Field Value Module dots-renderingVersion 3.32.3Effort highTools —
Keywords: 3D, DOTS, perspective, rendering
/t1k:unity:dots-rendering:perspective-3d
3D mesh rendering, LODGroup, skeletal animation, impostor billboard
Third-person camera, orbit camera, perspective projection
URP/Lit materials, shadow casting, light probes
Simulation : XZ plane (always). Standard 3D NavMesh pathfinding on XZ ground
Presentation : Perspective camera. 3D meshes (primitives or skeletal). LOD for performance
No coordinate remapping — what you see is what the simulation computes
Perspective camera with orbit/spring-arm or fixed angle
Adjust position based on battle AABB and FOV (same math as isometric)
Can use Cinemachine for production cameras
-> See references/camera-3d-guide.md
Meshes : Unity primitives (Capsule, Cylinder, Sphere) or Synty Polygon skeletal models
Shader : Universal Render Pipeline/Lit with per-unit color and smoothness
LODGroup : LOD0 = full 3D mesh, LOD1 = Amplify Impostor billboard
Shadows : Full shadow casting enabled (unlike 2D perspectives)
-> See references/rendering-3d-guide.md
NavigationAuthoring.EnableGrounding = true (default) — DOTSGroundingSystem applies Y
RangedAttackAuthoring.ArcHeight = 5f — parabolic arc projectiles
Standard NavMeshSurface with PhysicsColliders geometry
Full SonarAvoidance enabled (viable at 3D unit counts, typically < 500)
Skeletal: Animator + AnimatorController on mesh child (not DOTS-native yet)
Impostor: static billboard at LOD1 (no animation at distance)
Death: DeathAnimationSystem handles scale shrink (no alpha fade needed for 3D)
UnitPrefab (root — authoring components here)
├── Mesh (child — MeshFilter + MeshRenderer + LOD0)
│ └── Primitive mesh with URP/Lit material
├── LOD1_Impostor (child — MeshFilter + MeshRenderer)
│ └── Amplify Impostor mesh + tinted material
└── LODGroup (on root)
├── LOD0: transition at 0.15 → Mesh renderer
└── LOD1: transition at 0.0 (never cull) → Impostor renderer
var lodGroup = root. AddComponent < LODGroup >();
new LOD ( 0.15f , new [] { lod0Renderer }), // LOD0 = 3D mesh
new LOD ( 0f , new [] { impostorRenderer }), // LOD1 = impostor (never cull)
lodGroup. RecalculateBounds ();
// Per-unit material instance (don't modify shared atlas)
var tintedMat = new Material (baseMat);
tintedMat. SetColor ( "_BaseColor" , teamColor);
AssetDatabase. CreateAsset (tintedMat, folder + name + "_Tinted.mat" );
impostorRenderer.sharedMaterial = tintedMat;
# Issue Fix 1 Synty models: 5 SMRs + 50 bones per unit = 0.9 FPS at 100 units Use Amplify Impostor LOD1 at distance 2 Impostor atlas shared across units — team tint modifies all Create per-unit material instance with _BaseColor override 3 Animator on source model captures dynamic pose during impostor bake Disable Animator before baking impostors 4 Lightmap baking corrupts ChunkWorldRenderBounds with NaN Do NOT bake lightmaps for DOTS entities. Use realtime lighting 5 Mesh pivot at center, not feet — authoring offset needed Child mesh at (0, halfHeight, 0), root at feet
t1k-unity-dots-core-ecs — ECS fundamentals
t1k-unity-dots-combat-rpg — Combat, stats, navigation authoring components
t1k-unity-dots-rendering-graphics — RenderMeshArray, MaterialMeshInfo, LOD in ECS
t1k-unity-rendering-amplify-impostors — Impostor baking workflow, atlas config, URP shader
t1k-unity-dots-nav-battlefield — ArenaConfig, tree LOD with impostors
t1k-unity-rendering-urp — URP/Lit shader, shadow settings, pipeline config
t1k-unity-rendering-shadow-optimization — Shadow cascades, light probes
t1k-unity-dots-nav-agents-navigation — NavMesh pathfinding with grounding
t1k-unity-dots-core-performance — Draw call reduction, chunk utilization
File Coverage references/rendering-3d-guide.mdLOD setup, impostor integration, mesh rendering, materials references/camera-3d-guide.mdPerspective camera, orbit patterns, FOV-based framing
A visible change made with this skill (UI, sprites, 3D, shaders, VFX, animation, lighting, scene layout; audio excluded) follows core rule visual-change-discord-notice:
Capture OLD before the first edit and NEW after the last (screen: ScreenCapture in Play mode per t1k-unity-editor-cli, windowed or Xvfb only because rendering is needed; 3D, shader, VFX: same-scene frame grab; 2D asset: the file).
Return the paths and a draft Discord post to the orchestrator, which posts; the PR links the post or carries visual-notice-waiver: <reason>. Logic only: “no visible change”. No capture: “screenshots pending: ”, never a fake image.