LogInAlex Avdeev shared with us his experience creating the collectible-style project Ruined King Viego and modeling the body, clothes, and armor while keeping in mind the competition rules.
Introduction
Hello everyone! My name is Alex Avdeev. My fascination with 3D graphics has been with me for as long as I can remember, and I've been remembering for a good long while now. I haven't graduated from any specialized art schools. I haven't even taken any 3D courses. And I didn't have any structured self-study, either. I just watched how others did things and took something for myself from each one. Also, I have never worked in any 3D studio or participated in any projects. 3D graphics for me is a hobby.
Over the years of watching 3D content, I've gained a lot of knowledge about 3D, which is more than enough for my needs. And if something new appears, some technique or trick, you can always find it online. Lately, I've been consistently following the work of a wonderful 3D artist, Olya Anufrieva.
Despite my long-time acquaintance with 3D, strangely enough, I've hardly had any of my own projects. Not to mention finished and published works. I didn't even have an ArtStation page. And it might have continued like that if this project hadn't come along. The thing is, the Ruined King Viego project was born as part of a friendly challenge (JoeMADChallenge), organized by 3D/2D Artist Viacheslav Gedich. I found out about the challenge by chance, from friends.
Initially, I didn't plan to participate, as I wanted to finally finish a couple of my own projects that had become long-term constructions. And finally publish something. But then, after thinking a bit, I decided that participating in the challenge would help me get the job done and, at the same time, help test my organization and ability to work on a deadline. That, in fact, was the essence of this personal challenge for me: to see it through to the end, meaning to finish the model. And to understand, for myself, what I am ultimately capable of. I also had no prior experience participating in competitions.
A Bit About the Challenge
The essence of the challenge was to strictly follow the concept art by the famous 2D artist Joe Madureira (Joe Mad) to create a 3D model in the style of a collectible statue. Not fully ready for 3D printing, but giving the impression of a printed model. Further on, in my explanations about working on the model, I will add to the competition conditions.
Viego
I liked the Ruined King concept immediately. I won't say what else, besides the cool visuals, hooked me. Perhaps his sad story: the eternally sorrowful king, forced to constantly search for a way to resurrect his beloved. Also, when choosing a concept, I wanted as few participants as possible to choose the same one, but in the end, there were five people, including me. However, by that time, I had already finally decided to make him.
For the model's pose, I decided not to invent anything and chose what seemed, to me, a simple pose by another 2D artist (not Joe Mad). This was acceptable within the challenge rules.
Since the task was a figurine à la 3D print and the pipeline didn't imply retopology (just a sculpt in ZBrush), I decided to make the model in the final pose right away. That's why it was important to set up the model's blocks in a way I liked from the start, so as not to be distracted by edits later and fully focus on sculpting the model.
Even though the legs on the reference are visible only up to about halfway, positioning them correctly turned out to be not so simple. I even had to pose a bit in front of the mirror. When I was finally happy with the pose, it was time to move on to the next stage.
The jacket, pants, and belts were made using Extract. First, the part of the body where the clothing should be is masked, and an Extract is performed. It's inflated a bit with Inflate, and a ZRemesh is done. Then, the desired shape is given using the Move brush, periodically recalculating the mesh with ZRemesh. It's an old, simple, but proven and reliable method for modeling clothing.
The armor elements and small belt details were first roughly sculpted, then a Remesh was applied to them, and afterward, they were refined using ZModeler and the Move brush. Here is an example of the boot:
One of the elements where I decided to switch software was the sword. And it's not that I couldn't have done it in ZBrush. Just with my not very confident ZModeler skills, it would have been long and painful. So, at this stage, I decided to move to 3ds Max. The sword is almost entirely modeled in 3ds Max, except for the cracks and chips.
The cracks and chips, both on the sword and on the armor elements, were made using LiveBoolean. A very powerful and cool tool! It allows you to literally edit any chip or dent in real-time, as if you were sculpting it. Initially, I wanted to make the chips using special brushes with corresponding alphas, but that would have required highly detailing the model (increasing subdivisions), which I didn't want. LiveBoolean handles subtracting geometry perfectly, even on not very high subdivisions.
The torso was simply sculpted using anatomical references. Nothing much to explain here.
Head and Hair
I decided not to copy the face from the concepts but to create something collective, in my own way. I wanted to please the girls with the pretty face of a capricious prince. After experimenting a bit with the shape, I got a suitable result.
Since the model was being made for a collectible statue, no complex forms were needed for the hair, on the contrary: the simpler, the better. And also, according to the competition rules, we shouldn't deviate too much from the concept art. Therefore, the hair is made very simply with the Curve Tube brush with slight adjustments for the number of polygons, thickness, and taper curve.
A couple more elements made in 3ds Max were the base and the smoke. The smoke, because in 3ds Max it's convenient (for me) to work with curves, and it's very simple, using a modifier, to flow any geometry along those curves. Again, everything that I did in 3ds Max could have been done in ZBrush. Just for me personally, it's more familiar and faster to do such operations in Max.
As for the base, I wasn't very happy with it. I kind of forgot about it almost until the deadline itself, and at the end, I hastily put together something unappealing. It doesn't have much artistic value, although it was modeled in a rather tricky and labor-intensive way, which doesn't make sense to recount.
Texturing
According to the challenge rules, texturing in the usual sense, using UV unwrapping, was not allowed. The model had to be painted directly on the surface, so-called Polypaint. I won't advise on painting, as I'm far from an expert on this matter. For me, the difficulty when working with color is selecting colors and shades, and transitions between colors. I think only practice can help here.
Closer to the end of the challenge, the organizers allowed the use of one of the new features of Marmoset Toolbag 5, the so-called Tri-Planar Projection. The point is that for projecting Texture Maps onto the model's surface, no UV unwrapping is needed. This way, you can give the figure an additional interesting texture.
For rendering, I used Marmoset Toolbag 5. A simple and flexible render engine that produces high-quality images and video. Here I have to reveal a terrible secret: I didn't use a single additional light source. The scene is lit by Marmoset's standard HDRI Map, and the default one at that, which loads when the program starts.
Since I started setting up the materials under this HDRI, it turned out that I tuned the material settings for this default light. And also, the fact that the model was supposed to look like a 3D-printed figurine actually turned into an advantage, and that default lighting worked out pretty well for the final render. Overall, I think the render turned out acceptable. Maybe not perfect, but not bad. No post-processing was used.
Conclusion
In this competition, I became a finalist, and I'm more than happy with that, as the level of the participants was high and many of them have a lot of practical experience behind them. I finally learned in practice my strengths and weaknesses and made certain conclusions. I finally applied some tools to a specific task, not just on primitives, for example, Live Boolean. But the most important thing was to see it through to the end, to finish the work, not to give up. That is my main victory, for which I participated!
To all those who want to get into learning 3D graphics from scratch, I wish you patience! Don't give up, even if something doesn't work out, keep trying, try again and again! It's especially tough for beginners now; the level of work has increased a lot. But don't despair, you need to keep moving forward and everything will definitely work out! Study and analyze the work of other artists, pay attention not only to the technical side, but also don't forget about the artistic component.
I want to thank Viacheslav Gedich for the great organization of the challenge! Joe Madureira for the wonderful concept art! And my supportive guide, Olya Anufrieva, for her constant motivation and inspiration, and just for being who she is! Also, all the creative people from Olya's group, from Viacheslav's community, everyone I've known for a long time, and those I've met recently! Keep creating beautiful things, no matter what!