Making Of 'The Two Mighty Kings'
My latest work, "The Two Mighty Kings", was created for CGSociety's 23rd challenge, "Steampunk Legends and Myths". The project took me a little bit less than two months to complete. Photoshop and XSI were used for the creation.
I didn't spend too long trying to find the right legend because, to me, that's a waste of time. There's no particular legend that I think is better than any other. So I set my focus on the old Chinese legend Journey to the West and decided to make the monkey fight a giant Steampunk monster.
I'm not from an art background, so drawing is something I'm not very good at. To make my early concept, I grabbed a couple of elements from the Internet and put them together. The idea of making the concept image is to set my focus. I need to know what I will have to face further on in the process. I need to know how I want to deal with my composition, light and colour; the number, size and pose of characters. The more time I spend thinking before I start, the better result I will achieve. Jumping right into modelling a character is never a good idea (Fig.01).
There were two characters that I planned to include in the image, a big one and a small one. According to my concept, I knew that the big character would take up at least half of my image and he was surely where I wanted to focus the viewers' attention at first glance. Therefore I started by modelling him.
I modelled a simple base mesh of a gorilla, then rigged and posed him (Fig.02). The reason I did this is because I wanted to establish the silhouette of the character as soon as possible. It's just like drawing; you want to get the gesture of the character first before you draw each muscle.
I keyed the neutral pose of the character at frame 0, and keyed the action pose at frame 1 (Fig.03). I had my camera fixed after I set the pose (Fig.04). The low-res modelling, posing and camera fixing took me two nights.
After that, I modelled a bunch of gears and pipes so that I could "Lego" them onto the base mesh (Fig.05). I worked on the details at frame 0, and constantly checked the result at frame 1 to make sure that I could see all the details through the camera from that pose. I ignored the back of my monkey because I couldn't see those areas through the camera (Fig.06).
At the end, the monkey himself weighed about 1.8 million polygons (Fig.07).
I didn't take the traditional approach of making a 3D character. After modelling, I skipped texturing for two reasons: one, I'm not good at texturing and two, unwrapping 1.8 million polygons would have taken too long and I only had two months to work on the piece. So instead, I made three personal metal/bronze shaders for the monkey (Fig.08). Those shaders were procedurally generated, light and camera based. Therefore no texturing or serious unwrapping was required.
I used a total of six shaders, including three metal shaders, two lamberts, and one AO (Fig.09).
The lighting set up was quite easy. In fact, there were only two spotlights simulating the backlight and one directional light simulating the front light. It was kind of like the classic three spotlights rig with a little tweak.
There were six layers to render for each character. Each layer contained one shader and each procedural shader took about nine hours to render (my computer is OLD). My render resolution was 4k by 5k.
I had only two months for the challenge, so I tried to save as much time as possible. Everything except the characters was hand painted or photo edited in Photoshop. It was my first time painting for a final image. I spent a lot of time painting the hot fire in the monkey's eyes and mouth (Fig.10). I also had some trouble with the background, but eventually pulled it off (Fig.11 & Fig.12).
I hope you enjoy the image and found this making of useful. Thank you very much (Fig.13).