Hey Everyone! My name’s Lauren (or Laurbavour) and I’m a game artist based in the sunny coastal city of Brighton, UK. I’ve been working as a professional artist for a little over a year now. I mostly focus on character work but like to stretch my legs with some 2D and prop pieces from time to time, depending on the project. I’ll be talking you through the creation of my latest character, a happy little Orc.
You can see more of my work on my ArtStation page.
My model is based on a concept by the amazing Satoshi Matsuura called “Orc and Meat Spike”. I absolutely adore their style and have been a long-time admirer of their work, creating a couple of ZBrush sculpts based on their concepts before. But this time I wanted to make a game-ready, low poly character and couldn’t resist bringing this cheeky little orc to life!
I went about creating the model in a very “old school” way. First, I roughly sketched out what I thought a front and side view of the Orc would look like in T-pose and without his accessories. The concept isn’t very complicated so I felt confident I could box model the character and therefore not need to worry about re-topology at a later date. To do this I used Maya and, beginning with a plain old box, began extruding faces and adding edge loops to very roughly match the silhouette of my sketch. Paying attention to the overall shape and appeal of the character from all views, I moved the verts until I was happy—adding more resolution where necessary. I also added an extra edge loop or two around the elbows and wrists as I knew these would need to deform to match the pose.
Once I felt happy with how the main body was looking, I modelled the teeth and eyes separately, moving them into place and combining everything together. I adopted the same technique to create the Orc’s clothing and accessories. Each piece began as either a standard or a smoothed cube and I manipulated the edge flow until I achieved the shapes I was looking for; I then combined the pieces into relevant clusters (i.e., the meat and stick became one object, as did the belt, etc). With each element of the character now in place, it became easier to see any disparity between my work and the original concept. This was the perfect point to start drilling down into the details and tweaking the model to match Satoshi Matsuura’s illustration as best I could. I also used this opportunity to review the model’s topology and clean up any untidy sections. Finally, I added a little extra resolution to round out the nose and ears so that the model wouldn’t appear boxy from any angle.
After I felt happy with the modelling, I moved onto creating UVs using 3D-Coat. I personally find this software the fastest and most intuitive way to create UVs for any mesh as it allows me to place seams easily and unwrap a whole model with just a few clicks. 3D-Coat also provides me with a level of flexibility should I need to go back and tweak the UVs once I’ve started painting (spoiler: I did!).
I used one UV set for both the character and their accessories, being sure to carefully utilize as much of the tile as possible when arranging the UV islands. I’d planned out which areas would need to be painted in the most detail (head, chest, meat ends, etc.) and made these UV islands the largest whilst assigning smaller islands to less visible areas. Once all the UVs were mapped and packed nicely, I moved on to texturing.
To emulate the flat, cartoonish style of my reference, I chose to hand-paint the character. Most of my character work is colourful and stylized so I personally prefer hand-painting textures on an unlit model so I can fully control light, shadow, and colour. To begin this stage, I switched to the paint room in 3D-Coat and changed my view of the model to “flat shade” which eliminates any shadows being cast from outside lighting onto the mesh. This way I can be sure any colours I choose later will be accurately represented on my model and in the Sketchfab viewer. Next, I baked Ambient Occlusion (AO) and curvature maps to give me a starting point for shadow and highlight placements. These provide great points of reference for how the light would behave if the scene were lit. I spent some time painting over the maps to remove any blemishes the bake had created and to make sure the shadows appeared as I wanted them to. At this stage, I only wanted very subtle shadows/highlights to give the model a sense of form without losing that cartoon appeal. I find it’s easy to get distracted thinking about how a shadow “should” look in terms of the real world, when this may not necessarily be how it is represented in the original concept. I made sure to keep my reference close by whilst finishing up this section of the painting to avoid such pitfalls!
Once I felt happy with the AO and curvature, I moved onto adding colour. I did this on a separate layer so it could be independently edited and not become confused with the shadows and highlights. By adding blend modes to the layers, the AO and curvature layers remained visible whilst the colour could be applied to the model non-destructively and adjusted much more easily. After I’d finished laying the flat colours down, the final stage was to paint in the large areas of shadow, highlights, and detail. Again, I did this on a separate layer and simply painted in the details as I saw them from the concept. I was careful to keep the fine black lines across the model as crisp as possible to represent the sharp angles and hand-drawn outlines of the original character.
Now that the textures were looking good, I could finally move on to posing the Orc—this would add that extra level of character and bring him to life! To do this, I created a rig for the model in Maya. As I knew the Orc’s pose wasn’t an extreme one, I could get away with making a very basic rig. I created a couple of bones for each ear to achieve that subtle kink and asymmetrical angle from the concept. Three bones were required in each arm to create the wrist and elbow joints and an extra bone here and there to keep the character’s big belly, bum and snout in check. Besides those additions, the rig took the form of a very simple, bi-pedal skeleton. After skinning the character and tidying up some paint weights, I posed it to match the original illustration as best I could. In doing this, however, I noticed that part of the shadow cast from the character’s cape had now shifted as the arm deformed to match the pose. This looked messy and didn’t make sense for the character, so I popped back into 3D-Coat to paint over the affected area. With everything now looking good, I moved the model into Sketchfab.
Setting up in Sketchfab
Because I’d chosen to hand-paint the model in an unlit environment, I didn’t need to make many adjustments when setting it up in Sketchfab. I imported the mesh and changed my shading settings to “shadeless”, which removes cast shadows and creates the same effect as the “unlit” setting I used when painting in 3D-Coat. Next, I uploaded my textures. Everything was on one base colour set so that was nice and easy! I also made some tweaks to the sharpness and colour balance settings before adding a background colour to make sure I matched the concept as closely as possible. Finally, I added a baked ground shadow (thanks for the suggestion, Abby and James) to bring the character forward and really make it pop!
And here is the final result!
Thanks so much to my mentors Jack, Ashley, and Karen for their support and wise words throughout this project. This was my first ever upload to Sketchfab (but certainly not my last!) and it’s been an amazing opportunity to share my process with you all—thanks for reading!