⚙️ OS Configuration - Real-Time Audio Optimization¶

Critical: Modern OS audio enhancements destroy the precise timing required for audio plugin development. This guide disables all interfering features.

Goal: Achieve deterministic audio processing with minimal latency jitter.

🪟 Windows 10/11 Optimizations¶

Windows audio subsystem has layers of "enhancements" that introduce latency, jitter, and unpredictable behavior. Every enhancement must be disabled.

1️⃣ Audio Exclusivity & Enhancements¶

Why? Windows audio enhancements (EQ, reverb, loudness) add 10-30ms latency and process audio in the background, interfering with your plugin's processing.

Step-by-Step Configuration¶

Navigate to Sound Settings:

Control Panel → Hardware and Sound → Sound

Or use Run dialog (Win+R):

mmsys.cpl

For Each Audio Device (Speakers, Headphones, Line Out):

Right-click device → Properties
Navigate to Advanced tab

Enable these options:

☑️ Allow applications to take exclusive control of this device
☑️ Give exclusive mode applications priority

Set Default Format:
```
24 bit, 48000 Hz (Studio Quality)
```
⚠️ Do NOT use 44100 Hz - Most interfaces default to 48kHz, mismatches cause resampling latency.
Navigate to Enhancements tab

Critical: Check this box:

☑️ Disable all sound effects

Or individually disable:

☐ Bass Boost
☐ Virtual Surround
☐ Room Correction
☐ Loudness Equalization
☐ (All other enhancements)

Click Apply → OK

Verification¶

# Check if exclusive mode is enabled (PowerShell)
Get-ItemProperty -Path "HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\MMDevices\Audio\Render\*" | Select-Object -Property *Exclusive*

Expected: Both AllowExclusive and PriorityExclusive should be 1.

2️⃣ Power Settings (Critical for Real-Time)¶

Why? CPU throttling and USB power saving introduce latency spikes (5-50ms) that cause audio dropouts.

Power Plan Configuration¶

Option A: High Performance Plan (Recommended)

Open Power Options:

Control Panel → Hardware and Sound → Power Options

Or Run dialog (Win+R):

powercfg.cpl

Select High performance plan
If not visible, click Show additional plans
Click Change plan settings → Change advanced power settings
Configure these settings:

🔋 Processor power management
   ├─ Minimum processor state: 100%
   └─ Maximum processor state: 100%

🔌 USB settings
   └─ USB selective suspend setting: Disabled

🎨 PCI Express
   └─ Link State Power Management: Off

💤 Hard disk
   └─ Turn off hard disk after: 0 (Never)

🖥️ Display
   └─ Turn off display after: 10 minutes (or your preference)

💤 Sleep
   ├─ Sleep after: Never
   └─ Hibernate after: Never

Click Apply → OK

Option B: Ultimate Performance Plan (Windows 10 Pro/Enterprise)

Even better than High Performance. Disables micro-power-saving features.

# Enable Ultimate Performance plan (Run as Administrator)
powercfg -duplicatescheme e9a42b02-d5df-448d-aa00-03f14749eb61

# Set as active
powercfg -setactive e9a42b02-d5df-448d-aa00-03f14749eb61

# Verify
powercfg /list

Verify Power Plan¶

powercfg /query | findstr /C:"Power Scheme GUID" /C:"GUID Alias"

Expected output:

Power Scheme GUID: e9a42b02-d5df-448d-aa00-03f14749eb61  (High performance)

3️⃣ Real-Time System Tweaks¶

A. Disable Windows Search Indexing (Project Folders)¶

Why? Search indexing locks files during builds, causing compilation errors.

Method 1: Per-Folder 1. Right-click your project folder (e.g., C:\Dev\AudioLab) 2. Properties → Advanced... 3. Uncheck:

☐ Allow files in this folder to have contents indexed

4. Click OK → Apply to all subfolders

Method 2: Exclude from Indexing Service 1. Control Panel → Indexing Options 2. Click Modify 3. Uncheck your development drives/folders 4. Click OK

B. Exclude Build Folders from Windows Defender¶

Why? Real-time scanning of build artifacts adds 50-200ms per file. A full rebuild can take 3x longer.

