This is amazing! Please tell us, What programs where used to create these amazing animations?
I am continuing development on WorldKit as a solo endeavor now. Progress is a bit slower as I've had to take a more moderate approach to development hours. I took a short break following the failure of the commercial launch, and now I have started up again, but I've gone from 90 hour work weeks to around 40 or 50 hour work weeks. See my longer reply on the future of WorldKit here: https://www.youtube.com/watch?v=CAYgW5JfCQw&lc=UgxtXVCCULAyzrzAwvp4AaABAg.8swLeUjv7Fb8swt1875FAT I am hard at work with research and code, and am not quite ready to start the next fund-raising campaign to open-source, so I've been quiet for a while. I hope to have a video out on the new features in the next few weeks.
Someone please create open source world creator already in C/C++.
Awesome 3d character breakdown by Dan Lipson, talking about the way he modeled and textured this character.
My name is Dan Lipson. I’m originally from Indiana and studied film production at Indiana University. I took as many electives as I could in the art department while I was there, ending up with a studio art minor. When I started getting interested in 3D, I took a few online classes through AnimSchool on the evenings and weekends. While at IU, I had the opportunity to take a painting course with Tim Kennedy which has really informed how I approach and think about art. I decided that I wanted to pursue 3D as my primary focus instead of my secondary focus, so I sped up my program at IU and applied to Gnomon. I’ll be graduating in 2019 from their modeling/texturing track, personally focusing on texturing, look development, and rendering.
Concept Selection and Planning
This project was built for Robby Branham’s Texturing 2 which focuses on Substance Painter and V-Ray, so I was looking for a concept that would be simple to model but had a variety of materials to explore. I’ve been following Pierre-Antoine Moele for a while and thought his Sake Master concept was perfect. I love the distinct, hand-painted style. I thought it had some interesting specialty textures that would allow for storytelling through the wear of the materials.
Google and Pinterest are great resources for collecting reference, but I find a lot of good images on Flickr. Photos from Flickr are not indexed by Google Image search, so they’re often missed. It has huge group-curated photosets with photos from professional and hobbyist photographers. For product references specifically, I’ve also found a lot of good reference from eBay. I’d really encourage people to be creative with where they look for reference.
The modeling process was very straightforward because every component is a primitive shape. Especially with simple shapes, I find it’s important to model in the looseness from the original reference so that the shapes still feel natural. I decided to stay entirely within Maya for this project and found that sliding edges, pulling points, and lattice deformations were enough to get the shapes I wanted and add enough variation to repeated elements so that it doesn’t have that copy-paste look. I looked a lot at work from Laika with regard to how they keep simple shapes feeling loose and natural like their initial concept art.
Before laying in any textures, I want to find all of my photo sources and prep them so that I’m not interrupted looking for new textures later on. Woods, metals, and string are pretty common in CG, but it was harder to find and develop textures for the gourd and especially the bamboo. Because bamboo is round, I had to spend a lot longer photobashing and balancing color from a lot of different bamboo references to get a texture that was evenly lit. One of the challenges I enjoyed with this project is that some materials in the concept were not defined as real materials. The skin of these characters is not human skin, but I still wanted something that would feel maliable and soft. I ended up looking at black licorice for reference, specifically Panda Licorice. I remember creating that base material from a lot of different images, including sandpaper and carpet, to give it a natural noise. Developing these fictional textures that have to sit in the scene with real materials is the most fun part of this process to me because it allows for more experimentation and creativity when processing the images.
Once all of the image assets are prepped, Substance Painter can really shine. It is so quick to build and assign basic materials and make sure the textures and colors are working together. After getting textures assigned in Substance, I set up standard V-Ray materials for everything. It was useful to pay close attention to how the materials translate across programs, using remap HSV nodes to align my V-Ray scene a little closer to the color values that I see in Substance. That way I’m sure the work I’m doing will translate as I expect. And then, comes my favorite part of this project, hand painting in the detail and wear. For this, I really spent a while going through each asset thinking about how it’s been used. Does the gourd get set down at night in the dirt and leave stains underneath? Where is it going to scrape against stones as he walks? How has the paint on the mask worn? Has it faded from the sun in some areas more than others? What direction will the sake drip and stain the gourd? Substance can generate a lot of this material variation and wear through its smart masks, which can be a great starting off point, but it can’t replace the richness that hand-painting brings. These details are important in selling the scale too. Being close to something so small will make unique imperfections more visible.
V-Ray and Lighting
Once the textures are getting to a finalized point, I started using specialty materials and attributes in V-Ray. I used Fast SSS materials on the strings and ropes, as well as creating V-Ray fur to add small, fraying strands. V-Ray falloff transitioned the rounded edges of the fabric to white so that it had a softer read. My teacher, Robby Branham, helped me especially to figure out how to sell the little bit of glaze on all of the ceramic components by creating a slightly larger duplicate mesh, a sublte fractal displacement, and a purely refractive material.
The lighting on the scene is very straightforward to not distract from the textures. Aside from a key light to act as the sun and an HDRI, there are a few small lights to just kick some light into areas that would get too dark otherwise. These small lights help keep details from getting lost in shadow, but should never be noticeable in the end. Additionally, the rim light helps separate the characters from the background, but also provides a lot of the light that scatters through the Fast SSS materials and refractive elements.
Originally I had planned to render this in a grey void, like the original concept, but Robby pushed me to add a full environment, and I’m so glad that he did. Most of the plants are low-poly modified SpeedTree assets. Because I knew they would end up as blurry masses in the back, I really aggressively color shifted them to match the color reference I had collected earlier. Lastly, I needed ground elements to sell the integration and the scale. The rocks, sprouts, leaves, and flower petals do just that. Laying them out is a lot of time just shifting things back and forth, trying to avoid strange tangents and getting clean overlaps. The most important thing is to keep clear silhouettes. Here I was focused on the framing of the gourd and rope at the top, making sure they are not interrupted by trees and giving them a clean sky to really pop against.
For final processing, I knew I wanted to render a partial turn. At the time I had no experience with Nuke and was already very comfortable with After Effects, making it the clear choice for me. I rendered a still frame with all of my passes and composited it first in Photoshop, later replicating that layer stack in AE. I knew that the depth of field would be important to sell the scale, but the character itself has more depth than I was expecting, especially with the sidekick. This meant that doing too much blur in camera would blur the sidekick and the master’s hand and gourd too much. I wanted a little bit with these elements, but still wanted the textures to read, so I mixed the lens blur with depth map blur in the post so I had a bit more control. With color correction, I used a combination of curves, LUTs, blended fill layers, and more with very low opacity. I wanted to keep the light feeling natural, but tried to push the image to be really warm without losing the cool shadows. There ended up being one visible texture seam that I removed in AE, but otherwise, the comp stayed very much restrained to just color, a little bit of lens distortion and light leaking into the frame. I knew aggressive compositing would just distract from the texture work and draw focus. Because this piece is focused on the textures above all else, during the other parts of the process I tried to not to do too much that could distract. Complex modeling, lighting, and post-processing would only muddle the textures instead of supporting them.
Thank you to Kirill Tokarev for reaching out to me and sharing my work here. I’m really honored to be recognized here since I feel like I still have much to learn. If anyone has questions about aspects of my work, about Gnomon, or anything else, feel free to reach out to me through the links on my Artstation.