Very impressive article Jake! You are very talented.
nice article! i love seeing the breakdowns.
Former Ubisoft employee Miki Company Butler talks about building a big 3d multiplayer game Skara. In this post he talks about the choice of Unreal Engine 4, the most useful tools for environmental artists and gives advice on connecting your levels and game design.
About Miki Company Butler
My name is Miki Company Butler and I am a 3D environment artist from Ibiza, Spain.
Since I was a child I´ve always wanted to work in video games, I remember playing with my father´s Commodore, creating games with DIV studio, RPG maker and always trying to do stuff with my computer.
I´ve been working on game development since 2009 and started my career working for Ubisoft, I was there for 4 years and was involved in many of their titles. After that period I started the indie adventure with Skara. I’m working on this project with a very nice group of guys. We started out in 2013. Right now I am trying to make more time to get involved in other exciting projects.
The Story behind Skara
Skara is being developed since early 2013 with a very small team and not many resources at all. We created the studio from zero, even painted the walls and gathered all the desks and furniture. It was such an adventure and it´s incredible how the project has evolved with more or less a dozen people since 2013.
The inspiration for Skara can come from anywhere. As 3D artists we try to materialize the vision of our concept artists the best we can, but we try to put a bit of ourselves. In my case, I get a lot of inspiration from other environment artists, concept or traditional artists, and even places like the coves around the Balearic Islands, but inspiration can come at anytime, from anywhere.
The Choice of Unreal Engine 4
First of all, you can jump into the editor for free, and the code is open for anyone. Is something better out there?
Without doubt it was the right choice in my opinion. At the time we were in doubt whether to use Unity or Unreal, but UE4 was about to be released. We had worked 8 months in UDK and all the work we did, went into the trash. The fact that the project was so ambitious for such a small team, and that visually could compete with a AAA game, made us choose Unreal.
Unreal for Big Open Arenas
Unreal makes life much easier, while the editor it´s been improving, the size on the terrain you can work on it´s been augmented, landscape level of detail works so great etc.
When you´re working on a small map everything is much more controllable, in a big one you have to reutilize much more, without noticing the repetition.
Unreal has perfect tools for the creation of open spaces. Landscape and foliage tools are indispensable. In the end they are tools that have evolved from the previous unreal versions and they have been improving with every release.
At the beginning of the project we defined some master materials from we are going to create all the instances for all the materials that we applied on the game. There is always going to be extra needs, so in many cases we´ll need to create some specific materials for the desired result.
Also it is important to keep in mind the performance at the time we´re applying these materials, there will be expensive ones and cheap ones (more or less resources required) that we´ll have to apply with sense on our scenario. For example, it has no sense to apply a material with tessellation and many more instructions to an object that is going to be far away on the BG.
Regarding the VFX they are all created from zero inside Cascade. They´re all done by our wonderful VFX artist, Sergi Salvador!
Blueprints are the evolution of kismet and are very useful. It is true that you need a programmer-mind to take advantage of them, but as an artist you can add a lot of functionality to everything you are creating. Every department can take advantage of Blueprints.
Important Tools For Building Games
I´ve always used Maya as my main tool, I feel much more comfortable than with 3DS max. I´m also a great fanatic of the free software, so Blender it´s been my 3d swissknife tool for years. Xnormal is another indispensable tool for generating your maps from your Hi-res models, and also it´s free. With a HI-res sculpt, a bunch of good extracted maps, your paint software and a bit of work you´d be able to create a nice setup of textures. Nowadays there are tools that automatize this process so much, such as Substance Painter/Designer Quixel Suite etc.
If you´re working with landscapes and exteriors there are tools that are indispensable like Speedtree and World Machine.
Building Environments for Games
It depends on the company, or the project you´re working on and also, the number of artists that are involved. The fact of working in different levels it is very helpful at the time you distribute the work. Ideally, if you are more than one person is to work with different layers.
In our case, once the layout has been created with BSP and it´s been played and replayed 3 million times. What we add to it in most of the cases is just decoration. But it´s true that at the time we are dressing the map, the designer can come up with a new idea from the work we are doing. Then we could adapt some concrete areas, so they can interact with the gameplay. Sometimes, a lot of synergy can be created between the two departments.
In terms of gameplay, I think it´s better to guide the player with visual elements. I really enjoy games where it´s hard to get lost, and usually the player that is guided by the visuals has it clear where has to go. Using concrete marks like lighting, objects etc. Even if your game wants to be realistic, the player needs some of this feedback because at the end it´s still a game.
Advice for Newbies
Think from big to small, prototype everything, and when you have the big picture clear, go forward and start refining stuff.
When it comes to creating a scenario, you should think on situations and try to tell a story, don´t start thinking which objects you would like to create, (your artistic heart is trying to fool you again).
Keep it simple stupid. Try to simplify A LOT , work with realistic objectives and try to not reinvent the wheel, that means you work smarter not harder, if you can save a week of work buying something on the marketplace, do it. (if you are an indie of course).
A work is never finished, it just gets abandoned. When you lose perspective, go for a walk, grab some herbes dolçes, then come back with a fresh vision and you may find a solution for it.
Above all things. Enjoy the process!