Modeling & Texturing a Detailed Colt Single Action Army

Introduction

Hi, my name is Daniel Midghall, and I'm a mid-level Hard-Surface Artist working at Expression Games. This is my first role in the industry, and I've been here for almost 3 years now. So far, I have worked on live service content for the WW2 first-person shooter Hell Let Loose, as well as working on the recently announced Hell Let Loose: Vietnam, which has been an amazing (and challenging) experience.

I never actually planned on working in the games industry, and like many others, I ended up enrolling in a games development course shortly before the end of high school, as I had no real notion of what I wanted to do with my life (save for digging up dinosaurs, but I was too busy playing Mass Effect 2 for the 12th time to take that seriously).

I feel like a lot of people think that there must be some special sauce or button to press that they have not found that will elevate their skill set dramatically, and while that would be awesome, it's sadly not the case. To grow my skillset over the course of my career, I found that it comes down to three things:

• Practice: It's the number one thing that any professional will tell you, and there's a reason for that, it's because it really is true. As humans, we naturally get better at something the more that we do it, we remember things to avoid and pick up on things that maybe we didn't quite get the first time. That metal finish that you didn't quite crack the first time will look a lot better the next time you get around to trying it, as long as you just keep trying and learning.

• Surround yourself with like-minded people: A lot of the early momentum in my 3D journey was gained from late nights spent with friends in Teamspeak and Discord, just modeling and texturing while yapping away with one another. It really pays to have people that you can bounce ideas off of, support, and motivate. One particular person I'd like to shout out here is my lead, Dan Conroy. I owe that man a hell of a lot, both for giving me the opportunity I currently have and for mentoring and giving me the confidence to grow as an artist and as a game developer.
• Feedback is a blessing, not a curse: To have someone criticize something you've worked hard on can suck, but it's important to know this happens at all skill levels, no matter if someone is a junior or a lead. To be able to be proud of something but also acknowledge the areas that can be improved is a very useful skill to have in this industry, and a surefire way to become a better artist. No one is "too good" for feedback at the end of the day, so always leave your ego at the door.

Beginning the Colt Single Action Army

I decided I wanted to tackle something smaller-scale and quicker to make this time around, compared to my last project, which was far more complex. I chose the Colt Single Action Army because of the wide array of finishes that it comes in, and frankly, because it's just a really cool, iconic piece.

I used Miro for my reference gathering as it's what I prefer using (although PureRef has won my heart of late for work tasks). I made sure to split my ref into different sections according to the parts of the weapon to have some semblance of organization on this board. Originally, I started out intending to do a variety of finishes for this piece, but after seeing some reference images of a case-hardened Jan Gwinnell-engraved version of the SAA, I knew that that was the one I had to replicate due to how distinctive the finish was.

Admittedly, I had to come back to ref gathering a few times along the course of the project as I kept changing the scope of what I wanted to deliver, when really it would have been best to nail this all down at the beginning. Consider this a "do as I say, not as I do" kind of thing.

Finding a ref can be a tricky one, especially with firearms. They can vary wildly from obscure pieces that have barely any usable ref to widely known pieces that have almost too much ref, which can make parsing the right information from them a nightmare. I recommend perusing through forums, gunbroker sites, subreddits, and YouTube videos to try and get as complete a picture as you can about the variations of what it is you're going to make. This way, you get to know what little idiosyncrasies exist between them and what to watch out for.

I would also heavily encourage anyone looking to get into making weapons to download and use World of Guns Disassembly on Steam. It's an incredibly useful game that lets you break down weapons to their individual components, and while it might not be 100% accurate, it's very close in most cases and can provide you with invaluable knowledge of how firearms function in general, and how the specific one you've chosen works too. Just as a character artist benefits immensely from knowledge of anatomy, a hard surface or weapons artist benefits heavily from being able to understand how a firearm works instead of just merely what it looks like.

Modeling

Modeling wasn't too difficult, mostly a mix of SubD, some Blender remeshing, and some ZBrush sculpting. I don't find myself beholden to any one workflow, and I'm happy to switch and learn new ones on the fly, as ultimately they're just tools at the end of the day. I began with a blockout mesh taken to mid-poly level detail, leveraging heavily on booleans and bevels to get as many curves and indents as I could while remaining as non-destructive as possible.

Generally, starting with cylindrical objects like the barrel or cylinder, modeling to a realistic scale, and modeling over the top of a reference image are all good ways to keep your mesh looking to scale.

Once I got my blockout to a point I was satisfied with, it was just a case of refining those parts into high-poly pieces. This was a pretty straightforward mix of traditional SubD, as well as using Blender's remesh modifier workflow for some more complex pieces like the cylinder.

