r3f-performance
R3F performance optimization—LOD (Level of Detail), frustum culling, instancing strategies, draw call reduction, frame budgets, lazy loading, and profiling tools. Use when optimizing render performance, handling large scenes, or debugging frame rate issues.
What this skill does
# R3F Performance
Optimize render performance through draw call reduction, geometry optimization, smart loading, and profiling.
## Quick Start
```tsx
// Performance-optimized Canvas
<Canvas
dpr={[1, 2]} // Limit pixel ratio
performance={{ min: 0.5 }} // Adaptive performance
frameloop="demand" // Only render on change
gl={{
powerPreference: 'high-performance',
antialias: false // Disable for mobile
}}
>
<Suspense fallback={null}>
<Scene />
</Suspense>
</Canvas>
```
## Frame Budget
Target 60fps = 16.67ms per frame. Budget breakdown:
| Phase | Target | Notes |
|-------|--------|-------|
| JavaScript | < 4ms | useFrame logic, state updates |
| GPU Render | < 10ms | Draw calls, shaders |
| Compositing | < 2ms | Post-processing, overlays |
| Buffer | ~1ms | Safety margin |
## Draw Call Optimization
### The Golden Rule
**Fewer draw calls > fewer triangles**
A scene with 100 meshes of 1000 triangles each is slower than 1 mesh of 100,000 triangles.
### Reduction Techniques
| Technique | Draw Calls | When to Use |
|-----------|------------|-------------|
| Instancing | 1 per unique mesh | 100+ identical objects |
| Merged geometry | 1 per merged batch | Static scene parts |
| Texture atlases | Fewer materials | Many similar textures |
| LOD | Reduces complexity | Large/distant objects |
### Instancing (Best for Identical Meshes)
```tsx
// 10,000 cubes = 1 draw call
<instancedMesh args={[undefined, undefined, 10000]}>
<boxGeometry />
<meshStandardMaterial />
</instancedMesh>
```
### Geometry Merging (Static Scenes)
```tsx
import { useMemo } from 'react';
import { mergeGeometries } from 'three/examples/jsm/utils/BufferGeometryUtils';
import * as THREE from 'three';
function MergedScene() {
const mergedGeometry = useMemo(() => {
const geometries: THREE.BufferGeometry[] = [];
// Create many positioned geometries
for (let i = 0; i < 100; i++) {
const geo = new THREE.BoxGeometry(1, 1, 1);
geo.translate(
(Math.random() - 0.5) * 20,
(Math.random() - 0.5) * 20,
(Math.random() - 0.5) * 20
);
geometries.push(geo);
}
return mergeGeometries(geometries);
}, []);
return (
<mesh geometry={mergedGeometry}>
<meshStandardMaterial />
</mesh>
);
}
```
## Level of Detail (LOD)
Swap geometry based on camera distance:
```tsx
import { useMemo } from 'react';
import * as THREE from 'three';
function LODMesh() {
const lod = useMemo(() => {
const lodObject = new THREE.LOD();
// High detail (close)
const highGeo = new THREE.SphereGeometry(1, 64, 64);
const highMesh = new THREE.Mesh(highGeo, new THREE.MeshStandardMaterial({ color: 'red' }));
lodObject.addLevel(highMesh, 0);
// Medium detail
const medGeo = new THREE.SphereGeometry(1, 32, 32);
const medMesh = new THREE.Mesh(medGeo, new THREE.MeshStandardMaterial({ color: 'orange' }));
lodObject.addLevel(medMesh, 10);
// Low detail (far)
const lowGeo = new THREE.SphereGeometry(1, 8, 8);
const lowMesh = new THREE.Mesh(lowGeo, new THREE.MeshStandardMaterial({ color: 'green' }));
lodObject.addLevel(lowMesh, 30);
return lodObject;
}, []);
return <primitive object={lod} />;
}
```
### Drei LOD Helper
```tsx
import { Detailed } from '@react-three/drei';
function AdaptiveSphere() {
return (
<Detailed distances={[0, 10, 30]}>
{/* Close: high detail */}
<mesh>
<sphereGeometry args={[1, 64, 64]} />
<meshStandardMaterial />
</mesh>
{/* Medium distance */}
<mesh>
<sphereGeometry args={[1, 32, 32]} />
<meshStandardMaterial />
</mesh>
