World building with Gravity Sketch & VR

Introduction

In this tutorial I will explain what I think about as I create a world from scratch for any kind of intellectual property. This process will show you how I get from rough idea to a complete image that can be shown to an art director or creative team to kick start the look and feel of any production. This is the first step in a process that I like to use when tasked with creating different worlds and environments.

This tutorial will also include a little bit of VR and how I use it to help me create interesting and inspiring worlds.

For this project I want to combine the idea of windmills, Japanese architecture and waterwheels to create a fantasy world. The purpose of this painting is to kick start ideas and work as an initial concept to build on and inspire more paintings down the line.

Kaze Shrine - Inspired by windmills, Japanese architecture, intricate water wheels in China, and set in a fantasy setting.

Gather Reference

Arguably the most important step of the process is finding the right inspiration and images to kick start the process. I tend to spend a good amount of time during this step because the more images I can find and use to inspire me, the better the final image is going to be. Usually I will find three different topics that I can mix and match to create my initial idea to build from. The more unrelated the idea, the better. A key point in finding reference is figuring out how much the different inspiration points are reflected in the final idea. I usually organize the three images into big, medium and small. Big meaning it is the biggest influence and small meaning it’s the least influential.

The idea behind organizing your reference in this way is to establish a hierarchy so you can understand what parts to pull from your references. For the image that I did; the windmill is big, Japanese architecture is medium, water wheel is small.

Rough Sketch

I don't usually show anyone this first step because it's mainly just for me. After figuring out the initial idea during the reference phase, I block out an extremely rough idea of what I want the image to look like. Mostly intended for me to have something to aim for when I start building this in Gravity Sketch. When doing these rough sketches, I might go through five to ten different loose ideas to see if the idea in my head is worth moving forward with. I find that starting this way gives me something to anchor to when in VR; because if I didn't have it, I might just build something endlessly that ultimately might not work for the idea that I have.

When coming up with the initial idea, I really liked the main shape of the windmills and thought it would be cool if the architecture reflected more of a Japanese aesthetic.

Rough Blockout

Next Step is blocking out in VR. The program I use is Gravity Sketch. This is when I really start figuring out how this world is constructed. I particularly like using a virtual reality headset for this because I feel fully immersed into this world and can experience the space in first person view. Being able to place yourself into the world and see how it would feel like to be in a space like this is an incredible feeling and really gives you a strong understanding of the place you are building.

The main idea behind this step is creating compelling and interesting shapes. The key is to avoid getting too close in and detailing every single corner of this because we are just looking for the big read and initial impact.

When starting a piece like this, it can be hard to know where to start. Usually I will start from the central main object (a building in this example) and build out from there. Use this step to really explore and find shapes that work for you and your idea.

I thought it would be really cool if this whole windmill village was floating in the clouds and thought that the visual of the clouds kind of intersecting with the city would look really cool and interesting so I reflected that thought into the initial rough model.

Rough Color

After establishing a rough model of what I want, I bring the screen shot into Photoshop and begin to paint out the 3D feel of the shapes by adding color and light to establish rough lighting direction and material breaks ups.

When starting the painting process I try to balance the local values of each material as well as the light affecting them to make sure that the image doesn't get too muddy and hard to read.

Even though the model is figured out pretty well, I tend to move a lot of the elements around and change a lot of the shapes according to what I think will benefit the image the most. Throughout this painting I try not to be too precious with the initial placement of the buildings because I find that during the initial 3D process, it's very easy to lose track of the strong silhouette read and come out with images that have balance and readability issues.

Develop Shapes

The previous step was about getting the image out of the 3D phase; this step is more about getting stronger value groupings and setting the stage for the design phase. The reason I focus the first few steps on painting and composition is because I find that design is already very difficult on its own and fumbling with composition, color and light while doing design just causes things to be harder than it should be.

During this step I focus on creating more realistic cloud shapes and start dabbling in the architecture a little bit to get the ball rolling on the designs for this world.

Steps two to five are focused on trying to get the setting and space right for the image. The following steps are going to be more focused on design.

Develop Design

During this step I like to start designing single areas. It can be very overwhelming to start designing everything at once so I focus on one spot and grow out from there. In this case it’s the foreground walkway area. I really like how Japanese architecture uses a lot of wood joinery so I try to incorporate that into the different elements.

