Kay Tang tells us what inspired the Breakfast for Giants project and how every detail in the scene was carefully sculpted and textured using Blender and ZBrush.
In case you missed it
You may find these articles interesting
My name is Kay Tang and I am a Concept Artist based in the UK. I have previously worked as an intern at Opus Artz, and as an Environment Concept Artist at Grinding Gear Games after graduating. At these studios, I was able to contribute to a variety of projects including The Elder Scrolls: Legends and Path of Exile 2. At the moment I am just focused on my portfolio and personal work. I studied Digital Animation at the University of Hertfordshire, where I learned Photoshop as well as 3D programs such as Maya, ZBrush, and studied other areas of digital art.
When I was younger I would watch my older siblings play video games such as The Elder Scrolls III: Morrowind or Monkey Island, and then got into games myself later. I still love the feeling of some of those games and come back to them for inspiration. I also started drawing a lot when I was young, and when I was around 14 I slowly began to discover the world of concept art. My sister gave me an art book of Alan Lee's pencil drawings for The Lord of the Rings, one of my favorite film series as a present. Although it is not very common to use pencil for concept art nowadays, I still look through that book all the time because I love how explorative the sketches feel, and I enjoy seeing the sheer volume of work and persistence towards the vision for the trilogy.
After receiving this book, I filled sketchbook after sketchbook with pencil and ink drawings, just observing things outside in different places or images I found online that I thought were interesting. I started to do a little bit of digital art, first in MS Paint, then in Paint Tool Sai. In the beginning, I would either just use a mouse or this little tablet that my family bought to write texts in Chinese, it was just a few inches big, which did not work very well at all but I had fun at the time. Around this time I began to discover more artists online while searching for tutorials and also pored over 2 copies of the ImagineFX magazine that I found. I was especially inspired by Theo Prins' art that was featured in one of the issues, and his tutorial showing his work process that was on a CD at the back of the magazine.
Here is one of my sketchbooks from around 2013:
Thanks to all these sources of inspiration, I ended up being particularly drawn towards a fantasy-type setting in my work and wanted to create my own worlds, just like the ones that captured my imagination when I was younger. I feel that it was just in the last couple of years that I really honed in my focus towards design, although I think the years I spent doing lots of pencil sketching and painting were really beneficial to developing my personal taste and observational skills.
Creating Breakfast for Giants
My recent prop series is a continuation of a previous project, where I created an environment piece of humans administering traditional Chinese medicine to a giant.
I would say that a big inspiration for me is when I go back to my cultural upbringing. My family is from Hong Kong, and I was raised in the UK. I grew up experiencing both cultures, and I find that I do see a lot of art based on Chinese culture, but I wanted to show certain aspects of the culture that are not depicted as much.
In my more recent project Breakfast for Giants, I wanted to show some specific food that I grew up with, and my main inspiration for the series were these congee shops you can find in HK. These kinds of shops open up at the crack of dawn, and you can see other people grabbing big portions of congee, noodles, and other food before heading off for their busy day. A lot of the design is also inspired by my family's old takeaway business, as they had to handle and cook huge amounts of ingredients to serve all the customers.
Speaking of the giants, I thought that design-wise it would be very fun to play with scale in this world, and it acts as a central motif of the whole world and the project. I think having the main motif that informs all of the design decisions you make can be really helpful for making even mundane objects more interesting. The presence of the giants also creates a lot of satisfying problems to solve. I have a pretty keen interest in ancient history, which I studied as a main subject during my last years of school. I find that a lot of ancient civilizations have very interesting and advanced solutions for the problems, and I love reading about them to figure out how to create certain structures and mechanisms in my work.
The main program that I use for my work is Blender, and in this series, I first did a basic blockout of the breakfast trolley and the congee pot props. The third prop, the table with a lift system, was created last and had a much faster workflow because I had already created some assets from the previous two props which I dropped in straight away during the blockout stage. I treat this stage a bit like sketching, just trying to create the basic look and proportions of what I am envisioning, and rather than making a lot of design iterations for each prop, I put my energy towards one version and aim to get it right.
