Oh boy, the most difficult thing is to choose the right concept to start a work, and you know why because you want to make all of them! But seriously, choosing a good concept is very important because you need to be sure what kind of skill you want to show besides doing a cool piece.
For this specific project, I wanted to have a nice character where I could really evolve and show a different range of skills. Like distinct materials and different techniques as well. Modelling (Organic/Hard Surface), texturing (Skin/Other materials), grooming (Hair/Fur), lighting and rendering.
Once the concept is chosen I started to play around in ZBrush to define the first shapes and find the correct silhouette. The concept shows a very strong guy and with some exaggerated anatomy making him even bigger.
I decided that I wanted to have a more natural approach to his anatomy. Something extraordinary but at the same time believable for the audience.
It’s very important to find references to the real world to follow, instead of other drawings or models. I got a bunch of older bodybuilders, some strong men and fat guys to understand how the muscles and the skin behave with this kind of body.
Also for the fantasy part, I followed some Gio Nakpill drawings to understand these values applied to the fantasy world in a very natural way.
For the face, it wasn't clear in the concept what specific race I wanted for him.
I pinned some key points to help me sculpt it, like the flat feline nose (Like Azog from The Hobbit), the long oriental eyes and the Orchish pointed ears. Breaking the symmetry of the face gave him a more of a natural flow, it allowed me to thrive on adding some small marks and wrinkles to create his expression and make the first detail pass for the organic part.
Time to model the accessories and put everything together to figure out the whole silhouette and balance the character. I took a few personal decisions to change some parts from the concept.
Taking out the spikes on the skull of his shoulder and building a completely new right arm design. Breaking the symmetry telling a different and more concrete story behind the character.
The Hammer is a special case here because it’s HUGE, and an awesome piece to study the mix of organic and hard surface modelling. I decided to leave it until the end to model because I wanted to give special attention to its details and make sure it fits with the character and style.
I love elephants and relating the fangs on his helmet. I decided to do it my way and sculpt the animal on both sides of the hammer. That's very tricky because you need to have clear shapes but still consider the damage to the hard surface and how the form will behave with the shader.
But it’s clear that was a good choice in the end ;)
For a good texturing workflow, you must be organized and layout your UVs well. Have in mind which software will you use and how you will texture each part.
For example, I use to organize my UVs by grouping them by material type per row. This way it’s easier to know where are all the metals or fabrics are for example. Making the texture process way faster, once you can paint the same materials at the same time with the same quality.
In total there are 75 UDIMs in this project, and I tried to keep them big enough to have a good quality close shot. But at the same time not too big as I wanted to maintain the file size and not make it too complex.
First of all, considering the model is done it’s time to start exporting the displacement maps from ZBrush.
Remembering that all the micro details will be done in the texturing stage and combined together later in the Vray shader, the first displacement pass must be exported in a 32bit maps in .exr format to guarantee the largest amount of details from the sculpt. Once exported, the displacement maps can help you as a guide and display the small imperfections on the mesh helping to see exactly what you are painting.
Texturing the Skin
For texturing the skin I decided to use Mari for its power to project and handle huge 32bit maps. Plus the flexibility to edit and paint more details by hand across the 22 UDIMs I had for the skin.
Before I started painting, I separated the amazing XYZ skin maps that I wanted to use in my projections and composed them together in Photoshop. Why compose the maps together? Well, for a better control of my displacements I figured that I must have my skin displacements separated in layers. The secondary forms (wrinkles and the small deformities of the skin), The tertiary details (Pores and smaller wrinkles) and the micro (basically a noise to blend the general details and help to break the specular bright as well).
Combining these 3 detailed maps into one single file separated by channels, allows me to paint all the 3 at the same time in Mari with no concern if they will match with the texture. After projecting my displacements all over the character’s body, it was time for the colours.
Putting the bump maps underneath it was easier to project some polarized pictures and see the results immediately. This way I could project a real skin texture and make it look as much natural as I could for in the end just tweak the colours and transform all the set in blue.
Texturing the Props
For the props I chose to go with Substance Painter for its speed, and because I could use procedural textures for most of the props. With some small breakup on the top, and knowing that it doesn’t paint across UDIMs, it’s pretty easy to manage the seams.