{/* Far: low detail */}
<mesh>
<sphereGeometry args={[1, 8, 8]} />
<meshStandardMaterial />
</mesh>
</Detailed>
);
}
```
## Frustum Culling
Objects outside camera view are not rendered. Enabled by default, but:
```tsx
// Disable for objects that animate into view unpredictably
<mesh frustumCulled={false}>
<boxGeometry />
<meshStandardMaterial />
</mesh>
// Force bounding sphere update for dynamic geometry
useEffect(() => {
geometry.computeBoundingSphere();
}, [geometry]);
```
## Adaptive Performance
R3F's adaptive performance system automatically adjusts DPR:
```tsx
<Canvas
performance={{
min: 0.5, // Minimum DPR under stress
max: 1, // Maximum DPR
debounce: 200 // Debounce time for changes (ms)
}}
/>
```
### Manual Performance Control
```tsx
import { useThree } from '@react-three/fiber';
function PerformanceMonitor() {
const { performance } = useThree();
useFrame(() => {
// Check current performance
if (performance.current < 1) {
// System is under stress, reduce complexity
}
});
// Trigger performance drop
const triggerRegress = () => {
performance.regress(); // Temporarily lower DPR
};
}
```
## Lazy Loading
### Code Splitting with Suspense
```tsx
import { Suspense, lazy } from 'react';
const HeavyModel = lazy(() => import('./HeavyModel'));
function Scene() {
return (
<Suspense fallback={<SimpleLoader />}>
<HeavyModel />
</Suspense>
);
}
```
### Progressive Loading
```tsx
import { useGLTF } from '@react-three/drei';
function Model() {
// Preload in background
useGLTF.preload('/model.glb');
const { scene } = useGLTF('/model.glb');
return <primitive object={scene} />;
}
// Preload before component mounts
useEffect(() => {
useGLTF.preload('/next-model.glb');
}, []);
```
### View-Based Loading
```tsx
import { useInView } from 'react-intersection-observer';
function LazySection() {
const { ref, inView } = useInView({
triggerOnce: true,
rootMargin: '200px' // Start loading 200px before visible
});
return (
<group ref={ref}>
{inView && <HeavyContent />}
</group>
);
}
```
## Memory Management
### Dispose Unused Resources
```tsx
// Manual disposal
useEffect(() => {
return () => {
geometry.dispose();
material.dispose();
texture.dispose();
};
}, []);
// Drei helper for GLTF
import { useGLTF } from '@react-three/drei';
useEffect(() => {
return () => {
useGLTF.clear('/model.glb');
};
}, []);
```
### Texture Optimization
```tsx
import { useTexture } from '@react-three/drei';
import * as THREE from 'three';
// Compress and optimize
const texture = useTexture('/texture.jpg', (tex) => {
tex.minFilter = THREE.LinearMipmapLinearFilter;
tex.generateMipmaps = true;
tex.anisotropy = 4; // Lower = faster, higher = sharper
});
// Use compressed formats (KTX2)
import { useKTX2 } from '@react-three/drei';
const texture = useKTX2('/texture.ktx2');
```
## Profiling
### Stats Panel
```tsx
import { Stats } from '@react-three/drei';
<Canvas>
<Stats /> {/* FPS, MS, MB counters */}
<Scene />
</Canvas>
```
### Performance Panel
```tsx
import { Perf } from 'r3f-perf';
<Canvas>
<Perf
position="top-left"
showGraph // Show FPS graph
minimal={false} // Full or minimal view
/>
<Scene />
</Canvas>
```
### Manual Profiling
```tsx
import { useThree } from '@react-three/fiber';
function ProfileInfo() {
const { gl } = useThree();
useEffect(() => {
const info = gl.info;
console.log({
drawCalls: info.render.calls,
triangles: info.render.triangles,
points: info.render.points,
lines: info.render.lines,
textures: info.memory.textures,
geometries: info.memory.geometries
});
});
return null;
}
```
### Frame Time Measurement
```tsx
function FrameProfiler() {
const frameTimeRef = useRef<number[]>([]);
useFrame(() => {
const start = performance.now();
// ... your logic ...