After blocking out the first two props I started to go more in detail, initially getting in some placeholder textures either from Quixel or just flat shaders, and designing with the intention of creating something that looks like it was built and used by people through getting rid of the sharp 3D feeling by refining the geometry and textures, adding layers of narrative. I make a lot of smaller props to add to the scene and bring it to life, and I made sure to keep it specific again with particular foods, Chinese pots, and cups, cloth with signage to break things up.
I do my best to try and never create the same asset twice, so for example, once I made a plank of wood in a specific shape for a prop, e.g. cylindrical, rectangular, etc. I can reuse it again and again for the same prop or for my future props. I also have another Blender file open while working, and there I organize my own personal kitbash library of assets per project. I find this particularly useful for details such as the planks, rope, or cloth.
As an example of how I create my assets, for the wooden planks, I took geometry from Blender to ZBrush to create wear. I would then dynamesh and/or subdivide it, sculpt wear on the edges and corners, and optimize it as much as I could using Decimation Master before exporting it back to Blender. It seems like a very subtle detail that nobody would notice but I think in the end when all the planks are like this, it makes a big difference to creating a worn and used feeling and tricks the eye into thinking there is more detail. I try to vary the amount of wear I sculpt onto all the planks, some have more dents and damage and the others like this one have less.
Next, I automatically UV the plank and texture it by using the Texture Paint mode in Blender, using a photo with a few different wooden planks as a stencil. I find this to be a really cool way of texturing quickly and getting photorealistic detail in the model. I make sure to use the same photo for all the planks to keep it consistent in terms of hue, saturation, and values. I then create a shader using the texture map I just made using this method, adjust parameters such as specular/roughness, and add some color adjustments. I usually do this after placing the asset back into the main model, to make sure it fits in with the lighting and rest of the prop.
For some of the assets rather than using texture paint, I add texture by projecting a photo onto it, for example, I do it to bowls or pots where the curved surface makes the item a bit more difficult to paint the texture on it uniformly.
It may seem like a tedious or long way of working for concept art, and individually small assets like this do not look very remarkable. But I think that in the long run working this way adds believability and saves a lot of time as I can gradually build my own personal library of assets. For portfolio work, I also think that it can be good to invest a little more time into things like this.
The same goes for the areas where I have used rope in the props. I spent quite some time at the beginning adjusting curves a lot to get the rope feeling right, but now I have a lot of variations that I can now drag and drop. Adding in the rope is one of my favorite ways to get in layers of shapes that have a more organic and natural feeling to them, in contrast to all of the harder shapes in the model.
For figuring out mechanisms such as the elevator, I researched diagrams and real-life examples of older, more primitive elevators, and also some explanatory videos that go in-depth describing how they work. I think videos are a really great resource in general for finding references and information that does not pop up elsewhere. I really try to make sure that the mechanism looks like it could actually work while making considerations regarding how it would fit into this world.
I love to pay attention to some of the details in the examples shown on my reference board here: in one of the images on the right, you can see I have directly taken the way the rope is loosely wrapped around the bar and used it in my model.
The same goes for the trolley, as I looked at various images of wagons or carts and observed how the wheels are built, how the leaf spring mechanism works, what reinforcements or attachments are used.
A lot of the detail here behind the wheel ends up just being glints of shapes in the final render, but I think it was worth it even to get those small indications of a working wheel structure.
I also use a bit of 3DCoat in my work, specifically using the Stencil function for the areas with patterns. I create an alpha of the pattern first in Photoshop, then after stenciling out the geometry in 3DCoat using the Cut Off tool, I take the geometry into ZBrush to sculpt in some more detail and wear before putting it into the Blender scene. I find this a great way to contrast against the other more primitive shapes in the props.
When it comes to detail, I like to have areas of rest or repetitive shapes, versus smaller areas of high detail density. The areas of high detail often look better in specific places such as where two objects connect, and I also do my best to make it believable that those two objects are connected. I also like to consider where the most natural place is for detail if a real human were to interact with it, kind of like how on your own desk or bedroom, you will absentmindedly place random objects somewhere convenient or close to you. I do not think there is such a thing as "too much detail", rather, it is the placement and distribution of it that is important.
For example, here I kept the coiled rope around the ends of the wooden bar and left the middle area empty, as this also creates a layering of materials and reinforcement of the structure. I also placed tongs next to the ladder where I thought a human might put it while they are in a hurry. I rotated them randomly to make it feel more natural.
