Jump to content

Computer-playback vs streaming client: audio-stream in best quality?


Recommended Posts

Posted (edited)

It‘s too long ago, so I don’t properly remember all details: just, that Mac has advantages over Windows in handling audio. No „reprocessing“ of audio inside the operating system, afaik.

I am a Qobuz subscriber, so I have access to their library up to 192kHz/24bit. In my living room, I use a Linn Klimax DS for listening. 

In my study, I have a MacMini, a professional RME outboard AD/DA (attached via USB) and active Neumann speakers. I could play back Qobuz via their browser-based player. Considering the RME soundcard is top notch quality (including its internal very stable clock): what effect on sound quality does playback of the Qobuz-stream via a browser and on Mac-OS have? Is there any difference in the handling of the audio-stream between a dedicated streaming client and a Mac computer?

Edited by dvdr
Link to comment
Share on other sites

  • dvdr changed the title to Computer-playback vs streaming client: audio-stream in best quality?
2 hours ago, dvdr said:

Is there any difference in the handling of the audio-stream between a dedicated streaming client and a Mac computer?

Well you have the equipment, so you tell us :/

  • Haha 1
Link to comment
Share on other sites

13 minutes ago, Man in a van said:

you have the equipment

It sure sounds different - but is different speakers, different rooms😉 

Now, back to serious: I was asking technically (like in „Windows resamples, as long as you do not use ASIO, Mac is…“).

Link to comment
Share on other sites

Seriously neither a Windows PC or a MAC are anything like ideal for being a music source . Both are designed to do other things and are a very noisy place for any audio signal to come from.

A Raspberry Pi dedicated to being a music server with a digital output would sound much better use much less energy (and so be greener) and more improtant i am sure would sound better as well . The cost is not huge to construct one and then it just sits there and works.

  • Like 1
  • Upvote 2
Link to comment
Share on other sites

Well you can always move the Linn into your other room and compare

Windows does not have to have a fixed output

Mac OS will also perform with a fixed output unless reset

So both OS basically do the same in default state


There might be a technical difference in the way the  Qubuz browser app processes the the internet connection and the way the Linn does.

Linn will use an API from Qobuz to perform the necessary decoding and presentation of Album art and meta data

The Qubuz app will perform the same thing within the browser, what differences there are technically, performance and SQ wise, I have no idea.

There will for sure, be a difference in the design, layout and function ot the motherboards in the Mac-mini and the Linn, which may have some bearing on output, but who knows?

For example, I read on another forum that a Google Chomecast Audio  is basically a web browser

ronnie

Link to comment
Share on other sites

3 hours ago, dvdr said:

It‘s too long ago, so I don’t properly remember all details: just, that Mac has advantages over Windows in handling audio. No „reprocessing“ of audio inside the operating system, afaik.

I am a Qobuz subscriber, so I have access to their library up to 192kHz/24bit. In my living room, I use a Linn Klimax DS for listening. 

In my study, I have a MacMini, a professional RME outboard AD/DA (attached via USB) and active Neumann speakers. I could play back Qobuz via their browser-based player. Considering the RME soundcard is top notch quality (including its internal very stable clock): what effect on sound quality does playback of the Qobuz-stream via a browser and on Mac-OS have? Is there any difference in the handling of the audio-stream between a dedicated streaming client and a Mac computer?

Many of us would not be so far down the streamer rabbit hole (and equally the player software rabbit hole) if there were not significant differences in the sound of different streamers despite supposed ‘bit perfect’ playback. 
 

So you should hear those differences. Only attempt to find out why there are differences if you are prepared for the time and money consequences of then trying to improve the sound quality!

  • Upvote 1
Link to comment
Share on other sites

9 hours ago, bencat said:

Seriously neither a Windows PC or a MAC are anything like ideal for being a music source . Both are designed to do other things and are a very noisy place for any audio signal to come from.

