Section 4: Audio Endpoints

The Big Picture: Where Does the Sound Go?

An audio endpoint is "the thing sound goes to, or comes from". It is how Windows names the end of the audio path. Examples:

One physical gadget can show up as more than one endpoint. For example, a single laptop may list "Speakers", "Headphones" and a "Microphone array". Each endpoint has its own volume, mute button and format.

Endpoints are built by the AudioEndpointBuilder service from the information the drivers give (Section 2 and Section 3). Programs and the Sound settings only see endpoints, never the chip behind them.

4.1 Playback Devices

What it is: The list of endpoints that output sound.

Where to see them. There are two places, and both show the same devices:

In the classic window, the Playback tab lists each device with a green tick on the current default. A moving green bar beside a device means sound is flowing to it right now.

Properties of a playback endpoint (right-click a device, choose Properties):

• General: the driver name and which jack it uses.
• Levels: master volume and balance.
• Enhancements (or "Effects" on some versions): turn APO effects such as loudness equalisation on and off (Section 3).
• Advanced: the default format, for example "2 channel, 24 bit, 48000 Hz (Studio Quality)", and the exclusive-mode check boxes (Section 3.3).
• Spatial sound: options such as "Windows Sonic for Headphones" or other formats. Which choices appear depends on your Windows version and drivers.

Disabled and disconnected devices. Right-click an empty part of the list and tick Show Disabled Devices and Show Disconnected Devices. A device you disabled for a test will show up here, greyed out.

4.2 Recording Devices

What it is: The list of endpoints that input sound: microphones and line-in. In mmsys.cpl open the Recording tab. Speak into your microphone and the green bar beside it should jump.

Properties of a recording endpoint:

Privacy. Windows also has microphone privacy switches (Settings → Privacy & security → Microphone) that decide which apps may record. The ADC in the codec (Section 1) converts the signal, but Windows decides who may see the samples.

4.3 Default Device Selection

What it is: If you have several playback devices (speakers, headphones, HDMI monitor), Windows must pick which one new sounds go to. It uses a default device. There are in fact two defaults for each direction:

On the Playback tab, a device marked "Default Device" has a green tick. A device marked "Default Communications Device" has a green telephone symbol. The two can be the same device or different ones (for example the headset for calls and the speakers for music).

How to change the default:

1. In Settings → System → Sound, pick a different device in the Output box (or choose a device and click Set as default where the page offers it).
2. Or in mmsys.cpl, right-click a device and choose Set as Default Device or Set as Default Communications Device.

What Windows does by itself. When you plug in a new headset or turn on a Bluetooth speaker, Windows often switches the default to it automatically. When you unplug it, the sound moves back. If sound "comes out of the wrong thing" (the TV instead of your speakers, say), the default endpoint is almost always the reason.

Some apps pick their own device. Many apps (and the Volume mixer in newer Windows) let you choose an output per app, overriding the default. If one app is silent while others work, check its own device choice.

Jack detection (Section 1) can also change things: some drivers show a pop-up asking what you just plugged in, and then switch between "Speakers" and "Headphones" for you.

Summary