Audio in the latest nightly of Frodo
#31
(2013-01-05, 22:46)squarooticus Wrote:
(2013-01-05, 22:44)mudwater Wrote:
(2013-01-05, 22:39)uNiversal Wrote: I upgrade XBMC everyother day, compiling from GIT on Linux

Works fine here.

uNi

What is your Passthrough audio device set to?
b
FWIW, this made no difference in my case. No matter what the passthrough device was set to (though it should have been set to "HDA NVidia, HDMI 0"), I got no passthrough audio. Reinstall, and bam! Everything works with the defaults, which are now "ALSA:default" for PCM audio and "ALSA:hdmi_formatted" for passthrough.

And you have Dolby and dts capable reciever checked?
Reply
#32
Glad it's working.....but keep in mind the "drop of honey" will get better support than the rant posted earlier....
System: XBMC HTPC with HDMI WASAPI & AudioEngine - Denon  AVR-3808CI  - Denon DVD-5900 Universal Player  - Denon DCM-27 CD-Changer
- Sony BDP-S580 Blu-Ray  - X-Box 360  - Android tablet wireless remote - 7.1 Streem/Axiom/Velodyne Surround System
If I have been able to help feel free to add to my reputation +/- below - thanks!
Reply
#33
(2013-01-05, 22:49)mudwater Wrote:
(2013-01-05, 22:46)squarooticus Wrote:
(2013-01-05, 22:44)mudwater Wrote: What is your Passthrough audio device set to?
b
FWIW, this made no difference in my case. No matter what the passthrough device was set to (though it should have been set to "HDA NVidia, HDMI 0"), I got no passthrough audio. Reinstall, and bam! Everything works with the defaults, which are now "ALSA:default" for PCM audio and "ALSA:hdmi_formatted" for passthrough.

And you have Dolby and dts capable reciever checked?

Yes. Those were already checked by default. (I had to uncheck TrueHD and DTS-HD because the ION cannot output those over HDMI... the receiver interprets it as PCM and plays very loud static.) I also switched the speaker configuration to 5.1, but this occurred in the middle of my testing: it worked fine both before and after doing that. I don't think I've ever really understood what the speaker configuration does considering an AC3/DTS stream is completely opaque to the PC and has meaning only to the receiver.

Edit: and just to be clear: I looked at the lights on the receiver. It was definitely receiving a 7.1 DTS stream.


(2013-01-05, 22:58)DDDamian Wrote: Glad it's working.....but keep in mind the "drop of honey" will get better support than the rant posted earlier....
After being up until 2:30 am working on this and getting nowhere, I was a bit frustrated. I'm still frustrated that the only thing that worked was "reinstall". I think what I need to do is install a file system with a snapshotting facility so I can back up to a known-good state when something like this happens next time.
Reply
#34
Reinstall did nothing for me
Reply
#35
When starting from terminal get this message

Running DIL (3.22.0) Version
DtsDeviceOpen: Opening HW in mode 0
DtsDeviceOpen: Create File Failed

Cound this be the problem?
Reply
#36
You are cross posting, but see my answer in the other thread Smile

http://forum.xbmc.org/showthread.php?tid...pid1287658
If I have helped you or increased your knowledge, click the 'thumbs up' button to give thanks :) (People with less than 20 posts won't see the "thumbs up" button.)
Reply
#37
So, after a day of playing with various things, here's the big item on my wishlist for the audio engine: no matter what the reason for audio failing (blocking on an ALSA device, failure code from Pulse, etc.), it feels like XBMC should wait a few seconds before declaring failure; and then, if still failing, ask the user what to do. Here's a use case that currently gives it problems:

On my desktop machine, I am frequently running multiple audio applications at once: XBMC with a video, Chrome with Flash, pidgin with audio notifications, PulseAudio in general, etc. Currently, XBMC opens the PCM device and keeps it open until it has to switch to passthrough mode. This makes the audio device inaccessible to other applications. A logical solution to this problem would be to switch XBMC to talk to PulseAudio for PCM instead, but the problem here is that when XBMC switches to passthrough, it closes its connection to PulseAudio but PA doesn't release the device for another few seconds. XBMC fails to open the passthrough device, and bam: no audio.

So I guess what I'm asking for is better handling of these kinds of transient failure modes. Thoughts?
Reply
#38
(2013-01-06, 22:59)squarooticus Wrote: So, after a day of playing with various things, here's the big item on my wishlist for the audio engine: no matter what the reason for audio failing (blocking on an ALSA device, failure code from Pulse, etc.), it feels like XBMC should wait a few seconds before declaring failure; and then, if still failing, ask the user what to do. Here's a use case that currently gives it problems:

On my desktop machine, I am frequently running multiple audio applications at once: XBMC with a video, Chrome with Flash, pidgin with audio notifications, PulseAudio in general, etc. Currently, XBMC opens the PCM device and keeps it open until it has to switch to passthrough mode. This makes the audio device inaccessible to other applications. A logical solution to this problem would be to switch XBMC to talk to PulseAudio for PCM instead, but the problem here is that when XBMC switches to passthrough, it closes its connection to PulseAudio but PA doesn't release the device for another few seconds. XBMC fails to open the passthrough device, and bam: no audio.

