Fake mic is unreliable
Reported by
ben.brow...@gmail.com,
Jun 16 2017
|
|||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:54.0) Gecko/20100101 Firefox/54.0 Steps to reproduce the problem: 1. Start Chrome with the '--use-fake-device-for-media-stream' flag. 1. Go to https://webrtc.github.io/samples/src/content/getusermedia/volume/ 2. See the fake mic volume indication 3. Replace the url in the same tab with https://webrtc.github.io/samples/src/content/peerconnection/pc1/ 4. Create a call - Click on Start, Call, Hang Up. 5. Replace the url in the same tab with https://webrtc.github.io/samples/src/content/getusermedia/volume/ 6. See that the volume of the fake mic is 0. What is the expected behavior? The fake mic volume should change. What went wrong? Volume is 0. Did this work before? No Does this work in other browsers? N/A Chrome version: 59.0.3071.86 Channel: stable OS Version: 6.1 (Windows 7, Windows Server 2008 R2) Flash Version:
,
Jun 19 2017
,
Jun 19 2017
Able to reproduc the issue using latest stable #59.0.3071.104 on Win 10 , Mac 10.12.4 and Ubuntu 14.04 as well. Note: Observing the same behavior since M45. Hence considering this issue as Non-Regression issue and untriaged the issue so that it get's addressed. Unable to launch the older builds (Ex: M30). Issue is seen in M61 as well. Thanks!!
,
Jun 19 2017
AFAIK --use-fake-device-for-media is there only for testing, though I may be wrong. How are you using it? IIRC, Chrome's browser tests each get a new process to run in, to avoid any state changes in one affecting the other. From what I can see, this is what's happening here: the FakeAudioInputDevice has a BeepSource that is generally set to automatically generate beeps. Once the peerconnection page is loaded, that code instead instructs the BeepSource to manually trigger beeps in sync with the video. Once that's happened, the BeepSource will no longer automatically generate beeps and the volume test page will show zero. I've changed this to Available for now. I'm not sure if making FakeAudioInputSource more generally reliable is something we want to do, though.
,
Jun 26 2017
Given that use-fake-device-for-media is a flag intended for tests only (it avoids the need to grant camera/microphone permissions), I don't think you can have the expectations given in the initial report. The flag does not emulate a microphone but simply allows the user to use WebRTC on machines with no real audio/video hardware. |
|||||
►
Sign in to add a comment |
|||||
Comment 1 by guidou@chromium.org
, Jun 17 2017