I myself lean towards realism most of the time.
As for environment design, the aspects of the cinematics and my personal work are usually handled by amazing concept artists. The concepts are usually fleshed out by them, and all we need to do is to produce the assets from the concept and match it as much as we could. I haven’t really come up with original environments, but it is something that I am looking into and learning about lighting and composition of shots to make it more appealing to the eye.
Unfortunately, we are not allowed to share works in progress for most of the stuff, but some of them were approved to be shown, and they were featured by our director Sava Zivkovic in his behind-the-scenes for Outriders, which you can find here:
The process that is taken to produce them tends to be systematic. When the concepts are done by the art directors and concept artists, they are passed to us to replicate closely as possible.
Quite a lot of times, we would purchase a lot of assets from places like mega scan or evermotion as starting blocks because it saves us a lot of time. Some of the assets would work well and won’t really need to be fixed too much for the shots, however, more often than not, they would need some extra love to make it pop more or integrate more into the environment itself and on occasion, optimizing it for the pipeline.
For my personal projects like Warlock’s Hut, I was mainly doing them as tests for a game me and a friend are making. These were small tests that I was doing to try to find a style that looked appealing enough but also at the same time didn’t take too long to make as I am the only artist on the project, my friend being the coder.
I took the adorable cute concepts of JiaYing Liang, where the design was really fleshed out and looked really appealing, and did some experimentation to see what I could get.
I actually avoided doing sculpts in ZBrush for this since it takes a lot of time to sculpt everything and then make the low poly and create the assets that way. I just jumped straight to the low poly and made sure that the main shapes are there, took them inside of Substance Painter and relied heavily on the gradient filter to remap textures of grunge maps to stylistic color palettes.
One trick I did in Marmoset was that the grass appeared to be really dark no matter what I did. I cheated the daylight look by giving everything a bit of emission so that it doesn’t come out as too dark. All in all, I would say that it took me around 12 hours or so from start to finish to complete this little test of mine.
Adding the Color
I have to admit that I actually have quite a poor perception of color overall, which is emphasized a bit by the fact that I am slightly color blind. For this project, where I have a solid concept to work off from, it isn’t too hard since I can just sample, using the gradient remap filter inside of Substance, and grab the exact color, which the concept artists used for their concepts, so that it matches as closely as possible. A lot of the techniques that I apply to my personal work nowadays are bestowed on me by Sir Dima who has taught me a lot.
I always ask for lots of feedback from other people, people who are artists and also who aren’t for their point of view. That’s another way I go about tackling my weakness in this area when it comes to color palette choice as sometimes I simply can’t tell when colors are too similar.
For this piece, in particular, I mostly used a combination of fill layers with grunge maps, remapping the colors to an appealing color palette, where I sampled from concept art and then used a bunch of warp and blurs to achieve the results that you see. For pretty much the whole piece, it’s this same technique that’s been used alongside some blurred edge damage.
Biggest Challenges Overall
The biggest challenges that I face right now would be probably getting even faster at what I do as well as starting to look more into proper lighting as well as the ability to compose a shot nicely. I am of course still learning more about these fields, and while I haven’t overcome this hurdles completely, yet, I would say that practice makes perfect, and continuing doing personal projects outside of work is a really good way to improve even faster.
Thank you to 80.lv for giving me the opportunity to share my knowledge. I hope I explained everything adequately. If anyone reading this has any questions regarding anything mentioned, feel free to contact me and ask. I’d be more than willing to help out.