ASR Open Source Streamer Project

[phofman]

Though the installation is a little clunky, Moode on a raspberry pi has evolved to the point where it hits pretty much all of the requirements you put on the first post of this thread.

I agree. IMO such a project is not only about the initial development, but mostly about the long-term ongoing support. Which is a boring and unpopular chore. Moode seems to be able to provide that. A small but very interesting https://github.com/project-owner/PeppyPlayers.doc/wiki does that too. Also Volumio is open source. Lyrion Music Server is a large project with history (though Perl not being very popular nowadays). Mopidy https://mopidy.com/ is a huge long-running project with large community, very extensible. Likely there are quite a few more. Why building a new one, instead of helping one of the existing projects?

 

[ObiWanKenobi]

Yes. Since the streamer already has the digital data, it can process it on the way out to the DAC.

What do you do in case of DSD?

 

[miero]

I agree. IMO such a project is not only about the initial development, but mostly about the long-term ongoing support. Which is a boring and unpopular chore. Moode seems to be able to provide that. A small but very interesting https://github.com/project-owner/PeppyPlayers.doc/wiki does that too. Also Volumio is open source. Lyrion Music Server is a large project with history (though Perl not being very popular nowadays). Mopidy https://mopidy.com/ is a huge long-running project with large community, very extensible. Likely there are quite a few more. Why building a new one, instead of helping one of the existing projects?

Interesting for these project could be implementing a base for the Roon Ready certification. There are tenths of non-trivial requirements (afaik non-public) that mandates Roon playback to be a first class citizen on a device.
Small manufacturers have hard times to support Roon, for example:

We did it for the SHD, it was quite a lengthy and costly process in terms of engineering resources..

Source: https://www.minidsp.com/community/t...-setting-be-changed-via-usb.23238/#post-71109

 

[somebodyelse]

Though the installation is a little clunky, Moode on a raspberry pi has evolved to the point where it hits pretty much all of the requirements you put on the first post of this thread. CamillaDSP is integrated, it does synced multiroom, etc.

@phofman listed a few others already, but there are more out there - piCorePlayer and daphile spring to mind but there are others. Since this thread started several of them have added EQ options, either built in or as plugins or customisations. RoPieee primarily targets Roon by running Roon Bridge. Squeezelite (software squeezebox endpoint implementation) is a core part of some, and an option on many, also well supported by Roon for those that need it.

 

[mcdn]

What do you do in case of DSD?

You convert it to PCM, just like a DSD DAC does internally anyway. DSD has no useful features, it’s almost as pointless as MQA

 

[ObiWanKenobi]

You convert it to PCM, just like a DSD DAC does internally anyway. DSD has no useful features, it’s almost as pointless as MQA

That's your opinion, and it's wrong.
Most DACs do this conversion by my DAC plays DSD native and PCM native and there is no conversion between them.
I cannot comment about MQA as my DAC does not support it, and it's PCM thus I don't care about disc space it saves.
If you don't hear the difference between the formats, then it's probably due to the conversion in your DAC.

 

[mcdn]

That's your opinion, and it's wrong.
Most DACs do this conversion by my DAC plays DSD native and PCM native and there is no conversion between them.
I cannot comment about MQA as my DAC does not support it, and it's PCM thus I don't care about disc space it saves.
If you don't hear the difference between the formats, then it's probably due to the conversion in your DAC.

To clarify: "most" DACs are some form of Delta Sigma DAC, and internally they are around 3-6 bits per sample. ESS DACs specifically are thought to be 6 bits, but my source for that is only https://hometheaterhifi.com/30th-anniversary/looking-back/introduction-to-dac-boxes/

Some AKM chips do have a "DSD direct" option, but not many manufacturers offer it as a selectable option. Archimago had a look at a couple here: https://archimago.blogspot.com/2021/10/measurments-dsd-testing-with-sox-dsd.html

 

[ObiWanKenobi]

To clarify: "most" DACs are some form of Delta Sigma DAC, and internally they are around 3-6 bits per sample. ESS DACs specifically are thought to be 6 bits, but my source for that is only https://hometheaterhifi.com/30th-anniversary/looking-back/introduction-to-dac-boxes/

Some AKM chips do have a "DSD direct" option, but not many manufacturers offer it as a selectable option. Archimago had a look at a couple here: https://archimago.blogspot.com/2021/10/measurments-dsd-testing-with-sox-dsd.html

See my details and check the DAC I’m using and you’ll learn it’s not like most DACs.
It has two mono paths for PCM and two mono paths for DSD.
Saying DSD is dead is totally incorrect.
DSD being PDM is far superior to PCM.
You need to learn the difference between them beside the bitrate.
Just to simplify it, put an attenuator on DSD signal and you’ll get sound, unlike PCM which you need to reconstruct the sound wave.