const elapsed = performance.now() - start;
frameTimeRef.current.push(elapsed);
if (frameTimeRef.current.length > 60) {
const avg = frameTimeRRelated in Image & Video
watch
IncludedWatch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or Whisper API fallback), and hands the result to Claude so it can answer questions about what's in the video.
physical-ai-defect-image-generation
IncludedUse when the user wants to orchestrate defect image generation, run associated setup, or handle outputs on OSMO. The Day 0 path handles cold-start with USD-to-ROI, image-edit augmentation, and AnomalyGen to create initial PCBA datasets. The Day 1 path performs inference and labeling on real images. This skill helps with first-time asset setup, creation of finetuning checkpoints, and configuring deployment. Trigger keywords: defect image generation, dig workflow, dig pipeline, defect image detection workflow, aoi pipeline, aoi anomalygen, usd2roi anomalygen, day 0 pcba, day 1 pcba, day 1 real-photo alignment, day 1 manual roi, metal surface anomaly, glass defect, anomalygen finetune, setup_pcb, setup_metal, setup_glass, setup_pretrained, dig setup, dig datasets, dig pretrained checkpoint, dig image-edit endpoint.
accelint-react-best-practices
IncludedReact performance optimization and best practices. ALWAYS use this skill when working with any React code - writing components, hooks, JSX; refactoring; optimizing re-renders, memoization, state management; reviewing for performance; fixing hydration mismatches; debugging infinite re-renders, stale closures, input focus loss, animations restarting; preventing remounting; implementing transitions, lazy initialization, effect dependencies. Even simple React tasks benefit from these patterns. Covers React 19+ (useEffectEvent, Activity, ref props). Triggers - useEffect, useState, useMemo, useCallback, memo, inline components, nested components, components inside components, re-render, performance, hydration, SSR, Next.js, useDeferredValue, combined hooks.
elevenlabs-agents
IncludedBuild conversational AI voice agents with ElevenLabs Platform using React, JavaScript, React Native, or Swift SDKs. Configure agents, tools (client/server/MCP), RAG knowledge bases, multi-voice, and Scribe real-time STT. Use when: building voice chat interfaces, implementing AI phone agents with Twilio, configuring agent workflows or tools, adding RAG knowledge bases, testing with CLI "agents as code", or troubleshooting deprecated @11labs packages, Android audio cutoff, CSP violations, dynamic variables, or WebRTC config. Keywords: ElevenLabs Agents, ElevenLabs voice agents, AI voice agents, conversational AI, @elevenlabs/react, @elevenlabs/client, @elevenlabs/react-native, @elevenlabs/elevenlabs-js, @elevenlabs/agents-cli, elevenlabs SDK, voice AI, TTS, text-to-speech, ASR, speech recognition, turn-taking model, WebRTC voice, WebSocket voice, ElevenLabs conversation, agent system prompt, agent tools, agent knowledge base, RAG voice agents, multi-voice agents, pronunciation dictionary, voice speed control, elevenlabs scribe, @11labs deprecated, Android audio cutoff, CSP violation elevenlabs, dynamic variables elevenlabs, case-sensitive tool names, webhook authentication
humanizer
IncludedHumanize AI-generated text by detecting and removing patterns typical of LLM output. Rewrites text to sound natural, specific, and human. Uses 28 pattern detectors, 560+ AI vocabulary terms across 3 tiers, and statistical analysis (burstiness, type-token ratio, readability) for comprehensive detection. Use when asked to humanize text, de-AI writing, make content sound more natural/human, review writing for AI patterns, score text for AI detection, or improve AI-generated drafts. Covers content, language, style, communication, and filler categories.
generating-mermaid-diagrams
IncludedSalesforce architecture diagrams using Mermaid with ASCII fallback. Use this skill when generating text-based diagrams for Salesforce architecture, OAuth flows, ERDs, integration sequences, or Agentforce structure. TRIGGER when: user says "diagram", "visualize", "ERD", or asks for sequence diagrams, flowcharts, class diagrams, or architecture visualizations in Mermaid. DO NOT TRIGGER when: user wants PNG/SVG image output (use generating-visual-diagrams), or asks about non-Salesforce systems.