I decided I wanted to do a ZBrush pass on most of the metal parts anyway, as I really wanted to go overboard on this one and have the larger cuts and bits of wear actually affect the low-poly. I also used this as an opportunity to add in the crosshatching found on the hammer, since I knew I wanted at least some of that detail to be present in the low-poly. I didn't actually sculpt any of the engravings at this stage, as I wasn't originally planning on having engravings yet.

Low-Poly and UVs

After this, I moved on to low-poly creation. This was very simple for the most part, as it was mostly just refining my high-poly into a more workable lower-poly mesh. Some sections, like the grip and parts of the frame, were shrink-wrapped to the high-poly to factor in some of the bigger cuts to spice up the silhouette and make it look a lot more interesting, and a lot of parts were beveled in areas where there was some heavier edge damage to avoid hard edge seams running through them.

The final tri count and texture set ended out way higher than what would be acceptable in a production environment. Part of the joy of personal work is being freed from the shackles of having to make your work performant, within reason, of course. After finishing up my low-poly, I decided on how I was going to allocate my texture sets, of which there ended up being 4:

• Cylinder
• Frame
• Movables
• Grips

UVs were a pretty simple affair. I relied mainly upon UVPackmaster, ZenUV, as well as plugins like Textools to help with visualizing TD and assigning texture sets. I'd say these addons are borderline mandatory for unwrapping within Blender, as they make life so much easier compared to the out-of-the-box tools.

Because of how many texture sets I was using, it was super easy to keep a consistent texel density across the asset, although there were still some areas (such as the insides of the frame) that I chose to reduce the resolution of to eek out that little bit extra resolution for the areas that would be prominently seen in renders.

Texturing

After UVs, it was time to bake out my Mesh Maps, which were tackled in Marmoset Toolbag 5. The ability to skew paint normals and see your baked maps update in real-time is a game-changer. It feels super tactile and intuitive to use, and has honestly made it almost impossible for me to go back to baking in Painter.

Now for the part you're probably here for. Textures were handled within Substance 3D Painter, as it is the industry standard for this type of work. The first thing I did after getting my Mesh Maps baked out was add a layer that would sit atop all the other ones, which contained an Ambient Occlusion Map that had been baked with all of the pieces reassembled to add some extra punchy AO to the asset, as just baked AO tends to get a bit lost once you start with textures.

You have to be careful with this, as too much extra AO can make your asset look a bit cartoonish. Add just enough to give some subtle oomph to the asset without being too extreme. After this, I added in my painted height details, things like the grip checkering and text were pretty simple to take care of as they were just fill layers with either positive or negative height depending on what effect I was trying to replicate.

I then added a black mask to said layer to paint on to create the desired effect. After creating these layers, I made sure to anchor them to be able to refer back to them later to add highlights, recess definition, and to have them interact generally with layers that required it.

It was around this time that I decided to commit to tackling the engravings for the asset. To accomplish this, I used ref images that best showed side-on angles of the engravings and added them to the base color slot of a few fill layers, and then used the "Warp Project" projection mode in order to get the ref to roughly conform to the proportions of the mesh.

From this point, I was able to then create some more height layers for the filigree. For the actual pattern outline, it was mostly just tracing and refining the overlaid reference images as best I could with an Ink brush that had reduced spacing. This wasn't necessarily difficult, so much as it was time-consuming. I then added a few filters/levels nodes in order to add some extra interest to the filigree so it read a bit more like the ref images.

The ref also included these little balls nestled in some of the filigree. Creating those was pretty simple. I just created a fill layer with some height, multiplied a texture over the top that resembled the balls in the ref, added some filters to further aid the resemblance, and then simply painted them in the areas in which they belonged.

Following on from this came the hard part, replicating the case-hardened finish. This is a very unique heat treatment finish that's achieved by heating the metal with a mix of things like charcoal, leather, and bone in order to create that iconic nebula-like finish. Alas, I had access to none of those things, so some head scratching in Substance 3D Painter would have to suffice.

I originally tried the same technique I used for creating painted tank metal on Hell Let Loose, which used various fill layers referencing a base mask with different level clamps to isolate where I wanted the specific colors to be. This kind of worked, but it did not give me the effect I really wanted for this particular piece.

I happened to stumble across a post from user Sakata_Toji, who is also making a Colt Single Action army (which looks awesome, you should absolutely go and follow his progress as his is coming along fantastically). In his post, he made use of a gradient filter atop, a UV Border Generator, with a few grunges overlayed and a passthrough paint layer in order to refine the mask. Seeing this practically lit a lightbulb above my head, and gave me the idea to try this for myself and see if I could get the effect I wanted this way. Thankfully, it worked out amazingly!

