Okay, to start how did you start your career as a 3D Artist? What was your first 3D software, and what type of learning curve was there in the initial phases?
When I was in high school, I once picked up an opening cinematic trailer of Reign of Chaos, Warcraft 3. That was when I seriously considered working professionally in the entertainment industry. Maya was the first software I used since the college. In the initial phases, besides getting familiar with software user interface and such, I also needed to figure out what I really want to focus on. It took me a pretty much just one semester to find out what I wanted to do, becoming a 3D artist.
What sort of education or training did you seek out to further evolve your skills as a 3D artist? Did this schooling fully prepare you for working as a 3D artist?
I have majored in computer animation at a college. After just a few internships, I realized that my skillset was not enough to match the industry criteria. Thus I decided to enter Gnomon for Digital Production of Modeling and Texturing Track, which took me two years to complete. Even before I graduated from Gnomon I could already tell that my skills were seriously leveled up in many aspects. Apart from skills, Gnomon also emphasized to maintain a positive attitude under a great deal of pressure while working and to be friendly among other people, which also helped a lot during production work. I would say that Gnomon fully prepared me for working as a 3D artist.
What surprised you when you entered into the industry professionally that you were unaware of as a student?
One thing that surprised me is that clients tend to change their ideas a lot during production work. Apart from that, artists here in this industry need to learn new skills pretty much on their own, which didn’t surprise me but I think it’s important to bring this up. In most cases, lead artists and senior artists are probably too busy to help junior artists during production. Junior guys are supposed to learn on their own and ask questions proactively.
Alright awesome! Well to switch gears a bit could you tell us a bit about how you started your process on Hanzo? What inspired you to create this character?
So the way I started was to block out Hanzo‘s torso in Zbrush based on an existing base character I’ve done a while ago. I then brought it into 3DSMax to build his cloth and props in very low poly count based on Hanzo’s body. During the making of his props I just focused on the shapes and didn’t care much about topology because I knew that Dynamesh and Zremesher were able to resurface them with ideal topology for me. I tend to work with everything together before I went in deep because this could help me to grab an overall feeling of the whole character. When the high res model was done I then re-surfaced the whole thing with Maya Qual Draw tool. As far as the texturing went, I hand-painted the base color in Body paint mainly for the skin, some detail projections and seams. The rest I did all in Photoshop. The reason I picked up Hanzo was because he seemed quite different from the rest of the characters in Overwatch. Hanzo appeared very determined and a bit saddened in his face, which were quite different from others in terms of the facial expressions. We could tell that Hanzo’s facial expression changed very little and was naturalistic in the cinematic trailers. I like Hanzo’s Kimono elements, which separated him from most of the characters. Also unlike most of characters, Hanzo’s clothes broke the symmetrical pattern while most characters maintained symmetrical clothes pattern, which made him even more unique.
How were you able to sculpt and model the clothing to fit the style of the character? Is there a certain technique that you use when sculpting to be able to create stylistic shape languages?
In terms of clothing creation, I built the clothes based on Hanzo’s torso with low poly counts. While I was building clothes in Max in the beginning, I toggled Turbo Smooth tool to create the smoothed preview for the clothes to see if the they were going to fit onto the body or not. Like I said earlier, I'm not too concerned about the topology. If the topology I have isn't too messy then I'm happy. During the sculpting stage, I would block out the main cloth folds and I constantly viewed it from a distance. Once I was happy with the overall form I converted it to a Dynamesh and then continued sculpting. Dynamesh is a really good tool for stylistic shape sculpting. I had fun sculpting cloth folds in Dynamesh mode, which would always update the non-stretched four-sided faces for me while I was sculpting. I Zremeshed the clothes with a proper topology and projected all the sculpting detail to Zremeshed mesh. Then I just did some cleanup during the final stage.
Do you have a different process with different assets of the model? For instance, is your process different from modeling the hair and then the metal legs?
For Hanzo’s hair, I just created a sphere in Zbrush placing it in the right place and converted it to a Dynamesh. For other hair strings, I created another Dynamesh, blocked out the main shape and started duplicating them. That was how I started doing the hairs. And for the metal legs, I would just create base meshes in Max and brought it in Zbrush, converting them to Dynamesh to continue sculpting. From time to time, I would Zremesh those legs for a couple of times in order to do some cleanups. I could block out the hair base mesh in Max if I wanted to instead of doing it all in Zbrush. In terms of workflow, it is really up to the artist’s personal preference. There won’t be any right or wrong ways to do things.
Okay cool! So what techniques did you use to paint the basic color maps on the model for initial baking for in-game engines?
First I would define how many materials I need for painting on the character. I would rely on those normal maps, AO maps and Curvature maps to paint basic color in Photoshop. The way I started painting them was to layout the base color first and then brought in those baked maps on to the top of the base color layer. I converted the baked maps to multiply and/or overlay mode. Then I would continue hand painting. After that I would use some photo bashing method to layout some damages and dirt on areas like his metal legs and such. For Hanzo’s face including his arms, I used Body Paint 3D to paint it, but I would still do the base color in Photoshop first. Then I brought the skin base color to Body Paint to continue painting. That way I could view the character in 3 dimensional space, which would make the painting process much easier. I would also use some projection techniques to project some details onto the character and paint away some of the seams in areas like armpits.
What was your process on retopologizing the model for lower poly count? And how important is it for artists to be able to retopo manually?
I resurfaced the whole character in Maya using Qual Draw tool. It is very important for artists to understand that the resurfaced low res model must be rigged and animated properly during the production. Zremehser is a great tool for auto resurfacing; it gives the artists a quick and proper topology that met for further sculpting and cleanup related purposes. However, to come up with production ready models, one must be able to manually retopologize them in order for riggers and animators to use them during the production.
And finally, what advice would you offer young artists who are looking to work in games? What type of training would you recommend to become a better artist?
In order to keep being a great competitive artist in this industry, one important thing to keep in mind is to never stop learning. As we can see, every once in a while, new software will be invented and most likely replace some outdated programs that doesn't cut it anymore. As for a character artist, anatomy should not be ignored. Even the best character artists could not stress enough how important the anatomy is. What’s more, having outstanding mentors around would be a huge help for growing artists. I would recommend Gnomon for those who are seriously wanting to become professional artists in the entertainment industry. Honestly, I won’t be able to make this far without Gnomon's education. It’s definitely worth the energy, time and money. Also I would like to give a big thanks to Wits and Young, those two great character artists who guided me during the making of Hanzo Fanart.
CGSociety would like to thank Yingbiao for taking the time to explain his processes! We hope this encourages and inspires everyone, and we look forward to seeing more incredible work from Yingbiao Han!