Add Exclusions:

# Run PowerShell as Administrator

# Exclude build folders
Add-MpPreference -ExclusionPath "C:\Dev\AudioLab\build"
Add-MpPreference -ExclusionPath "C:\Dev\AudioLab\cmake-build-*"

# Exclude JUCE folder (if local)
Add-MpPreference -ExclusionPath "C:\SDKs\JUCE"

# Exclude Visual Studio cache
Add-MpPreference -ExclusionPath "$env:LOCALAPPDATA\Microsoft\VisualStudio"

# Verify exclusions
Get-MpPreference | Select-Object -ExpandProperty ExclusionPath

⚠️ Security Note: Only exclude paths you control. Never exclude system folders.

C. Registry Tweaks for MMCSS (Multimedia Class Scheduler Service)¶

Why? MMCSS prioritizes audio threads. We need to increase the priority boost.

⚠️ Warning: Editing registry incorrectly can break Windows. Backup before proceeding.

Backup Registry:

# Create registry backup
reg export "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Multimedia\SystemProfile" "%USERPROFILE%\Desktop\MMCSS_backup.reg"

Apply Tweaks:

# Run as Administrator

# Increase MMCSS priority for Pro Audio
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Multimedia\SystemProfile\Tasks\Pro Audio" /v "Priority" /t REG_DWORD /d 1 /f
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Multimedia\SystemProfile\Tasks\Pro Audio" /v "Scheduling Category" /t REG_SZ /d "High" /f
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Multimedia\SystemProfile\Tasks\Pro Audio" /v "SFIO Priority" /t REG_SZ /d "High" /f

# Increase system responsiveness (reduces background tasks)
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Multimedia\SystemProfile" /v "SystemResponsiveness" /t REG_DWORD /d 10 /f

# Increase network throttling index (reduces network interference)
reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Multimedia\SystemProfile" /v "NetworkThrottlingIndex" /t REG_DWORD /d 4294967295 /f

Reboot Required: Restart Windows for changes to take effect.

Verify:

reg query "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Multimedia\SystemProfile\Tasks\Pro Audio"

4️⃣ ASIO4ALL Configuration (If No Dedicated Interface)¶

Download: ASIO4ALL Official

⚠️ Note: ASIO4ALL is a last resort. Use a proper audio interface (RME, Focusrite, etc.) if possible.

Installation¶

Download ASIO4ALL v2.15+ installer
Run installer → Install for all users
Reboot if prompted

Configuration¶

Open ASIO4ALL Control Panel: - From your DAW: Options → Audio Settings → Click ASIO Control Panel button - Or from Start Menu: ASIO4ALL v2 Instruction Manual → Control Panel

Settings:

🎚️ Buffer Size: 64-256 samples
   ├─ 64 samples  = 1.33ms @ 48kHz (aggressive, may glitch)
   ├─ 128 samples = 2.67ms @ 48kHz (recommended start)
   └─ 256 samples = 5.33ms @ 48kHz (safe, higher latency)

🔧 Hardware Buffer:
   └─ Match ASIO buffer size (e.g., if ASIO = 128, set HW = 128)

⚙️ Advanced Options (click wrench icon):
   ☑️ Allow pull mode (WaveRT)
   ☐ Always Resample (keep OFF for best quality)
   ☑️ Force WDM driver (if needed for problematic devices)

Device Selection: 1. Expand your audio interface in device tree 2. Enable (click power button icon):

☑️ Input channels (e.g., Line In 1/2)
☑️ Output channels (e.g., Speakers 1/2)

3. Disable all other devices to avoid conflicts

Visual Guide (Text-based diagram):

┌─────────────────────────────────────────────────┐
│ ASIO4ALL v2.15 Control Panel                   │
├─────────────────────────────────────────────────┤
│ Buffer Size: [====|--------] 128 samples       │
│ Latency: ⬅ 2.67ms (input) + 2.67ms (output) ➡  │
├─────────────────────────────────────────────────┤
│ 🔊 Realtek High Definition Audio               │
│    ├─ 🔇 Microphone (Disabled)                  │
│    ├─ ✅ Line In 1/2 (Enabled)                  │
│    └─ ✅ Speakers 1/2 (Enabled)                 │
│                                                 │
│ 🎧 USB Audio Device (Disabled)                 │
│                                                 │
│ [Advanced Options ⚙️] [OK] [Cancel]            │
└─────────────────────────────────────────────────┘

✅ Windows Configuration Checklist¶

After completing all steps, verify:

Verification Test:

# Run LatencyMon for 5 minutes while DAW is active
# Download: https://www.resplendence.com/latencymon

Expected Results: - ✅ "Your system appears suitable for real-time audio" - ✅ Highest DPC latency < 500 μs - ✅ Highest ISR latency < 100 μs

🍎 macOS Ventura/Sonoma Setup¶

macOS has excellent audio support out-of-box (CoreAudio), but still needs configuration for low-latency development.

