Breakdown: Creating a Clay-Style Viking Cottage in Maya & Nomad Sculpt

Introduction

My name is Yoel Pereira, and I'm a 3D Environment Artist working in features since 2018. I started studying a generalist course in Spain, learning a bit of everything in graphic and 3D design, going from animation, video games, editing, coding, and web design, but early on, I found that modeling and environments in particular are what I enjoyed the most.

During the last few months of this course, I started an internship in a small studio in Santiago de Compostela (Spain) working on the feature film of Turu, and soon after, I started working as a Junior 3D Environment Modeler. After that, I have been working on projects in Europe and Asia for Netflix, Pokémon, Clash of Clans, Skydance, Sunrise Animation, and Adobe, among others.

Getting Started

It all started in 2019, after finishing my first film production in "Turu the wacky hen," I was left with the challenge of finding a new job. Modeling tends to be one of the first departments to finish production, and it may take months or even years until the movie comes out, and you can use your production work in your portfolio due to NDAs.

It was time to update my portfolio showing my latest work, and that's how I proudly finished "The Inn" in 2019, based on the concept of Greeimm Bae.

Fast forward to 2025, I found myself in a similar situation as in 2019, with an old portfolio that felt out of date. The portfolio is one of the most important aspects to find a job, and as some say, "your portfolio is just as good as your worst project." But at the same time, it's hard to just hide old work that made you proud and that took you weeks or months to complete.

So with that in mind, I decided to take another shot at this old project and do an updated version of it.

The main aspects that I wanted to achieve with this project were:

1. More cohesive style. Since it's a personal project, instead of just matching the concept, I wanted to take some freedom to design, adjust, and complement some parts of the original design. One of the main issues that I had in 2019 was designs that didn't belong in the same universe, which can be seen clearly on the fish and pork in the chimney, so consistency in the design and style was one of the priorities.

Since the theme of the project is a Viking cottage, messy and half torn apart, I decided to go for a more adult and detailed style instead of a preschool cartoony vibe.

2. Emphasis on storytelling. I wanted to steer away from rendering a few still frames and have more focus on telling a story through the project. While that's something that I intended to do in the 2019 version with the Worms area, it felt like it didn't happen much anywhere else. So I tried to achieve a micro story or some action in every corner, feeling like there is a story when doing a close-up in each corner of the scene, and that leads to the last point.

3. More detailed. If I wanted to do close-ups on each corner of the scene, I would have to make sure that everything keeps a consistent level of detail through the scene. While it would make sense to have some hero asset more detailed than the rest, having close-ups in every camera meant having to detail everything. And since it's a personal project with no real deadline, I wanted to enjoy the journey and detail everything as far as I was capable of.

Composition & Blockout

The whole idea of the project revolved around reworking the existing from 2019, so I didn't want to change too much of the original idea and the main cameras that I would render. I used the original environment as a blockout for the new version, and the cameras as a starting point for the new composition, mainly using a golden ratio to achieve an interesting composition, while trying to always maintain three levels of depth to the scene. 

Modeling

As of modeling, except for the pig in the chimney and the mouse character, which were done as a rough blockout in the iPad with Nomad Sculpt, everything was done polymodeled within Maya.

Wood assets were reused a lot, but that also meant that I should avoid visible repetition. For that, I polymodeled a few pieces with a clean topology and extracted a few sections that I could "kitbash" into new planks to add detail without having to model all of it from scratch.

While wood was reused throughout the scene, some assets had to be more unique, and that was the case with the table. Unlike the previous version, where the pattern seemed to be over the ring.

I wanted the new one to feel like it was carved out of a wooden trunk.

The pattern was modeled by tracing the concept, using instances while modeling, which helped ensure that I could create a seamless piece that I could weld together.

Finally, I applied a bend modifier to achieve the round shape, and added some final breakup details to avoid a perfect repetition and achieve more naturalness.

For the base of the table, I wanted to replace the old thin column as it felt a bit too unstable. I did a rough 3D model combined with some photobash to explore different options before committing to doing the final model. 

Finally, while the cloth would be the least visible part, I wanted to make a fabric piece with more sense of weight and gravity. For that, I did a cloth simulation with nCloth over a proxy mesh, and transferred the deformation to the high poly using a blendshape and wrap deformer.

Assembling the Final Scene

Considering the amount of detail that I was planning to add, I was expecting a quite dense scene from the beginning, so I used a Reference workflow, adding all the props and assets in their own scene and referencing them into the main assembly. This helped having a lighter scene to work on, and having a more narrow perspective on the assets helped push the detail of the assets further, since it's easier to mentally view each asset as a hero prop when you are only focusing on that one instead of having a big assembly on the background distracting you.

Lighting & Rendering

I have always been limited when it comes to lighting, so I prepared a simple setup to light every corner of the room without creating too many shadows that could hide the detail put into the assets. The materials were fairly simple too, just three aiStandard surface materials, one material serving as the clay base for the entire scene, and the other two to create brighter and darker accents of color and contrast in different pieces.

I wanted to take special care of the presentation for this project, not only in terms of renders, but also by emphasizing the story behind it. That's when I decided to add some characters and animations to it in order to make it all feel more alive. 

Conclusion

It's always hard to measure time for personal projects since you tend to do it on and off during your spare time, but I would say it was around two months of full-time work from start to finish, considering that a big part of the blockout was done from the 2019 version.

The main challenge here was to try to bring something new and conceptualise an idea, storytelling, and animation that felt cohesive, which took a long time of finding references and trying different cameras and perspectives.

Advice for beginning artists would be to focus on topology and especially observation, which doesn't mean only looking for references, but in trying to understand the logic and background of how objects, vehicles, characters, or environments are built and applying that logic and knowledge to whatever you create. 

Whatever software you use, never let that limit your creativity. Taking some time on your own to try new things, explore new tools, and try new workflows is what may help push your skills a step further.

Social media: Instagram, LinkedIn, ArtStation, and website.

Yoel Pereira López, Senior 3D Environment Modeler