Let me thank Javier Benitez for this fantastic opportunity to share my One Two scene in detail on his blog. My name is Daan Hesterman, and I am a 22-year-old junior 3D artist working in the AAA game industry, with close to a year of experience in game development. I have a degree in game art and have been modeling for around 5 years.
I started this project on November 2nd of 2018, and finished it on November 13th of 2020. I want to mention that I didn't track my progress throughout the scene's development, so some images can be slightly different from what you might expect when reading. These are just random screenshots I took throughout the development just for myself.
I worked on the scene when I was home and done with work for my study.
In this article, I want to tell you about the development of my scene.
This article isn't meant as a tutorial or guide but rather a closer look at the development from start to finish. I chose this because there are already so many tutorial-like articles out there, and I wanted to do something different for my own article.
I want this article to give the reader the entire story of the development of my scene, from beginning to end, while also sharing my thought and opinions on the workflows I used.
Throughout this article, I will touch on what I thought worked well for the development of my scene and what didn't.
So, in the end, I want the reader to know how to properly start working on their own scene if they are interested in making one.
When I started, the scene was very different from what it has become now. I had the idea to make a Chernobyl-style environment featuring an abandoned rehabilitation center.
Around 2-3 months into the development, I completely scrapped the scene I had built. This was because of the lack of creativity I was able to put into it. I felt very restricted working in this style. I had a lot of ideas that simply could not be applied to this scene, as otherwise, it would look out of place compared to the rest of the environment.
After I abandoned the scene, I decided to work some more on a door asset that I was also working on. I didn't really know what style I wanted the door to be in, but after a couple of days of reworking it, I finally came up with something I liked a lot. That door was the doorway to a whole new style I knew I had to explore further. This is what kicked off the development of my One Two scene.
I would like to state that you don't need a completely finished idea when you start working on a scene. Making a scene is a big-time investment and, of course, a scary thing if it is your first time doing so, but don't let that scare you away. Good ideas will come in time. You just have to start and go from there. I know that now because I have experienced that first hand.
Where to Begin with a Big Vision in Mind.
My idea was to make a big overgrown mansion. I started with a block-out of a room which I was planning on expanding later on. I started with the area that had the broken wall in it. I wanted this to look as good as I possibly could.
I had never done any interiors, so I was kinda anxious to get started. That's why I first wanted to make sure I could make something I was happy with before moving on to the rest of the mansion. I saw this as an area I could experiment with. What I did next was I focused entirely on this one area until it was finished. When I got there, a mistake I had made began to reveal itself. I had failed to realize how much time it would take me to make the rest of the scene as detailed as this spot.
What I should have done here was remove detail until I felt comfortable moving on. What I did instead was I moved on without changing a thing. I wanted to make the scene as impressive as possible, so removing detail wasn't something I felt comfortable doing.
I am still a big fan of the amount of detail I was able to put into this project, but if I could give my previous self some advice, I would suggest cutting the scene in half at least once while keeping the amount of detail. This would have helped a lot with not losing motivation, and it also makes the chance of finishing a scene way more likely.
I want to give this advice very early because this is the first big mistake you can make while making a scene. Making something too big or too challenging can quickly kill your motivation. That's why I highly recommend taking a realistic look at what you want and what your skillset/experience is. This way, you may be feeling a little disappointed in the beginning, but it is entirely worth it when you end up with a finished scene instead of never getting a finished product.
Also, at this point, I ended up scrapping the idea to make the scene from the outside.
This time I wanted to take a different approach. Instead of focusing only on one area, I decided to first place all the main objects like walls and furniture into the scene (only the lower floor cause otherwise it would still be too overwhelming) and add all the more detailed stuff like vegetation and smaller models later.
Throughout the process of making the lower floor, I started to get more and more ideas to add to the scene. I felt like I was just making an ordinary abandoned mansion even though that was not what I wanted. I finished the lower floor (up to a certain degree) and started adding some more creative and original ideas to the scene.
