Breakdown: Game-Ready Mining Drill Model Created with Blender & UE5

Introduction

Hey everyone! I’m Finn Bogaert, a 20-year-old second-year student at Digital Arts & Entertainment (DAE) of Howest University of Applied Sciences, studying the Game Graphic Production major with a Technical Art minor. I’m currently looking for an Environment/Tech art internship in Q1 2026! Feel free to connect with me on LinkedIn! My socials are the following: ArtStation, LinkedIn.

My interest in 3D and art started only recently. I joined DAE three years ago. With this being my first steps into the 3D world. I quickly fell in love with 3D art. What was first a “school study” instantly became my passion and obsession. I started working hard and doing my best to learn as much as possible, so I could eventually find an internship and work on amazing games in the future!

I’m still very new to the industry, but I'm eager to improve and learn new things every day. I hope this article inspires the newer artists in this industry to try out new things, explore different opportunities, and, most of all, not be scared of the challenges ahead of making their own projects. The project that I’m about to break down here was a tough challenge, since I challenged myself to go through the whole pipeline, start to finish.

The Daughter Drill

Inspiration & Concepting

This project was a rollercoaster. It was made for the Game Asset Pipeline subject at our school. The assignment was to make a game-ready asset (machine) and figure out your own pipeline. In this breakdown, I will talk mostly about my personal experience working through this project, how I found my way, and where I struggled. Since technically this definitely isn’t the best there’s out there, I think the journey still makes it worth the read. 

So I started looking for ideas. Personally, I am not that good at concepting, so this was a real struggle for me. I had no idea what I wanted to do, which direction I wanted to go, but I pushed through. Our school recommended using AI to come up with ideas, but this didn’t give me any results I wanted. Here’s a little look into my idea board:

Throughout my search, I kept going back to one specific concept; this one was by noax.design on ArtStation, for Concord:

I wanted my design to give off a similar feeling to the vibe I got from looking at this weapon, so I started concepting. My first thought was to find some concepts that I liked, some mechanical objects, and then I tried to photobash them together. I wanted to see if that could give me any results that I liked. After a while, I did come up with something that sparked my interest.

This final concept sparked my interest; I wanted to explore this further. Since concepting isn’t my strongest suit, I took it to 3D to explore more possibilities of what I could do with this idea. I took the concept and started modelling to make it look similar, but it didn’t work; a lot of the weight balance was wrong, the silhouette was boring, nothing made sense, and “flow” like I wanted it to. So I started iterating.

Modeling

Now that I had a concept I liked a lot, it was time to start modeling. While exploring the different workflows, I thought that making a “midpoly” as a base, and then making the highpoly and lowpoly both from the midpoly would work the best for me, so I started on that. For the midpol,y I wanted to make sure it was very easy to get to both the highpoly and lowpoly very easily, this meant: still have clean topology, make sure everything is still optimized.

The midpoly was fully created in Blender. Before this project, I used Maya, but I found Blender to be way quicker, together with the correct plugins like Machin3tools. I managed to speed up my workflow a lot, so making the midpoly was smooth sailing.

My approach to this was basic, so there’s not much I can share about this. Mainly working shape by shape, as the shapes I had were very simple, this process wasn’t too complex, and I could get through it pretty easily. 

One exception was the drillbit; for this, I had to find a way to copy them around in a logical order, while still being connected, and not having them deform too much. This was a tough challenge. I decided to do this with an array and curve modifier in Blender. 

To make sure this worked, I had to model one part, make sure it’s UV’ed and fully optimized, then array and curve it, since I wouldn’t be able to adjust it afterwards.

To create the high poly of this piece, I decided to use ZBrush, as this would make it way easier to get a good-looking high poly, without me having to intricately go through the trouble of making this with normal subdivision techniques.

For those who don’t know: ZBrush polishing is a really useful technique to make a smooth high poly from complicated meshes. This way, you don’t have to waste any time making polyloops, just have your shape, crease the edges, subdivide, and bring them to ZBrush.

In ZBrush, there’s an option to polish the edges, which makes for even smoothness across the whole meshpart -> easy highpoly creation. Many thanks to my friend Petru Grati for teaching me this technique. Here’s a tutorial he made that goes through the process:

Here’s a preview of what the high poly looked like before ZBrush, and after ZBrush. A big difference that only takes two minutes to do!

The rest of the high poly and low poly were mostly done manually from the mid poly. As seen in the pictures below, the shapes are pretty basic, so these could be done with normal Sub-D techniques. While modelling, I still tried to implement as many details as possible.

Retopology & Unwrapping

After the mid poly was finished, it was very easy to get to the low poly, just delete all unnecessary edges, and optimize the flat faces.