1️⃣ CoreAudio Optimization¶

A. Sample Rate Lock¶

Why? Prevents macOS from switching sample rates, which causes glitches.

Using Audio MIDI Setup: 1. Open Applications → Utilities → Audio MIDI Setup.app 2. Select your audio interface in left sidebar 3. Set Format:

Sample Rate: 48000 Hz (locked)
Bit Depth: 24-bit Integer (or 32-bit Float if supported)

4. Right-click device → Use This Device for Sound Output (optional)

Lock Sample Rate (Prevent Auto-Switching): - macOS will honor this setting for all apps - Some interfaces (RME, UAD) have hardware sample rate lock in driver settings

B. I/O Buffer Size Configuration¶

Why? Lower buffer = lower latency, but more CPU usage.

Method 1: Per-Application (DAW-specific) - Most DAWs (Logic Pro, Ableton, Reaper) have buffer size in preferences - Typical: 64-256 samples @ 48kHz

Method 2: System-Wide (via Terminal)

# Check current buffer size
system_profiler SPAudioDataType | grep "Buffer Frame Size"

# Set buffer size (requires app restart)
# Note: Not all devices respect this setting
defaults write com.apple.coreaudio HALBufferSize -int 128

Verification:

# Check if setting was applied
defaults read com.apple.coreaudio

C. Aggregate Device Creation¶

Why? Combine multiple interfaces into one virtual device (e.g., interface inputs + BlackHole outputs).

Steps: 1. Open Audio MIDI Setup.app 2. Click + (bottom-left) → Create Aggregate Device 3. Name the device: AudioLab Aggregate 4. Check boxes for devices to combine:

☑️ Focusrite Scarlett 4i4 (or your interface)
☑️ BlackHole 2ch (for loopback routing)

5. Set Clock Source: Select your primary interface (e.g., Scarlett) 6. Sample Rate: 48000 Hz 7. Click Use in your DAW's audio settings

Example Use Case: - Inputs: Scarlett mic/line inputs - Outputs: Scarlett main outs + BlackHole (for analysis tools)

2️⃣ Permissions (macOS Privacy)¶

Why? macOS Sonoma+ requires explicit permission for audio I/O and accessibility features.

Grant Microphone Access¶

System Settings → Privacy & Security → Microphone

Enable for:

☑️ Your DAW (Logic Pro, Ableton, Reaper, etc.)
☑️ Terminal (if running CLI audio tools)
☑️ Your audio plugin host (if standalone testing)

Grant Input Monitoring Access¶

Required for: Screen recording plugins, visualizer tools

System Settings → Privacy & Security → Input Monitoring

Enable for:

☑️ Terminal
☑️ Xcode (if debugging)
☑️ Any app that monitors audio/MIDI

Grant Accessibility Access (If Needed)¶

Required for: Automation scripts, keystroke simulators

System Settings → Privacy & Security → Accessibility
Click + → Add your app → Enable

3️⃣ Terminal & Development Tools¶

A. Install Homebrew¶

What? Package manager for macOS (like apt/yum on Linux).

# Install Homebrew
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

# Add to PATH (for Apple Silicon Macs)
echo 'eval "$(/opt/homebrew/bin/brew shellenv)"' >> ~/.zprofile
eval "$(/opt/homebrew/bin/brew shellenv)"

# Verify
brew --version

B. Xcode Command Line Tools¶

Required for: Compiling C++ code, git, make, clang.

# Install Xcode CLT
xcode-select --install

# Verify
xcode-select -p
# Expected: /Applications/Xcode.app/Contents/Developer
# or: /Library/Developer/CommandLineTools

# Check clang version
clang --version
# Expected: Apple clang version 15.0.0+ (for Sonoma)

Full Xcode (Optional, 40+ GB): - Download from Mac App Store if you need Instruments (profiler) - Not required for command-line builds

C. Rosetta 2 (Apple Silicon Only)¶

Why? Some legacy plugins/tools are Intel-only. Rosetta 2 provides translation.