This is where I began to feel like the scene became more and more like I wanted it to be. I didn't use many references to begin with, but at this point, I abandoned almost all references and started making the designs directly from my imagination.
This has its pros and cons. The good thing about it was that I was completely free in how I wanted the assets to look and wasn't restricted by any reference or art style I had to follow. But the tricky thing about it was that it took way more time and effort because I had to design it myself.
Designing the assets was kinda hard for me in the beginning, but it got easier over time.
The scene started to fill itself with cool ideas I had, like the design of the balcony, which is located above the door. The lower floor looked more and more like a finished product.
Even though I liked what I had so far, I knew it still didn't look as good as I wanted it to be, and after spending some weeks focusing on color-grading, I just couldn't get what I wanted. I also still was struggling with the thought of how much work still had to be done. This is where things started to kinda fall apart.
At this point, I had almost spent a year working on this project. I started to feel like I maybe needed to scrap the upper floor. The problem there was that the basics of the scene were already built for a two-story building. Scraping the upper floor would make it look weird and out of place. With that not being an option, I kinda lost motivation.
After a couple of months of not working on the scene, I finally decided to go back to it. After this break, I felt much better and less stressed. I had about a year until I finished my study, and that would also be the time I gave myself to finish the scene. After the study, I wanted to search for a job directly with my scene as my big portfolio piece. Without thinking too much about the deadline, I started working again.
Here I would like to talk a bit more about the importance of taking breaks from a project. Especially when you're working on something that is very big or takes a lot of energy out of you. I found that only 2-3 weeks weren't enough to clear my mind in terms of the project. In my case, I had to take a break for at least a month to really start fresh again. Of course, this time depends on the project and the person.
I do need to admit that I really struggled with taking breaks. The hardest part for me was leaving things unfinished while they weren't like I wanted them to be. I kept feeling the urge to keep going back and hopefully fix what I wasn't happy about. Now, because of my experience, I found that some things only can be fixed by having a fresh look at them. I concluded that breaks are needed even though you want to do the opposite and fix that thing you aren't happy about.
Looking back, now I can also see that this cycle was repeating itself throughout the development of my scene. I was so focused on making a particular area as perfect as possible that I totally forgot about the rest of the scene, which still was utterly empty. This was a real motivation killer for me because it felt like I had already spent so much time and energy on this project while getting almost nothing done except for that small area.
Next, I ignored what I already had made and started focusing entirely on filling up the scene with assets so that I only had to add detail and fix the color grading later when it was all done.
Back on track again. This was also where something very special started to happen.
While I was looking at scenes from other artists, I found something that was missing for me. All the scenes had either no story behind them or the one told isn't too big. I wanted to make a change in that and decided to go in a fully story-based direction. I quickly came up with some ideas I very much liked and added them to the scene.
While slowly but steadily filling up the scene with more and more assets like walls and props, I started to get more and more into the story of my scene. Because I only later came up with the idea of adding a story to the scene most of the story-related content is located on the upper floor.
Months of slow but steady development later, I found myself comping up with more and more original designs I didn't even know I could make. I started to get more confident and consistent in this style that I really loved. This was a massive deal for me as I began to see everything finally coming together as a whole.
It was mind-blowing to see what I had made during those months of hard work, as now it felt like I finally hit that level of quality that I was seeking from the very first day of development.
At this point, my scene was almost completely done. The last thing that had to be done was to fill up one more empty spot in the background. While I was looking around the scene, I found it to have a little bit too much of the same vibe. Everything was kinda the same color scheme, and for this last spot, I wanted to change that up while still staying true to the art style.
I ended up making a little sand garden that serves as a bridge between the two story-driven places. I was thrilled with this idea as it is different from the other stuff but still able to fit in.
I would like to point out here that I found it very refreshing to step out of my comfort zone and see what else I could make in this style. I can highly recommend that you take a clear look at your work and see if you might want to change something to create some more diversity once in a while.