I also hung things like steamer baskets or gourds on as many places as I could on the trolley because this prop would be moving, the humans would not be putting stuff on the floor so there is a limited amount of space to place objects in. I considered that for this reason, they would probably just hang stuff everywhere like on the corner posts, and because the humans would be working hard I added a compartment for them to have tea and human-sized snacks at the front, and a foldable bench.
For smaller ideas and details such as this, I find that a lot of the inspiration for these comes from just observing things in the real world keenly, making a mental note of how people affect and inhabit their environment.
For the lighting, I keep it very simple. In the congee pot prop, I just used an HDRI and 2 sunlights, and the other two props were similar with about 2 – 4 sunlights. I try to have 1 quite strongly lit plane of the object, and then have some other areas in shadow or add some weakly lit areas to make it pop out more. As you can see in this image, the HDRI I used was an interior with lots of wood, with a bright light source that I could utilize as a natural-looking rim light.
For the elevator table, you can see that again I have just 3 sunlights, and a couple of weaker lights to show the stairs (I made sure to put a lantern there so it makes sense). In Blender you can place the sunlight wherever you want, it is just the angle that matters.
The final angles are rendered in Blender Cycles, and towards the end of the project, I was using E-Cycles, which is an add-on you can get that makes rendering much faster. I rendered out 8 different passes to achieve the final look: the main render, Ambient Occlusion, Diffuse Direct pass, Shadow, Clown pass for accurate masking, as well as 2 different passes to add edge and leak wear on the geometry and a Dust pass. These last 3 were separate materials that I set on material override in the render layer properties.
The last 2% of the concept were done in Photoshop, where I then made some final adjustments and compiled the render passes.
I start out the post-processing by setting the Ambient Occlusion to overlay or multiply, and creating a Color Balance layer and a Blue fill layer that is adjusted under layer style so that it lightens up just the shadow areas.
Then I take the edge, leak and dust render passes, set them all to lighten, mask it all out and then paint/mask in the areas where I want to have wear and dust.
After this, I take another photo of a variety of wooden planks (different from the one I used to texture so there is more variation in the grain patterns) and wherever there is an area of the render that looks like it needs improvement, I warp a plank of wood onto it, mask it by selecting the object from the Clown pass with Magic Wand tool, and adjust the color and fill until it looks incorporated into the scene. This is done in the Normal Layer mode, I find the detail does not quite transfer with any other layer modes. I make sure that if there is a shadow falling on the wood, I mask it away as well.
Next, I take a grunge dirt texture and, in Multiply mode, place it on the rope, again masking it cleanly using the Clown pass and using the Brush tool to mask away some areas so the grunge is placed more towards the connection points.
And finally, I just make some very subtle adjustments with paint, making some edges a little bit more painterly, adding in some more wear again. If necessary, I will make subtle adjustments with the Diffuse Direct and Shadow pass, to make the forms pop a bit more.
I try to keep a structured workflow when it comes to the 2D post-render stage so I can emulate the same final look for all of the final angles and props, and normally it just takes a couple of hours to post-process each render in this manner.
As soon as I felt satisfied and the final images were all done and completely processed, I organized them on sheets and added some writing and my seal, keeping the same template and layout for all the sheets.
I think the main challenge for these props was reaching a point where I was satisfied with both visual and the technical part. It took a lot of figuring out how to make the props look believably worn and used, how to tell a story using just 3 props, what I wanted to achieve in the final look. For this reason, the first two props took about 4 weeks each, and then the final table elevator prop took 2 weeks.
I experimented with a lot of new things to add and refine in my workflow, and I am definitely happy with what I learned and consolidated from the project, I feel like I could explore this world more and build on my personal asset library.
The advice I would give to beginning designers is to always be open to learning, take your time and enjoy yourself when creating portfolio work. Be proactive and seek feedback and support from other artists and friends who are also learning design. This way you can push each project to the max, really finishing it because no matter how long it takes it will give you the satisfaction and the feeling that you were honest in your vision. This pushes your learning further ahead as opposed to what you would achieve if you made lots of fast and unfinished pieces.
Also, look inwards when coming up with ideas and try to filter out the noise from all the art you see everywhere and think about what you want to say and put on the table. I believe everyone can draw inspiration from their individual experiences and background and can come up with very interesting and unique ideas!