# Install Rosetta 2
softwareupdate --install-rosetta --agree-to-license

# Verify
arch
# Expected: arm64 (native) or i386 (Rosetta mode)

Running Apps in Rosetta: 1. Right-click app in Finder → Get Info 2. Check: ☑️ Open using Rosetta 3. Launch app

✅ macOS Configuration Checklist¶

Sample Rate Locked: 48000 Hz in Audio MIDI Setup
Buffer Size: Set in DAW preferences (128-256 samples)
Aggregate Device: Created if using multiple interfaces
Microphone Permission: Granted to DAW/tools
Input Monitoring: Enabled for necessary apps
Homebrew: Installed and in PATH
Xcode Command Line Tools: Installed
Rosetta 2: Installed (Apple Silicon only)
BlackHole: Installed for virtual routing (if needed)

Verification Test:

# Check audio latency
# Open your DAW, set buffer to 64 samples @ 48kHz
# Expected round-trip latency: 3-6ms with proper interface

# Monitor system performance
sudo fs_usage | grep audio
# Should show minimal disk I/O during audio processing

🐧 Linux (Ubuntu/Arch) Configuration¶

Linux offers best real-time performance but requires more manual configuration.

1️⃣ Real-Time Kernel (Highly Recommended)¶

Why? Real-time kernel preemption = lower latency jitter (±50μs vs ±5000μs on generic kernel).

Ubuntu/Debian¶

# Install low-latency kernel
sudo apt update
sudo apt install linux-lowlatency

# Verify kernel after reboot
uname -r
# Expected: 5.15.0-XX-lowlatency or 6.x.x-lowlatency

Reboot:

sudo reboot

Select Kernel on Boot (if not auto-selected): 1. Hold Shift during boot → GRUB menu 2. Select Advanced options for Ubuntu 3. Choose Ubuntu, with Linux 5.15.0-XX-lowlatency

Arch Linux¶

# Install real-time kernel
yay -S linux-rt linux-rt-headers

# Update bootloader
sudo grub-mkconfig -o /boot/grub/grub.cfg

# Reboot and select linux-rt kernel
sudo reboot

# Verify
uname -r
# Expected: 6.x.x-rt

2️⃣ JACK/PipeWire Setup¶

JACK = Traditional low-latency audio server PipeWire = Modern replacement (backwards-compatible with JACK)

Option A: PipeWire (Recommended for Modern Distros)¶

Ubuntu 22.04+:

# Install PipeWire and JACK compatibility
sudo apt install pipewire pipewire-jack wireplumber

# Enable PipeWire services
systemctl --user --now enable pipewire pipewire-pulse wireplumber

# Verify
pw-cli info all | grep "node.name"
# Should list audio devices

Configure Buffer Size:

# Edit PipeWire config
mkdir -p ~/.config/pipewire
cp /usr/share/pipewire/pipewire.conf ~/.config/pipewire/

# Edit ~/.config/pipewire/pipewire.conf
# Find default.clock.rate and set to 48000
# Find default.clock.quantum and set to 128 (or 64 for lower latency)

nano ~/.config/pipewire/pipewire.conf

Add these lines:

context.properties = {
    default.clock.rate          = 48000
    default.clock.quantum       = 128
    default.clock.min-quantum   = 64
    default.clock.max-quantum   = 2048
}

Restart PipeWire:

systemctl --user restart pipewire pipewire-pulse wireplumber

Option B: JACK (Traditional)¶

Install JACK2:

# Ubuntu/Debian
sudo apt install jackd2 qjackctl

# Arch Linux
sudo pacman -S jack2 qjackctl

Configure JACK (via GUI):

# Launch QjackCtl
qjackctl &

Settings (click Setup button):

Sample Rate: 48000 Hz
Frames/Period: 128 (or 64 for lower latency)
Periods/Buffer: 2
Driver: alsa (Linux native)
Interface: hw:0 (or your interface name)

Advanced:
☑️ Realtime
Priority: 70

Start JACK: - Click Start button in QjackCtl - Status should show "Started"

3️⃣ Real-Time Privileges¶

Why? Without RT privileges, JACK/PipeWire cannot achieve low latency.