It's important for you to keep things organized and have your UDIMS set in an easy way for you to share the materials and textures that you’ll create for each tile.
An important thing to keep in mind when you are texturing inside Substance Painter is that the generators and masks only work when you bake the utility maps.
Instead of having a baked low-res mesh I decided to bring a more high-res one, and use this very same mesh to generate these utility maps for my texture to work. That's easier to see the small details working “physically” the through a normal map.
The wonderful thing about texturing in Substance Painter is that you can create a very good base of different variations of the same material. Saving a leather base as a smart material, its then possible to replicate for the other pieces, changing properties to look like a different kind of leather, more aged or damaged. It'll save you a lot of time having not to make a new material for each part, and it's still important to make all the materials have the similar look and feel.
That’s exactly what I did in this project, create some different bases for the same material to have different aspects on different surfaces. Turning the look of each piece unique and at the same time correlated to the rest of the pieces.
When you have the same piece split into multiple tiles, you can share the layers of one tile through the others. So, when you change something in one part it will affect all of them.
Just don’t forget to set the same viewport resolution for all the tiles for you to see what’s going on in the whole thing.
For this project I chose V-ray to be my render engine, so first I had to export my maps from Substance and Mari in a proper way for it to work as I wanted it to.
First, for the skin, the displacements were exported in 32Bits from Mari, Spec and Gloss maps in 16Bits as the colour as well.
Inside Substance I chose the V-ray UDIM preset to export the props textures. That makes all the maps named with the respective tiles numbers and has all the maps I need just to plug in V-ray’s node inside Maya.
This preset from Substance could make your Specular map look lighter than usual, but it’s normal because it is controlled by the IOR map, that is also exported in the pack. And don’t forget to set the same resolution for all your tiles before exporting or you’ll freak out not knowing the drop of quality in some parts of your model. (Yes it happened to me in some tiles, and yes again you need to set all of them manually).
Once you have all the maps ready, it’s time to build your shader inside Maya. So for this, I’ll talk first about how I built my skin shader and then the rest of the props.
In the Image above you can see that in my VrayFastSSS2 I just have the diffuse and my bump channels loaded. It is this way because the V-ray skin material doesn’t handle well the glossiness and Spec. So it’s preferable to combine the Vray skin shader to a VrayMtl with the reflection maps into a VrayBlendMtl in additive mode, to have more control in each aspect.
For the last, I decided to create a dirt layer in a simple VrayMtl plugged into second VrayBlendMtl with a dirt mask that I painted inside Mari to give the skin a more natural dirt look. The displacement here I preferred to have it directly in the shader.
Making easy the process of combining the different displacements extracted from Mari and ZBrush into a single one. With this method, I can control the intensity of each channel of the map by a PlusMinusAverage node. The props Shader was way easier to set.
The displacements of all the props were assigned through a VrayDisplacement node directly in the Outliner. Plugging than the ZBrush maps to it, the construction of the shader I separated into three parts only: The body stuff, the hammer and the sword. As you can see in the image below the props shaders are just a VrayMtl with the proper maps assigned to it.
I chose to work this way to have the control of the maps all in one place, so if I need to adjust something, I go back to Substance Painter and exported the map again. Just for the bones and ivory, I used a VrayFastSSS2 as a base of the shader, so I can have better control of the scattering of this particular pieces.
Hair and Fur
Then once placed I reduced the fibers numbers and exported to Maya as guides to Xgen. The hair shader was a simple VrayHair shader with a colour variation.
For this particular scene I wanted something more dramatic. Darker but at the same time a light setup that directs the viewer's eyes to a specific point of the character. Creating some depth with the light/shadow variation.
I’m a huge fan of traditional art and the old masters. My favourite always was Caravaggio because of his unique view of the scene’s mood through the light focus.
This way I created a simple light setup, with a good HDRI that provided me most of the fill light of the scene, a key light that directed the viewer’s attention to the face of the character, and a strong backlight that cuts off the character from the background and helps to show a better silhouette.
I want to make a special thanks to my mentor Gustavo Groppo for the commitment and patience. He's an awesome artist and a good friend whose without his help I wouldn't make half of it.
Check more stuff like this and my reel on my page. https://viniciusfavero.artstation.com/
Reel - https://vimeo.com/291227406