This way, I had my clean game-ready topology in an instant. The workflow I use in Blender also makes UV’ing very easy! I just marked all my sharp edges as UV seams, and then with only a few manual seams, I had my UV layout ready. I also made sure to use overlapping for all the small parts; this way, I could save some UV space for better texture resolution.

To get through the unwrapping process easily, I used the Textools and UVpackmaster plugins for Blender.

Texturing

After UV’ing, the first thing I had to tackle was the baking. This is the second Game Ready prop I’ve made so far, and the first one I made brought a lot of baking issues with it, so I was pretty scared of this step. Luckily, I didn’t have that many issues here. Since I decided to go with the midpoly to low and highpoly workflow, my LP and HP aligned perfectly, and there were not that many issues to fix.

Now that the bake is done, it’s time to texture! First, I went looking for some reference, what kind of textures do I want? I quickly found a lot of styles I liked. I decided to go more into the grungy-dirty route. I loved the yellow colors I was seeing in the reference, so I decided to go this route first.

Here’s a little GIF that shows the progression during the texturing phase: 

The way I usually like to work on my textures is like this:

First, I get a base in for everything; this way, I know roughly what the whole model will look like at the end. Once I have a base going, I go back into every material and start refining it.

I started by mapping out the materials, what materials do I want, do the colors look good, is it still readable from a distance, etc. I felt like I had most of that correct in this stage, only the color I wasn’t too sure about. 

After experimenting with the yellow materials for a while, I realised it didn’t work (In my opinion), I decided to go for a more desaturated, cream-ish color. This worked better and aligned more with what I envisioned with the asset. Making this transition was easy, as I had good values down; I could just desaturate it, and I instantly had the white I wanted, but this also allowed me to explore different color options. Sadly, nothing clicked with me as much as the white one did, so I went with that.

I build up all my materials from scratch. I look at my reference and try to replicate it. For example, take a look at my paint material below.

I start with a base layer, which defines the basic color and roughness. I quickly follow up with some base dirt, and then I add gradients to make the transitions read better and make it look better overall.

After that, I decided to use a lot of stencils, since I haven’t done that much before. I found working with stencils nice, it gives way more control than generators. I attached an image to see what difference the stencils make. Here’s a video that explains the process:

I also used stencils for the paint peeling, local damage, oil stains, etc. This was my main approach for most of the localized wear on the drill. I liked this workflow a lot, and I’ll use it in my further projects. 

Another thing I want to highlight is the welds on the drill bit. These welds were done fully in Substance Painter using the Weld Brush. This allowed me to break up the harsh edges, which I intentionally didn’t bake in. The welds made the drillbit look way better, covering all the seams and making everything feel “whole.”

Finally, here are some screenshots in Substance 3D Painter of my final texture. Quite happy with how it turned out, but definitely still room for improvement! 

Lighting & Rendering

The rendering was done in Unreal Engine 5. I decided on Marmoset Toolbag since I felt way more comfortable in Unreal Engine 5, as well as I wanted to do some scene renders, and I had a specific UE5 scene in mind for this. 

For my studio renders, I decided to go with a dark background. I felt like this worked well with the drill and allowed me to make it pop. I started with a 3-point lighting setup. This didn’t get me all the way there, but it was a good base to start.

I decided to make the drill suspended by chains, which added some extra composition elements to the render and made it that little bit more “cool.”

After that, I decided to use more lights to accentuate the areas that I wanted, to give the render more “pop.” I decided to use quite a lot of lights, but this didn’t matter as it was just for this 1 shot.

I decided to use this render as my main render. It is built from only spot lights. I built on the main three-point light setup, and then I filled and accentuated the areas that needed it with some more small spotlights. I used a similar lighting setup for my other studio renders, with the mesh moved around a bit.

I also wanted to explore some scene renders, since I always love seeing those on other people’s posts. I had an idea to present this in a mine, this would fit perfectly with the vibe I was trying to portray with the prop, and the grungy vibe with the whole asset.

Conclusion

This project as a whole took me about 150 hours. I struggled with the concept as concepting has never really been my strongest skill, but for this project, I aimed to push my boundaries and to learn new things. Also, not having any direct reference for what I wanted to make was complicated; texturing based on different references, and kind of mixing and matching everything together to make it look like it works, was an interesting experience.

I learnt so much throughout this process, and I want to convince everyone reading this to try a project once where you have no idea what to do, a project where you feel totally lost. Pushing through the unknown is something that taught me a lot, and the struggles I had throughout this project made me realise more what I want to do and learn in the future.

Thanks for reading through my article. Hopefully, there was at least some useful information in there that you guys could learn from! Thanks a lot to 80 Level for having me!

Feel free to connect with me or check out my socials: ArtStation and LinkedIn.

Finn Bogaert, Technical Environment Artist