A. Create `realtime` Group¶

# Check if realtime group exists
getent group realtime

# If not, create it
sudo groupadd realtime

B. Add User to Audio Groups¶

# Add your user to audio and realtime groups
sudo usermod -aG audio $USER
sudo usermod -aG realtime $USER

# Verify
groups $USER
# Expected: ... audio ... realtime ...

C. Configure RT Limits¶

Edit /etc/security/limits.conf:

sudo nano /etc/security/limits.conf

Add these lines (at the end):

@audio   -  rtprio     95
@audio   -  memlock    unlimited
@audio   -  nice      -19

@realtime   -  rtprio     95
@realtime   -  memlock    unlimited
@realtime   -  nice      -19

Explanation: - rtprio 95: Real-time priority (max is 99, reserve 99 for kernel) - memlock unlimited: Allow locking memory (prevents swapping) - nice -19: Highest CPU scheduling priority for non-RT processes

Reboot Required:

sudo reboot

Verify After Reboot:

ulimit -r  # Should show 95 (rtprio)
ulimit -l  # Should show unlimited (memlock)

4️⃣ Disable CPU Frequency Scaling (Optional but Recommended)¶

Why? CPU governor switching introduces latency spikes.

Check Current Governor:

cat /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor

Set to Performance Mode:

# Temporarily (until reboot)
echo performance | sudo tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor

# Permanently (install cpufrequtils)
sudo apt install cpufrequtils

# Edit /etc/default/cpufrequtils
sudo nano /etc/default/cpufrequtils

Add:

GOVERNOR="performance"

Apply:

sudo systemctl restart cpufrequtils

5️⃣ ALSA Configuration (Advanced)¶

Tweak ALSA for Lower Latency:

Edit ~/.asoundrc (create if doesn't exist):

nano ~/.asoundrc

Add:

pcm.!default {
    type plug
    slave.pcm hw
}

ctl.!default {
    type hw
    card 0
}

Explanation: - Forces ALSA to use hardware device directly (bypasses dmix/dsnoop) - Reduces latency by ~5-10ms

✅ Linux Configuration Checklist¶

Verification Test:

# Test JACK latency
jack_iodelay

# Expected output:
# 2.67 ms total roundtrip latency (at 128 frames/period)
# extra loopback latency: XX frames
# use XX for the backend arguments -I and -O

# Monitor system performance
sudo rtcqs  # Install via: pip install rtcqs
# Should show all checks passed

🎯 Cross-Platform Verification¶

After configuring your OS, run these tests:

Test 1: Loopback Latency Test¶

Setup: Connect audio output → input (physical loopback cable or virtual loopback)

Windows:

# Use LatencyMon (GUI)
# Download: https://www.resplendence.com/latencymon

macOS:

# Use BlackHole + AU Lab (built-in)
open /System/Applications/Utilities/Audio\ MIDI\ Setup.app
# Create aggregate device with loopback

Linux:

# Use JACK's built-in test
jack_iodelay

Expected Results: | Buffer Size | Sample Rate | Expected Latency (Round-Trip) | |-------------|-------------|-------------------------------| | 64 samples | 48 kHz | 2.67 ms | | 128 samples | 48 kHz | 5.33 ms | | 256 samples | 48 kHz | 10.67 ms |

Test 2: CPU Load Test¶

All Platforms: 1. Open your DAW 2. Load 20+ instances of a reverb plugin 3. Play audio for 5 minutes 4. Monitor CPU usage

Expected: - ✅ CPU < 70% average (leaves headroom) - ✅ Zero audio dropouts/glitches - ✅ Stable latency (no jitter)

📸 Visual Configuration References¶

Windows Sound Properties - Advanced Tab¶

┌──────────────────────────────────────────┐
│ Sound Properties - Advanced             │
├──────────────────────────────────────────┤
│ Default Format:                          │
│ [24 bit, 48000 Hz (Studio Quality)  ▼]  │
│                                          │
│ Exclusive Mode:                          │
│ ☑️ Allow applications to take exclusive  │
│    control of this device                │
│ ☑️ Give exclusive mode applications      │
│    priority                              │
│                                          │
│ [Apply] [OK] [Cancel]                    │
└──────────────────────────────────────────┘

macOS Audio MIDI Setup¶

┌────────────────────────────────────────────┐
│ Audio MIDI Setup                           │
├────────────────────────────────────────────┤
│ Devices:                                   │
│ ├─ 🎵 Built-in Output                      │
│ ├─ 🎧 Focusrite Scarlett 4i4         ← Selected
│ └─ 🕳️  BlackHole 2ch                       │
│                                            │
│ Format:                                    │
│ ├─ Sample Rate: [48000 Hz ▼]              │
│ └─ Bit Depth: [24-bit Integer ▼]          │
│                                            │
│ Input Channels: 4                          │
│ Output Channels: 4                         │
└────────────────────────────────────────────┘

