Audio Sources and Capture

Audio Sources and Capture

Prism analyzes either a system output or an input device. It never inserts itself into the system's playback path, so changing a scope or display setting cannot change what you hear. Trim changes the level sent into Prism's analyzers and rolling recorder only.

Open Settings and scroll to Audio Source to choose a source and read its status.

Output and input sources

The source menu is divided into two groups:

Group	What Prism analyzes	Typical uses
Output Devices	The mix being sent to a system output	Music, videos, games, streams, mix references
Input Devices	A microphone, interface input, or line input	Live instruments, microphones, hardware returns, room measurements

Default Input follows the operating system's current default input. A named input remains selected until it disappears or you choose another source.

Prism starts capture automatically when the app opens and remembers the selected capture mode and device. Changing the operating-system default output may update the Default entry while a specifically selected output remains selected.

If native system-output capture is unavailable or cannot start, Prism automatically falls back to Default Input and shows an informational notice explaining why. Check the selected source before analyzing or enabling rolling capture: on a computer with a built-in microphone, the fallback can be live room audio rather than system playback. Dismiss clears the notice but keeps the fallback source; reselect an output and use Retry after correcting the original problem.

Reading capture status

Status	Meaning
Connecting	Prism is opening the selected source or changing devices.
Capturing	Audio frames are reaching Prism. Silence can still produce a flat display.
Idle	Capture is stopped. It can be started again from the tray menu.
Error	The source could not be opened or stopped producing valid data. Read the message shown below the source.

When a message appears:

• Retry attempts the same source again.
• Use Default switches away from a non-default source and retries the platform default.
• Dismiss clears an informational notice; it does not change the selected device.

The tray menu also shows the active source and can switch sources or start and stop capture without opening the main window.

Input trim

Trim ranges from -12.0 dB to +12.0 dB in 0.5 dB steps. Double-click the slider to return to 0.0 dB.

• Raise Trim when a quiet input barely moves the scopes.
• Lower Trim when a hot input fills the scopes or clips Prism's meters.
• Trim affects every analyzer and any WAV made by Rolling Audio Capture.
• Trim does not change system volume, the selected output device, or audio heard from other applications.

Positive trim can create digital clipping in the analyzed/captured signal even when the original source was not clipped. Use the meters to keep enough headroom.

Platform capture paths

macOS

System-output capture uses CoreAudio and requires macOS 14.2 or newer. macOS may ask for permission to record system audio. Input-device capture may also ask for microphone access.

If permission was denied, open System Settings > Privacy & Security, allow Prism in the relevant audio or microphone category, then quit and reopen Prism. A custom or locally built app can appear under a different signing identity from an installed release.

Windows

System-output capture uses WASAPI loopback on the selected output device. Prism follows the device's shared mix format; it does not take exclusive control of the device.

If an output disappears after sleep, a driver update, or an interface sample-rate change, choose Use Default and then reselect the device after Windows reports it again.

Linux

System-output capture uses PulseAudio monitor sources. PipeWire systems need the PulseAudio compatibility service. A normal output must expose a corresponding monitor source for Prism to capture it.

If outputs are absent, confirm that PulseAudio/PipeWire is running in the same desktop session as Prism. Sandbox policies in some community packages can also prevent access to the session audio server.

Device changes and saved choices

Prism stores the last system-output ID and input-device ID separately. Switching from output to input does not forget the previous output selection, and switching back restores it when the device still exists.

Device IDs are local to a computer. They are not part of shareable .prsm profile files. Importing a profile changes the rack, not the recipient's audio-device choice.

Capture and performance

Hidden or removed scopes stop requesting their analyzer stream. Popped-out scopes continue receiving the stream they need. The Performance frame target changes how often scopes draw, not the audio capture rate or the audio content being analyzed.

If the display stutters while capture remains healthy, lower the frame target or reduce expensive FFT/clarity settings. See Understanding the Scopes.

Troubleshooting a silent rack

1. Confirm that Audio Source says Capturing.
2. Verify the selected output is the one actually playing audio, or the selected input has signal.
3. Return Trim to 0.0 dB.
4. Try Use Default, then reselect the intended device.
5. Check platform permissions or the Linux session audio service.
6. Keep the exact error text if the problem continues and use Troubleshooting.