Throughout this final phase, I started to remaster some stuff that I thought looked outdated. I edited some assets, textures, and materials, etc. I also added more detail to get everything to the same level of quality. When everything was checked, I started to render shots of the scene. Each render I later edited in photoshop to give it a crisper look. When everything was done, I posted it on Artstation.
.01 - Software.
For my workflow, I used the one I usually use for making assets. I used Maya to first make the low-poly, and afterward, I made the high-poly. After being done with the low poly, I sent it to Zbrush to create a high-poly from it. I do this by DynaMeshing it and then using the Deformation tool to Polish it. This will give the mesh nice smooth edges.
After that, I start with unwrapping and then applying the materials to it. I bake the high-poly into the low-poly through Marmoset Toolbag, which has amazing tools to fix additional bugs you may encounter after the bake. Most of the time, after the baking process, I drag the normal map into photoshop to fix some minor issues.
When that is done, I jump into Substance Painter.
Sometimes some assets require a little more attention. This was mainly on the floors and walls. To get the material needed for those assets is where Substance Designer comes into play. In this program, I made some custom materials that I couldn't get by simply opening Painter. There are many other options to get materials like marketplaces, but as always, I didn't want to make use of any outsource work from other artists.
The vegetation is primarily painted in photoshop (I think I did one plant through sculpting). For example, the grass is made by drawing lines and adding some grunge maps over it. After that, I dragged the base-color texture into Designer and made the rest of the PBR maps. I did this by using the information from the base color and adding level and grunge notes to it until I got a result that looked right.
I also sometimes used Adobe Illustrator for some complex/detailed shapes to later use as brushes in Painter or Designer. After finishing the textures in Painter, I exported them to Unreal Engine to see how they looked in my scene. This was a crucial step as sometimes some colors were too dark or too light with my current lighting settings. Based on this, I edited it until it looked correct. I already imported the mesh into the engine multiple times throughout this process to check for scale and different bugs. This workflow worked very well and helped me to deliver the quality I wanted.
Shapes made in Illustrator:
You can see the three shapes used in the wall asset shown here:
A material I made in Substance Designer:
.02 - Design.
The way I made the designs for the assets is kinda the opposite of the industry standard.
Typically you would first design your asset by making a sketch and a block-out.
I took a different approach here cause I wanted to stay true to how I designed the first asset I made in this art style: the door. I didn't really know how to reproduce this style because I only made one asset like it. I decided to just follow the steps that I took before while making the door and hoped to get an asset in the same kind of style out of it. This ended up working really well.
Here's a bit of a closer look at how I designed the assets. First, I started by figuring out the scale/space needed. I started off in Maya with a cube to mark the size the prop needed to be. Then I imported that cube into Unreal Engine to check the size compared to the rest of the scene. I kept repeating this and adjusting the size in Maya till it took up the right amount of space in the engine.
Then using this cube as a reference for the size, I started off with a new cube/cylinder and reformed it until I had a shape I liked. Keeping in mind the prop I was trying to make, I continued until the whole space was filled. Throughout the process of making the shapes, I kept a close look at how they would fit together and look as a whole. This was very important cause otherwise, I would have ended up with shapes that didn't match/connect with the other forms or the overall look of the asset.
In short, my workflow for designing the assets is making different shapes that come to mind and putting them all together until they look right. It sounds kinda directionless, but I always ended up with something I really liked.
Using these workflows together let me do everything I wanted without any restrictions. I was delighted to see how rather easy it was to go from a workflow I never used to a fully workable one that worked perfectly for what I wanted.
I also started experimenting with other work I made and fusing it together. In the picture below, you can see some wall parts are filled up with a drawing I once made.
The wall with picture parts in it:
Let's talk about Unreal Engine. Let me start by saying that it didn't take long for me to fall in love with this program. Before this project, I only used Unity, but I wanted to try Unreal for this one.
