Jump to content
IGNORED

Qobuz FLAC decompress


Recommended Posts

11 minutes ago, KingRex said:

Yes Roon is a resource drain.  That is why I try to  break up the processing and use the server in Roon Server mode and use another computer ethernet connected to operate Roon and manage the library.  Sonically its audible to operate this way.

 

You aren't actually breaking up processing, apart from the user interface - the Roon Server is still managing the library. All you are operating on the other machine is its remote control!

We are far more united and have far more in common with each other than things that divide us.

-- Jo Cox

Link to comment
45 minutes ago, Cebolla said:

That was indeed what I was referring to - the point being MQA is the PCM audio signal source (albeit with something done to it) and therefore not a file wrapper full stop. So, MQA can be losslessly carried uncompressed by WAV & AIFF,  losslessly carried compressed by FLAC & ALAC and even carried lossily (God forbid) compressed by MP3 and the like.

 

Well, sure - but the issue is that this MQA thing that's being losslessly carried by those wrappers is already lossy in and of itself, and is already compressed. 

Link to comment
10 minutes ago, Cebolla said:

 

You aren't actually breaking up processing, apart from the user interface - the Roon Server is still managing the library. All you are operating on the other machine is its remote control!

I dont think so, but don't fully know.  In server mode there is no GUI and the other machine is the machine I manage the library on.  Its a very audible change to using my server in Core mode where I do manage the library from the server and have a GUI up.  

Link to comment
6 hours ago, mansr said:

MQA is a partially compressed format. In a typical 24-bit 48 kHz MQA stream, the upper 15 bits contain regular PCM data for frequencies up to 24 kHz. The low 8 bits are a compressed representation of the 24-48 kHz frequency band. The decoder decompresses this and joins the two bands to produce a 96 kHz audio stream with a resolution of approximately 16 bits. The compression must necessarily be lossy as, due to the fixed compression ratio, the number of possible inputs is greater than the number of possible outputs.

The quote above got me thinking about how my non-MQA DAC, when receiving a decoded MQA stream (first unfold), shows different numbers for bit depth from my different decoders.

 

Example:

I have an Oppo 205 feeding S/PDIF coax into an RME ADI-2 DAC, and I have a Bluesound Node 2i feeding Toslink into the same ADI-2 DAC. When streaming a 24/48 MQA song from TIDAL (mconnect controlling the Oppo; BluOS Controller for the Node 2i) and not using the decoders’ DACs to fully render, but rather sending the decoded stream (call it what you will, first unfold, MQA core) to the ADI-2 DAC, the external DAC reports 16/96 for the Oppo 205’s decoded stream, but shows 24/96 for the stream coming from the Node 2i. 

 

No matter what 24-bit MQA track I stream, the decoded stream from the Oppo 205 is reported as 16 bits and the Node 2i’s as 24 bits. The fr is always the same between the 2 decoders (88 or 96). 

 

Why the difference in bit depth?

 

For the record, I’m not an MQA endorser, nor am I a detractor. I’ve just been curious to see if I could notice a difference with MQA (decode only or fully rendered) vs 16/44.1 vs hires flac from Qobuz, and during my testing I noticed this bit-depth anomaly.

 

cheers

Link to comment
48 minutes ago, Gus141 said:

The quote above got me thinking about how my non-MQA DAC, when receiving a decoded MQA stream (first unfold), shows different numbers for bit depth from my different decoders.

 

Example:

I have an Oppo 205 feeding S/PDIF coax into an RME ADI-2 DAC, and I have a Bluesound Node 2i feeding Toslink into the same ADI-2 DAC. When streaming a 24/48 MQA song from TIDAL (mconnect controlling the Oppo; BluOS Controller for the Node 2i) and not using the decoders’ DACs to fully render, but rather sending the decoded stream (call it what you will, first unfold, MQA core) to the ADI-2 DAC, the external DAC reports 16/96 for the Oppo 205’s decoded stream, but shows 24/96 for the stream coming from the Node 2i. 

