1037000 - WebRTC Audio fails to pick the default microphone on Windows

Status: VERIFIED FIXED

(Core :: WebRTC: Audio/Video, defect)

Description

[Mark Banner (:standard8)]

STR:

1) Open up Firefox, and visit gum_test.html
2) See what the default audio device is
3) Change the Window's default recording device
4) Reload gum_test and see what the default audio device is

Expected Results

=> Matches the new setting

Actual Results

=> Always stays the same

Comment 1

[Randell Jesup [:jesup] (needinfo me)]

Attachment: Include "default communications device" in the audio input enumeration for gUM

Comment 2

[Randell Jesup [:jesup] (needinfo me)]

Attachment: Include "default communications device" in the audio input enumeration for gUM

disable on Android/B2G until investigated.  Works on Windows, Linux and Mac.  Note that typically, the default device will show first, and again later in the list.

Comment 3

[Randell Jesup [:jesup] (needinfo me)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/f2e5fb7f242c

We'll want this for 32 if at all possible.  Huge improvement for users (especially on windows) for changing 0 to -1.

Comment 4

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/f2e5fb7f242c

Comment 5

[Randell Jesup [:jesup] (needinfo me)]

dangers of arrow keys...

Comment 6

[Randell Jesup [:jesup] (needinfo me)]

Comment on attachment 8453872 [details] [diff] [review]
Include "default communications device" in the audio input enumeration for gUM

Approval Request Comment
[Feature/regressing bug #]: N/A

[User impact if declined]: the first device found will continue to be the default, requiring the user to change the audio device on every browser restart on the first call, which causes major user confusion.  (On mac, this already somewhat happens due to how the OS exposes things).

[Describe test coverage new/current, TBPL]: On m-c for a month.  Normal getUserMedia test coverage.

[Risks and why]: extremely low risk - effectively a single line change of 0 to -1 (which is a special value which has always been supported by the underlying code).

[String/UUID change made/needed]: none.

Comment 7

[Sylvestre Ledru [:Sylvestre]]

Randell, the patch landed initially in 33. Why do you request the uplift for aurora? Mistake?

Comment 8

[Randell Jesup [:jesup] (needinfo me)]

AH, yes - didn't notice (was going through all unresolved webrtc/screenshare/openh264-uplift whiteboard tags.)  Thanks

Comment 9

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-beta/rev/ee74d30a8968

Comment 10

[Paul Silaghi, QA [:pauly]]

(In reply to Mark Banner (:standard8) from comment #0)
> 1) Open up Firefox, and visit gum_test.html
The test page is N/A and it seems I can't reproduce the problem on http://apprtc.webrtc.org/.
Is there another way to reproduce it?

Comment 11

[Paul Silaghi, QA [:pauly]]

Forget about comment 10.
I can reproduce the problem and it doesn't look fixed in the latest nightly too, if considering the default device as the first one from the share audio device list. Tested on Win 7.

Comment 12

[Mark Banner (:standard8)]

(In reply to Paul Silaghi, QA [:pauly] from comment #11)
> Forget about comment 10.
> I can reproduce the problem and it doesn't look fixed in the latest nightly
> too, if considering the default device as the first one from the share audio
> device list. Tested on Win 7.

This works on my windows machine using win 7. You set the default device via Control Panel -> Sound -> Recording.

Using 2014-08-10 The default device correctly gets listed as the first item in the drop-down menu. I can switch between internal microphone and my headset, and it correctly picks up when I restart capturing the audio.

Comment 13

[Paul Silaghi, QA [:pauly]]

Thanks Mark.
I was using the right click menu/Set as Default Device. I see it only works if using the second option - "Set as Default communication device", not sure what's the difference between them.
32b5, 33.0a2 (2014-08-11), 34.0a1 (2014-08-10), Win 7 x64