While finding it kinda hard the first couple of weeks, it really ended up being completely worth it. From using the shader graph to making use of volumetric lighting/fog and screen-space reflection. It is all very user-friendly and way easier to set up than I initially thought.
Lighting, this was something I struggled with a lot. Because I had chosen to make an indoor scene instead of an outdoor one. I couldn't just add a Directional-Light (sun) and make it look good. Also, adding a Point-Light (lightbulb) didn't correctly light up the whole room.
I tried bake lighting, the most commonly used way of using lights in a game-ready scene, but I couldn't get it to look like I wanted. I used a lot of real-time lights, which helped me deliver exactly what I wanted in the end. I also made air/dust particles and air-fire particles.
Because this is an old abandoned mansion, I decided I wanted to create some cracks in the floor, and some parts were completely ripped out. Doing this with mesh would take a very long time and also would be really inefficient if you want to change it later on.
After some research, I found a way to turn a plane into something with depth by using the shader graph. After some editing of a texture, I already made, I ended up with a good-looking result that is also game-ready. I only need 1 plane for this to work.
I am very happy that I chose to try unreal as it is now my main engine to go to.
Once you have a couple of weeks of experience in the engine you should feel right at home. I even ended up doing things like creating this material with depth that I first thought was too complicated but ended up being very easy.
First of all, if you made it this far, I would like to thank you very much for reading the whole article! It took a while to find all the images, as it is such a long time ago from the start, but it was an enjoyable trip down memory lane.
I am very passionate about my scene and the game I am working on. I hope that I got that through to you, the reader. I also hope that I inspired you to create your own environment and made you hyped to start creating your own world.
After this crazy development-journey of ups and downs, I finally published the scene I worked so hard and long on. As you probably could've guessed by now, this wasn't really the best way of making a scene. That said, I have learned so much throughout these two years, not only about art and workflows but also about myself.
I would like to end this topic with giving the reading and also myself the advice to first plan the amount of detail you want to put into your work and maybe make a (very) small environment just to see how long it will take for you to make it from start to finish before beginning working on the final product.
If I had known it would take such a long time, I would have made it smaller even though I am secretly kinda happy I didn't. Also, I want to mention that there is no shame in starting over and that the first idea you come up with doesn't have to be a big amazing thing. Speaking from experience, the best ideas will come in time while already working on a project. All very cliché, but it really is the way to building your dream scene.
The start of creating a world is opposed to where you end up with. You start with a blank canvas without any direction to lead you in. While otherwise, you would have your reference, now you only have your mind, and that's all you can rely upon up until the very end.
Official renders of the scene: https://www.artstation.com/artwork/6aDYJ6
Story and Game
As I mentioned earlier, I really wanted to focus on the story as that is something that, in my opinion, turns a scene into a believable world. After I released the scene I wanted to do more with the story I created, I came up with the idea of making a game. I feel like I only just scratched the surface with this style which is why I want to fully commit to making a game around it.
The world from the game will be entirely in the style of the scene and will tell an intriguing story while you are finding out what is going on in this universe. The game will be completely FREE and won't have any additional cost. I am planning on releasing it in 2022. It will be a pc exclusive.
If you are interested in me, my scene, or my game, feel free to join my discord:
Here you can ask me everything about anything, work-related stuff, feedback, or something entirely different, for example, about my personal life. We have a tiny community where we talk about art, play games, and just chat in general.
One of the priorities of the server is to keep people up-to-date with my game. There is a channel that is dedicated to my game and the progress I am making. I am updating this from time to time and will give announcements when I hit a milestone, like being finished with all backgrounds, for example.
I would love to have you, and I am looking forward to your art and/or portfolio. Hope to see you soon!
One of the many backgrounds for my game (Still work in progress, not finished yet):
If you wish to stay up to date with the latest articles, and speak with the people writing these articles, don't hesitate to join our discord!