 

No matter what 24-bit MQA track I stream, the decoded stream from the Oppo 205 is reported as 16 bits and the Node 2i’s as 24 bits. The fr is always the same between the 2 decoders (88 or 96). 

 

Why the difference in bit depth?

That's strange. The output from the core decoder should normally be 24-bit to include the "renderer" instructions. The Oppo must be doing something unusual. If you have an S/PDIF capture interface, it would be interesting if you could grab 30 seconds or of the same track played by both devices so we could compare them.

Link to comment
2 hours ago, Gus141 said:

The quote above got me thinking about how my non-MQA DAC, when receiving a decoded MQA stream (first unfold), shows different numbers for bit depth from my different decoders.

 

Example:

I have an Oppo 205 feeding S/PDIF coax into an RME ADI-2 DAC, and I have a Bluesound Node 2i feeding Toslink into the same ADI-2 DAC. When streaming a 24/48 MQA song from TIDAL (mconnect controlling the Oppo; BluOS Controller for the Node 2i) and not using the decoders’ DACs to fully render, but rather sending the decoded stream (call it what you will, first unfold, MQA core) to the ADI-2 DAC, the external DAC reports 16/96 for the Oppo 205’s decoded stream, but shows 24/96 for the stream coming from the Node 2i. 

 

No matter what 24-bit MQA track I stream, the decoded stream from the Oppo 205 is reported as 16 bits and the Node 2i’s as 24 bits. The fr is always the same between the 2 decoders (88 or 96). 

 

Why the difference in bit depth?

 

For the record, I’m not an MQA endorser, nor am I a detractor. I’ve just been curious to see if I could notice a difference with MQA (decode only or fully rendered) vs 16/44.1 vs hires flac from Qobuz, and during my testing I noticed this bit-depth anomaly.

 

cheers

Just wondering on your OPPO 205 which firmware version you have. https://www.oppodigital.com/blu-ray-udp-203/blu-ray-UDP-20x-Firmware.aspx

The Truth Is Out There

Link to comment
1 hour ago, mansr said:

That's strange. The output from the core decoder should normally be 24-bit to include the "renderer" instructions. The Oppo must be doing something unusual. If you have an S/PDIF capture interface, it would be interesting if you could grab 30 seconds or of the same track played by both devices so we could compare them.

I should be able to record the S/PDIF inputs of the RME ADI-2 DAC (via USB connected to my laptop according to the RME manual); I haven’t explored this function of the DAC so it might take me some time.

 

I think you are correct: the Oppo is doing something unusual. It’s why I prefer to use the Node 2i to stream to the external DAC. 

 

I’ll try sending TIDAL MQA streams to the Oppo 205 via the desktop app over USB (which I know yields a full MQA decode & render over the Oppo’s analog outputs) and see what the S/PDIF output shows on the external DAC for just the core unfold. I expect it to show 24 bit.

Link to comment
2 minutes ago, mav52 said:

Just wondering on your OPPO 205 which firmware version you have. https://www.oppodigital.com/blu-ray-udp-203/blu-ray-UDP-20x-Firmware.aspx

UDP20X-60-0625 (June 29, 2018) with USB DAC firmware USB-0118 for MQA playback support via USB DAC input (which I have confirmed works as advertised). I haven’t tried the newest public beta.

 

I started using mconnect recently to act as a DLNA controller to the Oppo (mconnect has TIDAL and Qobuz support built in) and MQA looked like it was being decoded and rendered correctly when using the Oppo analog outs. When I connected the Oppo’s S/PDIF OUT to the external DAC, I was suprised to see 16-bit showing on the screen when playing an MQA track via mconnect connected to the Oppo over DLNA. 

 

 

Link to comment
10 hours ago, KingRex said:

I dont think so, but don't fully know.  In server mode there is no GUI and the other machine is the machine I manage the library on.  Its a very audible change to using my server in Core mode where I do manage the library from the server and have a GUI up.  

https://kb.roonlabs.com/Software_packages

It seems server puts GUI and controls elsewhere.  library and output are still on the core.  Amazing so little processing power can affect the sound so much. 