 

[Newman]

DSD being PDM is far superior to PCM.

You have been drinking someone's koolaid. DSD has nothing to offer to the audiophile who prioritises sound quality.

 

[phofman]

It has two mono paths for PCM and two mono paths for DSD.

Well, I was trying to learn more about the "vector step resistor network" Holo Audio claims to use for their DSD DA conversion. Google could find only quotes taken from Holo marketing materials, no relevant technical discussion https://www.google.com/search?q="vector+step+resistor+network" . I really wonder what it does, please anyone any more info? How is a resistor network used for converting a single-bit signal Thanks.

 

[mcdn]

See my details and check the DAC I’m using and you’ll learn it’s not like most DACs.
It has two mono paths for PCM and two mono paths for DSD.
Saying DSD is dead is totally incorrect.
DSD being PDM is far superior to PCM.
You need to learn the difference between them beside the bitrate.
Just to simplify it, put an attenuator on DSD signal and you’ll get sound, unlike PCM which you need to reconstruct the sound wave.

Um, the Holo Audio May is an R2R DAC. A very good one, but by definition it’s a PCM device.

 

[phofman]

Um, the Holo Audio May is an R2R DAC. A very good one, but by definition it’s a PCM device.

Well, this is what the vendor says:

HoloAudio – Cyan 2 DAC (NOS – R2R – DSD1024)

www.magnahifi.com

Equipped with R2R network and vector step resistor network at the same time. The R2R network is used for primary digital-to-analog conversion of the PCM stream, and the vector step resistor network is used for primary digital-to-analog conversion of the DSD stream.

That's why I was asking about the vector step resistor network for DSD.

 

[ObiWanKenobi]

You have been drinking someone's koolaid. DSD has nothing to offer to the audiophile who prioritises sound quality.

I would agree with you but then we would both be wrong.

 

[ObiWanKenobi]

Um, the Holo Audio May is an R2R DAC. A very good one, but by definition it’s a PCM device.

You need to get educated on it as you’re wrong

 

[Thomas Savage]

I would agree with you but then we would both be wrong.

Well I'm all for being wrong , as far as im aware there's no basis in audio science that backs the idea DSD is superior to PCM in terms of fidelity . Possibly for copyright protection at some point and lately for marketing .

If you have some actual evidence please share, the onus is on you as your claim is extraordinary . I know this can be frustrating but I'd appreciate your cooperation

You need to get educated on it as you’re wrong

Pray tell, would you mind sharing details .

I'd ask for some respect for the OP but I think the projects been a no go for some time and there's no real reason to strictly observe the original subject . Fill your boots with the off topic.

 

[voodooless]

Just to simplify it, put an attenuator on DSD signal and you’ll get sound, unlike PCM which you need to reconstruct the sound wave.

So? What's your point? Less is more?

 

[PureLIN]

Just to simplify it, put an attenuator on DSD signal and you’ll get sound, unlike PCM which you need to reconstruct the sound wave.

totally wrong, DSD can't be simply attenuate.

 

[ObiWanKenobi]

totally wrong, DSD can't be simply attenuate.

Not directly but unlike PCM it's almost
You should listen to this guy explains it far better than I can

 

[phofman]

Not directly but unlike PCM it's almost
You should listen to this guy explains it far better than I can

Hm, I could not find the part where he talks about attenuating DSD. He talks about avoiding the modulator, I did not catch a word about attenuator. Maybe you meant a standard analog attenuator, after DSD is converted to analog (with a plain capacitor, as he talks about).

 

[PureLIN]

Not directly but unlike PCM it's almost
You should listen to this guy explains it far better than I can

He is talking about "DSD CAN be easier convert to analog than PCM" which is true, one example is PowerDAC from E1DA, it even don't have a traditional DAC chip in it.

But to make your statement (if your "attenuate"=decrease audio volume) be true, we have to stay at digital domain, since after DAC, PCM and DSD are both convert to analog wave and no different here to attenuate.

In digital signal domain, to attenuate a PCM, a easiest way is float number multiply (I suppose may need some dither after that to ensure audio quality but that's another topic), for DSD, that's a lot things need to do.
That's why when product DSD music, a common practice is convert to DXD(super hires PCM) first, then convert final result back.