can you provide log information in chrome://media-internals?

Something like this?:
----

Player Properties

Property	Value
audio_codec_name	opus
audio_dds	false
audio_decoder	OpusAudioDecoder
duration	862.821
event	PLAY
found_audio_stream	true
found_video_stream	true
pipeline_state	kPlaying
player_id	36
render_id	38
url	blob:https://www.youtube.com/bee73635-9022-4915-a61b-6c7d76b7d51a
video_codec_name	vp9
video_dds	false
video_decoder	VpxVideoDecoder
Log 
property filter

Timestamp	Property	Value
00:00:00 00	pipeline_state	kCreated
00:00:00 00	event	WEBMEDIAPLAYER_CREATED
00:00:00 00	url	blob:https://www.youtube.com/bee73635-9022-4915-a61b-6c7d76b7d51a
00:00:00 00	pipeline_state	kInitDemuxer
00:00:00 12	duration	863
00:00:00 13	duration	862.821
00:00:00 16	found_video_stream	true
00:00:00 16	video_codec_name	vp9
00:00:00 17	found_audio_stream	true
00:00:00 17	audio_codec_name	opus
00:00:00 17	pipeline_state	kInitRenderer
00:00:00 24	audio_dds	false
00:00:00 24	audio_decoder	OpusAudioDecoder
00:00:00 25	video_dds	false
00:00:00 25	video_decoder	VpxVideoDecoder
00:00:00 25	pipeline_state	kPlaying
00:00:00 44	event	PLAY
00:00:03 753	event	PAUSE
00:00:18 770	pipeline_state	kSuspending
00:00:18 770	pipeline_state	kSuspended
00:02:10 583	event	PLAY
00:02:10 583	pipeline_state	kResuming
00:02:10 585	audio_dds	false
00:02:10 585	audio_decoder	OpusAudioDecoder
00:02:10 585	video_dds	false
00:02:10 585	video_decoder	VpxVideoDecoder
00:02:10 585	pipeline_state	kPlaying

Controller 29:2 Properties

Property	Value
channel_layout	STEREO
channels	2
device_id	default
device_type	pcm_low_latency
effects	NO_EFFECTS
frames_per_buffer	960
render_process_id	38
sample_rate	48000
status	started
web_contents_title	DIT IS ALLEMAAL ZO NIEUW! - H1Z1 King of the Kill - YouTube

-----

Stream 1:2 Properties 

Property	Value
channel_layout	STEREO
channels	2
device_id	{0.0.0.00000000}.{89e68197-17cd-4534-accc-3f77954bb334}
device_type	fake
effects	NO_EFFECTS
frames_per_buffer	960
sample_rate	48000
status	started
volume	1

Thank you for providing more feedback. Adding requester "yiningc@chromium.org" for another review and adding "Needs-Review" label for tracking.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Dale, I don't have hardware to repro this bug. can you diagnose by looking at log at #2, #3?

+audio folk. Do you have one of these in STO?

No, we don't have that one here.

Have a speculative CL to fix this, https://codereview.chromium.org/2683033003 but am waiting on testing from folk with discrete setups. I don't have one at this time. If anyone has a Chrome build environment and one of these cards patching that CL and seeing if it fixes your problem would be helpful!

Whoops, sorry, c#8 is incorrect, it won't fix this issue, I meant to put that comment on  issue 675787 . 

rbroek@ has this started working in the meantime? There are a few OSX updates that may have been related.

OSX? As in the Mac OS? I'm on win10.
But no, doesn't work yet.

If I can try anything else I'm all for it btw.
Although I'm probably not allowed on your test-env. :P

Heh, sorry, I'm all sorts of confused on this bug. I saw OS version 10 and was still confusing this with the other bugs. I don't have any fixes here unfortunately.

What you might try is installing Sawbuck or enabling logging from the command line, https://www.chromium.org/for-testers/enable-logging

and seeing if we spit out any more useful error messages. We've left a lot of them in the WASAPI path to find corner cases.

Hmm, yeah there are some, but no idea what they mean.
(See attach.)

Deleted: chrome_debug.log 194 KB

chrome_debug.log 194 KB View Download

Screenshot of what the cmd prompt thingie shows when browsing over YT.

