Even With AMDGPU DC, HDMI/DP Audio Isn't Working Out For All Radeon Linux Users

[partizann]

Do not waste poor APUs or any high performance computers cpu cycles to hdmi audio. Hdmi audio code is large and complex, see the source code. PC motherboards have had hifi quality analog audio outputs for a decade. Only the background music is mixed to back speakers in most of the movies, so multi channel audio is useless. Invest to a pair of good speakers.

My audio setup:
Amplifier: S.M.S.L SA-36A Pro
Speakers: B&W DM 601 S2

I would really not call Audio Out from any motherboard Hi-Fi, as usually any DAC over 120$ will deliver a perceivable improvement, if the rest of the audio chain is good enough to deliver it. And regarding your audio setup, I think that the amplifier dont do justice to those speakers.

A quality stereo/binaural headphones can deliver basically as much space perception as any 5.1, 7.1,... system out there.
Has to hear yet a 7.1 or other high speaker count audio system to deliver better sound experience as a binaural/true audio session on a good headphones or a good stereo speaker system.
My setup is:
Speakers: Alesis M1 active MK2
Headphones: Audeze LCD-2.2
DAC: DIY dac with AK4497
Headphone amp: DIY headphone amplifier AMB Beta22

[FireBurn]

I'd be curious to know where folk are getting these cool sound sources from? Is that what's built into 4K Blurays?

[libv]

In another thread about raven ridge, Bridgman and Alex blame the device makers for the poor raven ridge support. Wasn't atombios, which DC apparently still depends on, supposed to make such OEM woes go away? Or is this one of the things not hardcoded in the BIOS? If so, where is the documentation so that users can at least attempt to fix this issue?

[M@GOid]

In another thread about raven ridge, Bridgman and Alex blame the device makers for the poor raven ridge support. Wasn't atombios, which DC apparently still depends on, supposed to make such OEM woes go away? Or is this one of the things not hardcoded in the BIOS? If so, where is the documentation so that users can at least attempt to fix this issue?

Are you still bitter about it? Jesus man, is 2018 already, time to move on.

[holunder]

Software decoding existing for some (or all?) of these formats. The only film I've encountered that didn't include one of the older codecs was Ghost in the Shell. I think it was DTS-MA and VLC stuttered all over the place due to some bug that's probably fixed now. Fortunately mplayer had no trouble.

As far as I know, the only player that supports DTS-HD Master Audio software decoding is Kodi 17, VLC can only play Dolby TrueHD (which you will find on fewer discs), Dolby Atmos or DTS-X can only pass-throughed with the Android version of Kodi and there is no software decoding yet.

[endrebjorsvik]

Do not waste poor APUs or any high performance computers cpu cycles to hdmi audio. Hdmi audio code is large and complex, see the source code. PC motherboards have had hifi quality analog audio outputs for a decade. Only the background music is mixed to back speakers in most of the movies, so multi channel audio is useless. Invest to a pair of good speakers.

My audio setup:
Amplifier: S.M.S.L SA-36A Pro
Speakers: B&W DM 601 S2

Those speakers were good stuff 15-20 years ago. Today the rubber has probably become brittle from age, and is visibly cracking when you move the elements. At least my DM603 S2 are. And technology has evolved as well. I have transitioned to active studio monitors and I am very happy with M-Audio BX5 carbon.

[starshipeleven]

In another thread about raven ridge, Bridgman and Alex blame the device makers for the poor raven ridge support. Wasn't atombios, which DC apparently still depends on, supposed to make such OEM woes go away?

What does atombios have to do with OEMs screwing up their board ROMs?

[agd5f]

In another thread about raven ridge, Bridgman and Alex blame the device makers for the poor raven ridge support. Wasn't atombios, which DC apparently still depends on, supposed to make such OEM woes go away? Or is this one of the things not hardcoded in the BIOS? If so, where is the documentation so that users can at least attempt to fix this issue?

The register headers are all public and support for all of these formats is in place in the driver. I pointed out the relevant code in both the hda driver the gpu driver in the bug report. Anyone that wants to can take a look at it.

[direx]

Hmm...I guess I am lucky... I got no problems with sounds over HDMI on my Fury and Nano cards running Arch with 4.15.

So you are not dealing with HBR audio then.

This is something which needs to be fixed or lots users will loose interest in their APUs.

Exactly. APUs historically have been the stuff that HTPC people are keen on. And the newer AMD APUs are simply not an option for Linux HTPC people right now due to these sound issues.

The register headers are all public and support for all of these formats is in place in the driver. I pointed out the relevant code in both the hda driver the gpu driver in the bug report. Anyone that wants to can take a look at it.

Are you referring to the sound registers? From Takashi Iwais post on the Kernel Bugzilla (https://bugzilla.kernel.org/show_bug.cgi?id=196637) it looks like this is not an ALSA issue. Intel also has HBR support in place for many years now, so this is not a general driver infrastructure thing. What do you mean by "Anyone that wants to can take a look at it"? You mean AMD folks, right? Because I've tried helping with this on the FDO bug report with some printks, but things did not really lead anywhere...

[pal666]

"Anyone that wants to can take a look at it"? You mean AMD folks, right?

i think he means anyone with problematic hardware