3D Modeling Best Practices

A CAD model is not complete when it looks right. It is complete when it can change without breaking. Too many models are built like sketches with volume. Features added in whatever order is convenient. Dimensions tied to edges that might disappear. Patterns referencing geometry that will not survive revision. It works… until revision two. If you want a model that survives real engineering changes, the process matters. Here is the discipline I follow: Start with design intent, not geometry. What surfaces control function? What interfaces matter? Build stable references first. Primary datums. Core planes. Base features that will not disappear when details change. Create the fundamental shape before adding detail. Extrusions and revolves that define envelope and mass. Add functional features next. Holes, pockets, bosses tied to stable datums, not random edges. Pattern and mirror late in the tree. Patterns should depend on controlled sketches or datums, not finished faces. Apply fillets and cosmetic features last. If fillets fail, your model should still rebuild. Every dimension should answer one question. If this changes, what else must move with it? Parametric modeling is not about clicking tools. It is about building logic into the part. After decades in manufacturing and CAD, I have learned that the real test of a model is not how it looks on release day. It is how calmly it handles revision three. For engineers and designers. What is the most common rebuild mistake you see in poorly structured models?

𝐀 𝐜𝐥𝐞𝐚𝐧 𝐌𝐚𝐲𝐚 → 𝐔𝐧𝐫𝐞𝐚𝐥 𝐄𝐧𝐠𝐢𝐧𝐞 𝐰𝐨𝐫𝐤𝐟𝐥𝐨𝐰 𝐦𝐞𝐚𝐧𝐬 𝐛𝐞𝐭𝐭𝐞𝐫 𝐪𝐮𝐚𝐥𝐢𝐭𝐲, 𝐟𝐞𝐰𝐞𝐫 𝐞𝐫𝐫𝐨𝐫𝐬, 𝐚𝐧𝐝 𝐟𝐚𝐬𝐭𝐞𝐫 𝐫𝐞𝐧𝐝𝐞𝐫𝐬. Most visual problems don’t come from Unreal Engine. They come from a broken pipeline between tools. Save this if you want a smooth, studio-level workflow. 1. Modeling in Autodesk Maya (Foundation Stage) • Work in real-world scale (centimeters). • Maintain clean topology with no ngons or non-manifold geometry. • Freeze transforms before export. • Delete construction history. • Set correct and logical pivot positions. If scale or pivots are wrong in Maya, Unreal Engine User Group will amplify the problem. 2. UVs and Textures (Quality Lives Here) • Use a single clean UV set unless UDIMs are required. • Avoid unwanted overlaps. • Maintain consistent texel density. PBR texture set: • Base Color. • Roughness. • Normal map (OpenGL format). • Metallic. Most “bad lighting” issues are actually texture problems. 3. XGen Hair to Unreal Engine • Do not export raw XGen geometry directly. • Convert XGen to cards or groom properly. • Use Unreal Engine Groom system for cinematic characters. • Keep hair density realistic to control performance. Unreal hair quality depends more on density and lighting than on the groom itself. 4. Export from Maya (Non-Negotiable) FBX export settings: • Units set to centimeters. • Smoothing groups enabled. • Tangents and binormals enabled. • Avoid unnecessary animation baking. One incorrect export setting can completely break shading. 5. Importing into Unreal Engine • Verify scale immediately after import. • Check normals and smoothing accuracy. • Assign correct material instances. • Disable auto exposure while working. Never begin lighting before materials are correct. 6. Lighting in Unreal Engine • Decide on one motivated primary light source. • Use fewer, larger lights instead of many small ones. • Lock exposure before final lighting. • Use Lumen or Ray Tracing intentionally, not blindly. Flat lighting ruins realism faster than low-quality models. 7. Camera and Color Control • Set realistic camera values for FOV and aperture. • Use filmic or ACES color pipeline. • Avoid heavy bloom and excessive sharpening. If a shot only looks good after post-processing, the lighting is weak. 8. Rendering for Best Quality • Test renders without denoiser first. • Reduce noise through proper light balance. • Render short test shots before final sequences. Clean lighting always renders faster and looks better. "Studio Rule That Never Fails" • Maya builds form. • Unreal Engine builds mood. • Mixing responsibilities breaks pipelines. When the workflow is clean, imports are predictable, lighting stays stable, and renders look cinematic instead of game-like. #maya #unrealengine #autodesk #xgen #3danimation #vfxpipeline #lightingworkflow #cgartist #cinematicrender

There is no single “correct” topology or polycount that works for every project. One of the biggest mistakes artists make is trying to follow universal rules without understanding the actual needs of the project. Topology and polycount always depend on where and how the model will be used. A cinematic sculpt can be millions of polygons. A game-ready asset must be optimized. A prop might be all triangles. A high-end character might need clean quad flow only in deformation areas—and nowhere else. Optimization doesn’t mean “low poly.” It means using only as much geometry as the project requires. If something works visually, deforms correctly, bakes well, and performs properly in its target engine or pipeline, then it is correct—regardless of whether it’s quads, triangles, or dense meshes. Ask yourself: Will this model be animated or static? Does this area deform? Is this for games, film, or sculpting only? What level of detail does the project actually need? 📌 Pro tip: don’t optimize blindly and don’t overbuild either. Use geometry where it matters—silhouette, deformation, and readability—and simplify everything else. There are no absolute rules in topology—only solutions. If it solves the problem and meets the project’s requirements, then it’s the right choice. 🧩⚙️ #Topology #3DWorkflow #GameArt

Whether designing 3d models for large stone architectural pieces or something small enough to fit into the palm of your hand there are a lot of similarities in the design process, and of course, there are some differences. Both kinds of pieces start out the same with making sure that with whatever 3D program is being used that the settings are scaled correctly. If you are trying to match a cad drawing background reference with it, then this step is extremely important to make sure that things are lining up correctly from the start because that can be a pain to change later on. From there you make sure that each piece is being designed with max editability. For myself throughout the design process, I'm always thinking about things like "what if my client wants to add a bit more depth to this area?" or "what if my client wants this part to rotate a bit?" etc. So it's best to design with revisions in mind, just in case. Another thing to keep in mind is the thickness of parts. For CNC milling, depending on the type of material(limestone, marble, etc.), you want to make sure that the parts that protrude from the building, memorial, etc. are thick enough so that it won't break from the CNC or hand milling process, or be worn down quickly by the weather(rain, snow, etc.). For 3D printing it's kind of the same concept since, depending on the type of 3D printing material that you're using, parts can break off during the printing process, or when removing the supports, or simply when being handled because parts can become a bit brittle or worn out over time. For CNC milled stone, I usually can't add as many small details as I can for my 3D printed pieces or coin designs, mainly because the large CNC machine won't fully pick them up or those details won't even be visible from the ground. I usually also keep an eye on undercuts, especially if it's for something like coins(which also require a small draft around the edges so that they can be pulled from a mold properly). These three images show a unique CNC milled limestone keystone piece that I'd 3D modeled for Yale University's newer buildings around 2016 or so, and the smaller piece is my own design for my personal brand that I'd 3D printed in resin. For that piece I was curious to see how much of the details from the brush strokes and acanthus flourishes would show through at that scale. #cnc #cncmilling #stonework #construction #3dprinting #limestone #marble #architecture #fabrication #production #digitalsculpting #3dmodeling #zbrush #maxon #yale