A Raspberry Pi dedicated to being a music server with a digital output would sound much better use much less energy (and so be greener) and more improtant i am sure would sound better as well . The cost is not huge to construct one and then it just sits there and works.

I'm sat here listening to music being streamed via a Raspberry Pi 3b+ with an Allo Digione Signature 'HAT', running the free Moode Audio software and using the BubbleUPnP software on my tablet or phone. The Allo HAT has 2 boards, borh being powered by a Temple Audio Supercapacitor Supercharger (essentially a battery supply). The music is via Qobuz or ripped cd's on an attached flash drive. I can honestly say that I've not heard a more realistic reproduction. It seems that the less work the renderer's CPU has to do, the better the sound that it delivers.  increasing the buffer sizes was the final icing on the cake. Digital glare, brightness and compression has now gone. All-in, the streamer including power supply cost me just over £550, a bargain for the results it's giving.

  • Like 2
Link to comment
Share on other sites

Posted (edited)

My experience is as follows. When getting into Roon and Qobuz I had Roon core on a Microsoft Surface pro connected to a Benchmark DAC2 via USB cable.

On switching to the core on a i7 NUC via ethernet to a Stark Audio Link II I detected no change is quality but a big increase in convenience. 

Regards Andrew 

Edited by andrew s
  • Upvote 1
Link to comment
Share on other sites

On 24/05/2021 at 13:45, dvdr said:

It‘s too long ago, so I don’t properly remember all details: just, that Mac has advantages over Windows in handling audio. No „reprocessing“ of audio inside the operating system, afaik.

I researched this topic when I rebuilt my system in 2017: $MSFT have removed the problem areas with progressive releases of Windows 10. They first introduced flac playback, then reworked WASAPI to make it much faster (not as fast as ASIO ... but fast enough for non-specialised applications such as streaming) and finally upgraded the USB audio driver to Class 2 ($APPL beat them to Class 2 by many years). We don't have to worry about WASAPI because there is a application switch now to bypass it and go straight to the USB driver. Moreover the mixer is 64-bit, so conversion is bit-perfect for lower bit rates in any event. See:

Desktop Qobuz: PC playback modes.

I don't think Windows will handle 24-bit audiostreams yet though, remaining with the PCM standard at 16-bit ... but here Foobar2000 comes to our rescue. That freeware can be configured to use the WASAPI bypass and play 24/96, 24/192 ... and all the way up to SACD's.

The resources required to support audio playback are small, computing-wise: I use a LINX1010E tablet as a streamer (£100 second hand from eBay). This device collects data by wi-fi from a music server in the next room and pumps it to the nearby DAC via a USB cable.

I have a document explaining all this if anyone is interested.

  • Like 1
Link to comment
Share on other sites

On 24/05/2021 at 13:45, dvdr said:

In my study, I have a MacMini, a professional RME outboard AD/DA (attached via USB) and active Neumann speakers. I could play back Qobuz via their browser-based player.

I do not know the architecture of Mac's ...

... but flac playback from a browser is not bit-perfect on Windows 10.

The BBC had a trial whose technical description explains why browser modifications are required. See:

BBC Taster - Radio 3 Concert Sound

Link to comment
Share on other sites

13 minutes ago, JackoUK said:

I do not know the architecture of Mac's .. but flac playback from a browser is not bit-perfect on Windows 10.

That's why I asked. I use a Mac and a bigger Windows machine for music production (ASIO drivers, ProTools, RME AD/DA), so that part definitely is up to professional studio standards. What I do not know, thought, is how - so to say - "routing" of a browser-based playback is handled. I can imagine - since this is the default in my system - it also might be routed directly to the ASIO drivers on my Mac - but I can only guess. There may be "intermediaries" involved - hence my question. I definitely will look into your BBC link, thanks for posting it.

Link to comment
Share on other sites

Looking at this page ...

How do I experience Hi-Res on Mac? – Qobuz Help Center

... and similar for other operating systems ...

... it seems to me that the data interface for Qobuz is a flac audiostream.