When thinking about different parts of your image. Try to think about who lives there and how they might build things. I tend to do a lot of research into the different inspirations that I start with so that I can make sure that I try to incorporate those ideas into my work as well to create a richer and full world.

In the next steps you will see that I ultimately end up changing the design but it was a good initial pass to set up the designs for the other sections. Don't be afraid to explore different ideas within the same painting. This painting changes a couple more times as I get further into the process. The main idea of a painting like this is to inspire more paintings down the line, rather than nail the design prompt in one go, so a little bit of evolution and change is very welcome here.

Refine Composition

Like I said in the previous steps, this painting is going to be moving around a lot as I progress through it. I wasn't too happy with the composition earlier and wanted the image to show off the architecture a little more so I added some elements on the left and right to help with that. I wanted the location to feel fuller and complex, similar to how the water wheel reference looks, so I added more levels and sections around the base of the main structure.

When thinking about what to add into your designs, I tend to think about what kinds of feelings I want to create if the person was inside of this world. For this image I wanted it to feel windy, I wanted the viewer to hear the wood creaking from the windmills turning, and overall I wanted to create the feeling of wonder through how these buildings and paths were floating above the clouds. Once I had those ideas in my head, it got a lot easier to come up with different things to add and move around to create those feelings that I wanted to convey.

The foreground railing was further designed from the previous step helping establish the look and feel for the rest of the walkways so I began working roughly on how the architecture might look (building on the right side.)

Refine Idea

Similar to the previous step, I felt that something wasn't quite right with the image that I was making. After a little bit of problem solving, I came to the conclusion that the main structure felt a bit small. I was sticking too closely to the initial proportions of the 3D model.

When designing in 3D it can be very easy to lose track of big silhouette read and forget about how things look in frame. This is the main reason that I don't like to take my 3D phase too far as I know that it will change drastically in the final image.

To make the main structure feel bigger, I expanded the building to crop more on top of the frame, creating a feeling that it's too big to fit into a frame. Little purposeful choices can go a long way when composing your image; graphic design and cinematography deal with these principles a lot.

Establish what kind of feeling you want your image to convey and make changes accordingly.

Finalize Design

Now that everything was placed where I wanted them to be, next was to finalize form language and design.

Similar to when designing the foreground element, I just focused on the building to the right for all the architecture details. Buildings that are in the same location tend to always follow the same form languages so establishing one building really helps to define how the other buildings will look.

I tend to try to figure out things one at a time in images like this because it can be very overwhelming to bounce around different parts of the image and design everything at once.

Finalize Image

This is probably the biggest quality jump out of all the steps but realistically it’s the least impactful step in the process. Since all of the leg work had been done earlier in establishing lighting, color, walkway design, and architecture design, it was pretty straight forward to just apply all of those things to the main structure and everything else.

The main idea when finalizing the image is to preserve all of the planning and work established in previous steps. It's very easy to lose track of your design and form language as you go through different parts of the painting but since it gets very organized in the beginning during the reference gathering process with establishing the big, medium and small influences, it gets a lot easier to stay on track.

During this step it’s up to you how far you want to take it but keep in mind that when creating initial images for world building. This image is meant to inspire dozens of paintings after it rather than be a painting that starts and stops with itself. Hopefully this image can be further dissected into smaller more detailed paintings to really show how a painting like this can be used in a production type of process in future tutorials.

When starting your own projects I would recommend starting with blue-sky images like this and working to more and more specific ideas, essentially working from big main ideas to small detailed ideas. It gives you a target to aim for and acts as a guiding light to your projects.

Top Tip 1 - Learn Foundations

Learning your foundations is probably the most important thing an artist can do, being able to understand perspective, color, light, anatomy, composition, etc. at a high level frees up your mind to focus on design solutions rather than thinking about the basics. The more you are thinking about perspective, the less you are thinking about design.

Top Tip 2 - Stay organized with your references

When world building, it is extremely rare to nail the design in one go. So once you have your first image change the proportions of the initial design influences and make another. The more organized you are in the reference phase, the easier it will be in the future to make completely different worlds using the same three design influences. Knowing where each element comes into play makes it easier to change for the next painting.