Linux QjackCtl Setup¶

┌─────────────────────────────────────────┐
│ JACK Audio Connection Kit - Setup      │
├─────────────────────────────────────────┤
│ Driver: alsa                            │
│ Interface: hw:0                         │
│ Sample Rate: 48000 Hz                   │
│ Frames/Period: 128                      │
│ Periods/Buffer: 2                       │
│                                         │
│ ☑️ Realtime    Priority: 70             │
│ ☐ Verbose                               │
│                                         │
│ [OK] [Cancel]                           │
└─────────────────────────────────────────┘

⚠️ What NOT to Touch¶

Windows¶

❌ Do NOT enable "Loudness Equalization" (destroys dynamics)
❌ Do NOT use "Balanced" or "Power Saver" power plans
❌ Do NOT enable "Fast Startup" (causes driver issues after reboot)
❌ Do NOT enable "Core Isolation" / "Memory Integrity" (adds latency)

macOS¶

❌ Do NOT enable "Use ambient noise reduction" in Sound settings
❌ Do NOT force quit coreaudiod (it auto-restarts but causes glitches)
❌ Do NOT run heavy background tasks (Time Machine, Spotlight indexing) during audio work

Linux¶

❌ Do NOT use PulseAudio for real-time work (use PipeWire or JACK)
❌ Do NOT set rtprio to 99 (kernel needs priority 99)
❌ Do NOT disable swap entirely (memlock unlimited handles this)

📝 Notes¶

Last Updated: October 2024 Tested On: - Windows 11 23H2 (Build 22631) - macOS Sonoma 14.5 - Ubuntu 22.04.3 LTS + 6.5.0-lowlatency kernel - Arch Linux (kernel 6.6.10-rt)

Recommended Reading: - Windows MMCSS Documentation - CoreAudio Programming Guide - Linux Audio Wiki

Next: IDE Setup - Configure Visual Studio, VSCode, and CLion for AudioLab development

⚙️ OS Configuration - Real-Time Audio Optimization¶

🪟 Windows 10/11 Optimizations¶

1️⃣ Audio Exclusivity & Enhancements¶

Step-by-Step Configuration¶

Verification¶

2️⃣ Power Settings (Critical for Real-Time)¶

Power Plan Configuration¶

Verify Power Plan¶

3️⃣ Real-Time System Tweaks¶

A. Disable Windows Search Indexing (Project Folders)¶

B. Exclude Build Folders from Windows Defender¶

C. Registry Tweaks for MMCSS (Multimedia Class Scheduler Service)¶

4️⃣ ASIO4ALL Configuration (If No Dedicated Interface)¶

Installation¶

Configuration¶

✅ Windows Configuration Checklist¶

🍎 macOS Ventura/Sonoma Setup¶

1️⃣ CoreAudio Optimization¶

A. Sample Rate Lock¶

B. I/O Buffer Size Configuration¶

C. Aggregate Device Creation¶

2️⃣ Permissions (macOS Privacy)¶

Grant Microphone Access¶

Grant Input Monitoring Access¶

Grant Accessibility Access (If Needed)¶

3️⃣ Terminal & Development Tools¶

A. Install Homebrew¶

B. Xcode Command Line Tools¶

C. Rosetta 2 (Apple Silicon Only)¶

✅ macOS Configuration Checklist¶

🐧 Linux (Ubuntu/Arch) Configuration¶

1️⃣ Real-Time Kernel (Highly Recommended)¶

Ubuntu/Debian¶

Arch Linux¶

2️⃣ JACK/PipeWire Setup¶

Option A: PipeWire (Recommended for Modern Distros)¶

Option B: JACK (Traditional)¶

3️⃣ Real-Time Privileges¶

A. Create realtime Group¶

B. Add User to Audio Groups¶

C. Configure RT Limits¶

4️⃣ Disable CPU Frequency Scaling (Optional but Recommended)¶

5️⃣ ALSA Configuration (Advanced)¶

✅ Linux Configuration Checklist¶

🎯 Cross-Platform Verification¶

Test 1: Loopback Latency Test¶

Test 2: CPU Load Test¶

📸 Visual Configuration References¶

Windows Sound Properties - Advanced Tab¶

macOS Audio MIDI Setup¶

Linux QjackCtl Setup¶

⚠️ What NOT to Touch¶

Windows¶

macOS¶

Linux¶

📝 Notes¶

A. Create `realtime` Group¶