I suspect Qobuz have done what the BBC did and the web player will feed flac data to an output target of either a DAC or a Chromecast device.

However your RME AD/DA kit is something else altogether and potentially incompatible.

You need to talk to Qobuz technical support to confirm the web functionality and see if they have any ideas about your RME AD/DA kit. 

Link to comment
Share on other sites

  • Moderator
On 25/05/2021 at 16:51, JackoUK said:

($APPL beat them to Class 2 by many years)

Not strictly true. Class 2 enables plug and play. But Windows devices come with drivers that enable the full capability of the associated device. The result is that while Class 2 operation can actually bottleneck performance, Windows devices have always been able do everything so long as the right drivers were used. You are correct in that Windows have only recently added a generic class2 driver to their OS, but the lack of that was never a problem because of the device specific drivers which come with every sound card and DAC.

  • Like 1
Link to comment
Share on other sites

  • Moderator
On 24/05/2021 at 13:45, dvdr said:

It‘s too long ago, so I don’t properly remember all details: just, that Mac has advantages over Windows in handling audio. No „reprocessing“ of audio inside the operating system, afaik.

I am a Qobuz subscriber, so I have access to their library up to 192kHz/24bit. In my living room, I use a Linn Klimax DS for listening. 

In my study, I have a MacMini, a professional RME outboard AD/DA (attached via USB) and active Neumann speakers. I could play back Qobuz via their browser-based player. Considering the RME soundcard is top notch quality (including its internal very stable clock): what effect on sound quality does playback of the Qobuz-stream via a browser and on Mac-OS have? Is there any difference in the handling of the audio-stream between a dedicated streaming client and a Mac computer?

The defining factor here is not the computer you use, but the DAC. The RME has a very good USB input and so will offer excellent quality. You could get small improvements by using different processor platforms as an intermediary. The Pi based platforms are mentioned here with good reason. But you end up with additional devices and their power supplies which make the whole thing more unwieldy. It kind of goes against the very simple source/DAC/active speakers principle. Plug, play and off you go. The caveat is of course this is my experience and opinion. If you wanted to experiment with a Pi and Moode as an example I wouldn’t think that was wrong. Just different to what I would do now.

Link to comment
Share on other sites

"($APPL beat them to Class 2 by many years)." "Not strictly true."

$APPL released their Class 2 driver with Snow Leopard 10.6.4 in June 2010. $MSFT released their driver with Windows 10 Version 1703 in April 2017. I'd say my claim was strictly true: can we agree 'many= about 7'?

The primary goal of both the 'USB' standard and the import of the prefix 'Class' applied to drivers is interoperability ... designed to sate we poor consumers desire for convenience (and build up profitable ecosystems!). Coupled in the best designs with backwards compatibility.

"... the lack of that was never a problem because of the device specific drivers which come with every sound card and DAC."

Rather than lumping the issue with other 'problems', let me expand by linking it to inconvenience and a hidden unwelcome cost.

As you say Class 2 enables plug and play: that's what we lazy saps want. I do not want to have to read a DAC vendor's manual explaining how to install a Windows driver - didn't I just buy an expensive piece of audio kit which says USB good to go? Installing a vendor-specific driver contravenes the interoperability goal ... and can lead to problems down the road when technology changes. Whatever you think of $INTL and $MSFT generally, those 2 companies have achieved a lot with compatibility and interchangeability.

So, if the operating system lacks the necessary class 2 driver, where does a DAC vendor obtain the same for his kit? He pays a software house to develop a bespoke class 2 driver! And who ends up paying for this development? You do ... but not in an obvious way: the DAC vendor hides the licensing fee somewhere in his pricing structure. Now that $MSFT has eventually 'done the right thing' we ought to push DAC vendors to adopt the $MSFT Class 2 driver. I wrote to mine - no reply.

A similar situation obtains for ASIO.

"The result is that while Class 2 operation can actually bottleneck performance ...". Please expand.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...