Deleted: error.png 13.9 KB

	error.png 13.9 KB View Download

That's really helpful, for some reason the channel count is reported as zero. I'd guess we don't understand the channel layout. If you click on your speaker icon in the tray and open properties and then configure speakers, does it show a layout?

Possibly we need to add support for arbitrary layouts instead of restricting to the set here:

https://cs.chromium.org/chromium/src/media/audio/win/core_audio_util_win.cc?l=38

Hmm, thought the issue might be we don't guess layouts based on channel counts or that this device had a crazy channel count, but neither is true and it looks like it might just be a bug in how we report layouts:

https://codereview.chromium.org/2687873006

Hehe, no it does not.
Thing is, this soundcard is for DJ'ing; it has 2 channels, one for main output, one for headphone. Yet Chrome is the only thing that doesn't work on it tho ;)

Deleted: speakerSetup.png 17.7 KB

	speakerSetup.png 17.7 KB View Download

Heh, well hopefully my patch will help out with it. It should show up on canary by Sunday/Monday; likely 58.0.3010.0 or higher.

Cool, will check for sure :)

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/f062c42d082ebf9c2f2e39ac6683f7755d232ac2

commit f062c42d082ebf9c2f2e39ac6683f7755d232ac2
Author: dalecurtis <dalecurtis@chromium.org>
Date: Sat Feb 11 02:09:31 2017

Always return CHANNEL_LAYOUT_UNSUPPORTED for unsupported layouts.

The old code would sometimes return CHANNEL_LAYOUT_NONE which we
don't handle fallback to guessing the layout for. Instead always
return CHANNEL_LAYOUT_UNSUPPORTED so that we will properly attempt
to guess the channel layout based on channel count.

This is a speculative fix for broken audio on Traktor 2 Audio.

BUG= 651172 
TEST=none
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2687873006
Cr-Commit-Position: refs/heads/master@{#449831}

[modify] https://crrev.com/f062c42d082ebf9c2f2e39ac6683f7755d232ac2/media/audio/win/core_audio_util_win.cc

rbroek@ did that patch help?

Omg yes!

*Uninstalls Firefox* 

Thanks so much man!! =D

Great! Will merge back to M57 so it'll roll out with the next stable update in a few weeks, using canary/dev as a daily driver isn't everyone's favorite time :)

No problem till then. I'm a professional software tester anyway, hehe ;)

Your change meets the bar and is auto-approved for M57. Please go ahead and merge the CL to branch 2987 manually. Please contact milestone owner if you have questions.
Owners: amineer@(clank), cmasso@(bling), ketakid@(cros), govind@(desktop)

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

If possible, could you please merge your CL into M57 branch 2987 before 5 PM PT today, Tuesday (02/14/17). Thank you.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/d37b8a11a391cc19ff7f9caac521fc1408d73218

commit d37b8a11a391cc19ff7f9caac521fc1408d73218
Author: Dale Curtis <dalecurtis@chromium.org>
Date: Tue Feb 14 19:49:56 2017

Merge M57: "Always return CHANNEL_LAYOUT_UNSUPPORTED for unsupported layouts."

The old code would sometimes return CHANNEL_LAYOUT_NONE which we
don't handle fallback to guessing the layout for. Instead always
return CHANNEL_LAYOUT_UNSUPPORTED so that we will properly attempt
to guess the channel layout based on channel count.

This is a speculative fix for broken audio on Traktor 2 Audio.

BUG= 651172 
TEST=none
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2687873006
Cr-Commit-Position: refs/heads/master@{#449831}
(cherry picked from commit f062c42d082ebf9c2f2e39ac6683f7755d232ac2)

Review-Url: https://codereview.chromium.org/2699463002 .
Cr-Commit-Position: refs/branch-heads/2987@{#505}
Cr-Branched-From: ad51088c0e8776e8dcd963dbe752c4035ba6dab6-refs/heads/master@{#444943}

[modify] https://crrev.com/d37b8a11a391cc19ff7f9caac521fc1408d73218/media/audio/win/core_audio_util_win.cc