Learn How To Create A Realistic Dragon With ZBrush & Mari

Introduction

My name is Nicolas Morel and I am a French Senior Creature Artist based in Sydney, Australia.

After Graduating from ESMA in Toulouse in 2017, I spent four years in London, where I worked at MPC and One of Us. Then, following COVID-19 and the democratization of remote work, I got more and more offers to do freelance work, which suits me much more, as I can arrange my hours the way I like.

I then decided to move to Sydney, where I have been since November 2021. I worked one year at Animal Logic on shows that were unfortunately canceled and one year at Luma Pictures on TV shows like Halo season 2, The Boys season 4, and Star Wars: The Acolyte. My last contract to date was at Cumulus VFX, where I worked as a Senior Creature Artist on a really cool project I hope I'll be able to talk about really soon!

Cave Dragon Project

Like often with me, this first started as a "quick" concept head sculpt that I did in a couple of days.

But when I finished the first version, I thought that this creature would be even cooler if it had a body, so I did some tests.

One thing led to another. And before you know it, a couple of months had passed, and I was deep into one of my biggest projects to date. So what started as a concept sketch transformed into a big heavy project where I wanted to see how far I was able to push my texturing and lookdev skills.

Modeling

Scales on reptiles have always been a great challenge for me. I've been spending the last few years trying to understand and replicate that organised chaos that they are. At first glance, it looks completely random. But the more you look at it, the more you realise they are all organically connected to one another. For this project, I wanted to see if I could nail that.

However, while working on it, I realized the head design was very similar to one of my previous projects: Dromignon. So, after a few changes to the head design, it was time to work on the scales.

For this, I decided to create my own alphas. I didn't want to use maps, as they already have flows. Now, I know that the following method is not the quickest way to create scales, but sometimes, you have to spend time studying something to understand it.

I had three variations of the same scales with different widths. That way, I was able to follow each feature of the face.

I got myself a few key references that I liked. Amongst them are real reptiles, of course. But also some sculptures made by the incredible Stan Winston School team. Their approach to scale sculpting is so cool and I wanted, with this project, to honour them and their work.

Then from there it was simple. Adding scales, one by one and letting myself be guided by the shapes of my dragon. For the head, the whole process took about a week.

And here is the final render of the head:

For the body, to save some time, I used a tool in Houdini that creates scales based on topology. With a few iterations and a few topology adjustments, I was able to create an interesting flow fairly quickly.

But this method came with a lot of problems. I had to split the model as it was too heavy for the tool, and it needed a lot of adjustments. On top of that, projecting these scales gave me a very messy result with lots of holes everywhere. And it took me a few days to adjust and clean them.

But even if this workflow is still a work in progress, it definitely saved me quite some time. Though it has a lot of technical limitations and I still needed to do a lot of sculpting on top of it to get it to look good.

The final sculpt looks like this:

Topology

For the body, I did a very simple retopology in Maya. For the head, I simply did a ZRemesher in ZBrush and cleaned it a bit. When it comes to personal projects, I like to keep it simple, when I can. Since this model wasn't going to be used in production with complex shapes, there was no need to have an advanced topology.

I did the UVs in Maya and have the left and right identical so I can then copy textures in Mari.

Texturing

I first started texturing this Dragon in ZBrush, as I have been doing for all my projects for the last four years.

But I wasn't really happy with the result, and I felt a bit stuck. I actually stopped working on it for almost four months. During that time, I started working at Cumulus VFX, where I got to texture some cool creatures in Mari under the supervision of Mathias Lautour, who is an excellent texture artist, technician, and leader. I also had the opportunity to work with him back at Animal Logic, and his technical knowledge really helped me theorize my texturing approach.

The first step was to restart from scratch in Mari and lower the values of the clear parts as they were way too bright, and that's why I was stuck, as I didn't have enough range.

While working on the lookdev, I also switched lightrig often and changed the angle to get a fresher look as our eyes quickly get used to something and it makes it more difficult to spot issues.

From there, once again patience and handwork were key. I didn't use any textures that I projected. I painted everything by hand. For the actual painting work in Mari, this would request a whole tutorial to cover it. But I am actually thinking of creating my own Mari tutorial to dig deep into my texturing process.

Rendering & Lighting

Like in all my projects, I rendered in Maya using Arnold. At first, I had this idea of an underwater creature, and I had done a few lighting tests.

But as the projects went on, it kind of drifted apart from that idea. And what was originally a water dragon transformed into a cave dragon. I had the idea of having this huge scary beast peacefully sunbathing in its cave. Here are some of the references I used for the posing.

For the environment I mainly used scans that I purchased on LotPixel.com. I find it can be extremely time consuming to sculpt rocks and in that case it saved me a lot of time.

It took me roughly a week to build the environment and do the lighting.

Here are a good few rules for lighting that I like to follow:

• Direct lights, like the sun, generally have sharper edges. To achieve that, the lights need to be smaller;
• Ambient lights generally have softer edges. To achieve that effect, the light needs to be bigger;
• Your key light should be the strongest light. In my case, that's the sun coming from the top, so all the other lights are not as strong;
• Notice how far my fill light body is. That’s because I want it to spread across the whole scene. So, I keep it far away from my model and have it stronger so it's more uniform;
• By opposition, I want the fill light head to only hit the head. So, by keeping it close to the head and reducing the strength, it doesn't impact the rest of the asset.

Animation

As I said earlier, I didn't build this asset at all to be rigged and animated. So, I was quite limited here. For instance, there is no inner mouth. Plus I had spent enough time already on this project so I wanted to keep it simple.

So, I simply did a 120-frame loop of my dragon breathing and opening/closing its eyes. For that, I used a few shapes that I sculpted using ZBrush. I have three different shapes for the breathing loop so I can offset the animation, giving a more natural look.

Conclusion

I started this project in early November 2023 and finished it in mid-July 2024. But it didn't really take me 9 months to complete as I didn't work on it for four months.

The body was obviously a big challenge, but the most challenging part was that, until almost the very end, I didn't like what I had. I just wasn’t happy with the quality of the sculpt and the texturing. I almost gave up on that project so many times. But I kept on coming back at it and didn't hesitate to restart what I knew wasn't working.

I spent weeks refining the first head concept but decided to restart it as I didn't like the final look. I threw away the texturing I spent weeks working on in ZBrush just because I knew I wouldn't be able to push it as far as I wanted.

I think a good message to beginners is: don't give up because you don't like what you are doing, it is okay to struggle on a project. If you feel stuck on something and you feel like what you are doing is just not working, don't hesitate to take a break, work on something else, and restart a few parts. It can be very frustrating, but the second time you'll probably do it better and faster. So it is worth it!

Nicolas Morel, Senior Creature Artist

Interview conducted by Gloria Levine