So I guess what I'm asking for is better handling of these kinds of transient failure modes. Thoughts?

The devs aren't stupid, the audio device hogging issue is quite well known: http://forum.xbmc.org/showthread.php?tid=150963

As with most audio problems on this forum, I blame Ubuntu. It is probably the biggest reason why Linux audio has gotten the voodoo badge. It's braindead easy to get perfectly working audio over HDMI when using other OSes (I'm not talking Windows here) that don't screw you over with bullshit like Pulseaudio (although I see why you would want it in your situation, using XBMC on a desktop as a standard media player (which is the real WTF)), weird ALSA pseudo-devices and other general breakage.

It is probably a fact that users of other distros (Debian, Gentoo, Arch) are more competent when it comes to fixing Linux issues, but still, there is a clear "lack" of threads from users of these distros having audio-related problems. Go figure.
Reply
#39
@negge Rofl at that comment for different reasons Wink

The only reason why some Linux users are more competent than others its not because of Distro they use, its because some, will actually dowork() others will expect dowork() to be done for them.

It has nothing to do with brains, its more laziness.

uNi
Reply
#40
I agree to some degree although by original point was that problem-solving on Ubuntu is of a completely different caliber than problem-solving on any other distribution because Ubuntu manages to break/screw up so many things.
Reply
#41
(2013-01-07, 00:49)negge Wrote: The devs aren't stupid, the audio device hogging issue is quite well known: http://forum.xbmc.org/showthread.php?tid=150963
It may be quite well known (though I find it amusing that you pointed me at a thread from earlier today), but it isn't quite well prioritized. I'm not blaming anyone here, FWIW: XBMC is not intended to be a general media player and not intended for desktop use (though I use it on the desktop for video because its video player is far better than any standalone player I've used). Its primary use seems to be as a set-top box, which it does admirably: certainly better than any other open source package I could name. That's precisely why I classified this as a "wishlist" item.
Quote:As with most audio problems on this forum, I blame Ubuntu.
Ironically, Ubuntu is the first distribution that got audio right for me out-of-the-box. Now this was way back in the pre-PulseAudio days, but it was the first time I didn't have to spend hours tweaking stuff to get audio to work after an install. Even now, with the various problems I've had with PA, Ubuntu doesn't exactly make me nostalgic for those mid-1990's days of trying to figure out how to get OSS to work with my latest sound card. What a nightmare. A lot of it was probably simply that ALSA got a lot more mature in the mid-2000's, so I'm not sure Ubuntu really had all that much to do with it. But it was a nice correlation, at least.
Quote:It is probably the biggest reason why Linux audio has gotten the voodoo badge.
...
It is probably a fact that users of other distros (Debian, Gentoo, Arch) are more competent when it comes to fixing Linux issues, but still, here is a clear "lack" of threads from users of these distros having audio-related problems. Go figure.
I suspect that is the result of selection bias. Everyone I know who uses Linux on the desktop these days uses Ubuntu. I don't know of a single person who uses Gentoo or Debian or Arch or SuSE or Fedora or whatever on the desktop anymore.

And frankly, if a better audio solution made it to one of those other distros, it would have ended up in Ubuntu eventually. The fact is that audio on Linux, as far as it's come in the last 20 years, is still a mess. PulseAudio is a valiant, if perhaps less-than-ideal, attempt to fix that problem.
Reply
#42
FYI

2032 (GH issue)
2074 (GH issue)
http://forum.xbmc.org/showthread.php?tid=147938
http://forum.xbmc.org/showthread.php?tid=150963
http://forum.xbmc.org/showthread.php?tid=148428
http://forum.xbmc.org/showthread.php?tid=136140
http://forum.xbmc.org/showthread.php?tid=150791
http://forum.xbmc.org/showthread.php?pid=1266056

I've openend ticket for listing dmix devices which would be a workaround for these issues.
http://trac.xbmc.org/ticket/13966

Let's hope someone finds the time to fix it. I'm still searching for it, time that is.
Reply
#43
@Falcon1, you do that again and you will get a 10 day ban.
Reply
#44
@davilla,

Pardon me. I was not trying to bother anybody. I was only trying to unify the issue a bit.

Obviously people perceive it otherwise. Sorry about that.

Kind regards,
Reply
#45
Seems like AE or XBMC ignore the alsa config it some point, ignoring the default pcm and going (in my case) for the "front" pcm that dont have a dmix asociated.
Usually default "default" pcm have a dmix included at some point of the initialization (or the black magic/rune throwing/rain dance alsa does underground :-P), so opening other pcm's means losing the share device thingy.
Add this to /etc/asound.conf:
pcm.FixXBMCAE{
type plug
slave.pcm dmix
hint{
description "Create an alias of dmix that appear on XBMC audio setting, for not hogging the sound device ...."
device_output 0
device_input 0
}
}

It will create one more entry to pick up on settings/system/audio device/card etc. The name will be similar to the other entries, but this one will let you use sound in other apps
Reply

Logout Mark Read Team Forum Stats Members Help
Audio in the latest nightly of Frodo0