I also cloned this layer and made an alternate version once I was happy with the mask. This alt version had a reduced amount of colors in the gradient while also making use of some different color tones, as some areas of the ref had these softer, greener areas that I didn't want going across the whole model.

After I was happy with the base case hardening finish, it came time to add some real character to the textures. A problem I ran into here is that the ref I was using was very clean and pristine. Makes sense, right? You don't want to drop thousands on a bespoke piece like this in real life just to scuff and bang it up. This, however, would look a bit plain and boring in a game art sense, so I consulted other ref images in order to inform things like surface noise, specs, general imperfections/discoloration, as well as things like machining.

After the metal came the wood material for the grips. Wood can be pretty challenging when you haven't replicated it much before, but it's actually pretty simple once you get the base logic down.

Working from a planar projected wood pattern (the basic one Painter comes with will do) with a directional noise multiplied over the top is a solid starter for a lot of wood. The nuance here comes from getting the planar projection at just the right angle so that the bands get projected correctly across your mesh, so you may have to set up a few different wood projection layers and then consolidate them together with Anchor Points, like I have done above.

The grain was pretty simple too, just a few projected directional noise layers of different widths that were painted according to how I saw them on my ref. Generally, the grain on wood like this gets a lot more heavily tiled and rounder when you get to round cut areas like the bottom of these grips, for instance, but this will obviously depend on the kind of wood you're replicating and how it has been cut/treated.

The rest of the wood was pretty simple, mostly just referring back to my previous Anchor Points and overlaying grunges in order to build up subtle height and roughness.

Getting your material to read correctly can be a really daunting and hard task, especially for newer artists. Studying your ref is a key aspect. Not just merely looking at it but analyzing it, paying attention to the nuances such as specs, subtle hue changes, height, grain, and taking a moment every so often to compare what you're making to the reference, and asking yourself, "Does this actually resemble the ref?" This is where feedback and peer review are critical, getting fresh eyes on it from more experienced artists can often be what takes an asset from looking good to looking great.

After creating my base materials, it was time to finish up with a wear pass. I like to work in stages like this as it's very, very easy to get lost in the sauce too early by throwing in wear and dirt and grime before you've got a good base to build off of, and this can often result in some pretty noisy texturework. I didn't want to go too heavy, as I still wanted this to come across as a valuable antique piece, but a nice helping of edge wear, oils, and some dirt can ground your asset and really tell a story.

Nothing too special with how I went about tackling edge wear, mostly just hand-painted stencilwork. Same with the oils. The dirt/hairs and whatnot were a mix of hand painting, but also using an AO generator as a base, as well as using my previous anchor points from the text and filigree in order to get dirt going into those crevices, too. There's a fantastic method of doing dusty dirt by Adrien Roose, which I took a lot of inspiration from.

Rendering

By this point, I was definitely starting to run on fumes. Luckily, there was still enough in the tank to be able to quickly make a display case and stand for this piece. I'm a huge fan of diorama shots nowadays, as I really think they can elevate your presentation above the rest if they're done correctly.

I hate working with complex lighting setups and prefer to keep things simple. Based on my key ref image, I knew I wasn't going to need anything more than a basic HDRI and a Spotlight in order to get that light falloff, as well as fine-tune the diameter and size of the Spotlight to get the exact reflections I needed. Generally, the rule is that the wider the light, the flatter the reflections, whereas the smaller and thinner the light, the sharper the reflections will be.

As for post-processing, my setup was pretty standard. ACES tone mapping, depth of field when the shot called for it, some light grain, as well as general tweaks to contrast/exposure as the shot demanded. I do most of my shots at a 120mm lens, as that just tends to look the best for creating proper render shots in my opinion. I also added a subtle fog to the scene too just to help with light dispersion.

Conclusion

This project was definitely a lesson in the importance of proper planning for me. Truth be told, a lot of decisions like the engravings, finish, and the diorama at the end were made in the midst of the project when, in an ideal world, I really should have nailed all that down at the beginning, so I could have been much more efficient with my time. Still, I think this piece came out very well and is a great statement for what I can do as an artist.

In terms of advice for beginners, I would say stick to your guns (literally and figuratively). A good chunk of my early game developer career was marred by indecision and doubts about being able to be a weapon artist, and I instead spent a good bit of time vying to be an Environment Artist based on people's insistence that that was the discipline that would get me in the industry the fastest.

Weapons and hard surfaces are my passion, and once I leaned into that and embraced it, I was able to get myself up to a level where people started to take notice of me quite quickly. I guess that's a roundabout way of saying make the stuff that you love and find cool, as opposed to what you think other people want to see, and eventually you'll get to a point where people will take notice of you.

Daniel Midghall, Concept Designer & 3D Artist

Interview conducted by Emma Collins