Skip to main content

Best Blue Iris Settings (Real World)

Overview​

These are real-world Blue Iris settings designed for:

  • Stable multi-camera systems
  • Accurate AI detection
  • Reduced false alerts
  • Efficient storage usage

This is not theory — these settings are based on actual deployment behavior.

🎯 Objective​

  • Improve detection accuracy
  • Reduce CPU load
  • Optimize storage
  • Create predictable system behavior

📊 Related Setup Guides​

🧠 Core Principles​

  • Motion triggers first, AI confirms
  • Direct-to-disk reduces overhead
  • FPS should be controlled (not maxed)
  • Simpler systems are more stable

Video​

  • Resolution: Native camera resolution
  • FPS: 10–15 FPS
  • Codec: H.264 (preferred for stability)
  • Keyframe: 1–2 seconds

Recording​

  • Direct-to-disk: Enabled
  • Pre-trigger: 2–3 seconds
  • Combine/cut: 15–30 seconds

Motion Detection​

  • Use zones (not full frame)
  • Object size filtering enabled
  • Sensitivity moderate (not max)

AI Settings​

  • Use AI to confirm motion: Enabled
  • Confidence: ~60% starting point
  • Model: YOLOv8 (if available)
  • Use main stream for AI: Enabled

Storage​

  • Separate New vs Stored
  • Limit New folder size
  • Monitor alert folder growth

Performance​

  • Hardware acceleration: Enabled
  • Substreams: Enabled
  • Avoid maxing all cameras

⚠️ Common Mistakes​

  • Running max FPS on all cameras
  • Using full-frame motion detection
  • Relying on AI without good motion setup
  • Ignoring storage growth

🧠 When to Use These Settings​

Use these settings when:

  • Running 4+ cameras
  • Using AI detection (CodeProject or ONNX)
  • Experiencing false alerts
  • Seeing high CPU usage
  • Storage is growing faster than expected

These settings are designed for real-world systems — not lab conditions.

📊 Related Systems​

✅ Result​

A stable, predictable, and efficient Blue Iris system that performs well in real-world conditions.