Link to comment

I did some hard listening today.  Music on my nas is much better than flac files on Tidal or Qobuz.  Now I'm not sure how much this has to do with going through my router and modem. I do have a pretty decent network. I have Blue Jeans cables and a Linear Solution switch.    I made my comparison to nas music instead of on the harddrive to make sure some amount of the comparison is through my ethernet.  

 

Then again is could be decompressing FLAC files.  Now its time to see how my offline Qobuz library sounds compared to on the NAS.  

Link to comment
On February 6, 2019 at 4:58 PM, KingRex said:

FLAC is lossless compression like MQA.  My setup is designed to do best up to 24/192 pcm.  No decompressing or upsampling.  Thats my setup.  If I purchase high resolution FLAC,  I need to use DBPoweramp to decompress and turn it onto WAV.  

 

Now I need to find out if I can put Qobuz offline library files onto my drive pre, decomptessed.  

 

MQA is lossy, not lossless. 

Link to comment
  • 3 weeks later...
On 2/9/2019 at 8:14 AM, KingRex said:

I did some hard listening today.  Music on my nas is much better than flac files on Tidal or Qobuz.  Now I'm not sure how much this has to do with going through my router and modem. I do have a pretty decent network. I have Blue Jeans cables and a Linear Solution switch.    I made my comparison to nas music instead of on the harddrive to make sure some amount of the comparison is through my ethernet.  

 

Then again is could be decompressing FLAC files.  Now its time to see how my offline Qobuz library sounds compared to on the NAS.  

Any joy getting streaming to match file playback? 

Link to comment
13 minutes ago, Rexp said:

Any joy getting streaming to match file playback? 

No.  Actually been playing files stored internal and they are by far the best.  I may get a Gigafoil and Keces PS.  I am hearing they are adding to streaming playback with little if no degradation in sound.,  Actually everyone I  hear raves about the performance and says it makes it better with no losses.  I'm always cautious but who knows.  Maybe this is the one.

Link to comment
On 2/8/2019 at 10:32 AM, KingRex said:

https://kb.roonlabs.com/Software_packages

It seems server puts GUI and controls elsewhere.  library and output are still on the core.  Amazing so little processing power can affect the sound so much. 

 

I'm not sure calling GPU accelerated UI to be using a little processing power is proper, anyway, here's what Roon CTO said:

 

https://community.roonlabs.com/t/roonserver-over-roon-core-on-macmini/3793/18?u=wklie

 

https://community.roonlabs.com/t/build-99-does-it-sound-as-good-as-the-previous-build-to-you/7268/21?u=wklie

 

If your GPU has a fan, GPU activity may have an even bigger impact.

Peter Lie

LUMIN Firmware Lead

Link to comment
9 hours ago, wklie said:

 

I'm not sure calling GPU accelerated UI to be using a little processing power is proper, anyway, here's what Roon CTO said:

 

https://community.roonlabs.com/t/roonserver-over-roon-core-on-macmini/3793/18?u=wklie

 

https://community.roonlabs.com/t/build-99-does-it-sound-as-good-as-the-previous-build-to-you/7268/21?u=wklie

 

If your GPU has a fan, GPU activity may have an even bigger impact.

I don't use a stand alone graphic card.  Just what is on the motherboard.  I agree, graphics can be gigantic power soaks.  Ibelieve in doing everything to eliminate noise in my setup.  That is why I don't upsample.

 

  In a couple weeks I am going back to Linux.  Its close now.  Software is almost complete.  Once I hear that version I will decide if I need the Gigafoil with LPS.  For the last week I have been moving my favorite music from my NAS to my internal drive.

Link to comment
45 minutes ago, KingRex said:

I don't use a stand alone graphic card.  Just what is on the motherboard.

 

I know.  Even if you don't have a standalone graphics card, you still have a GPU inside the processor on your motherboard.

Peter Lie

LUMIN Firmware Lead

Link to comment

I tried to look in my resource monitor to see what the internal graphics were using but nothing stood out or looked to be graphics related.  Roon and HQ player were using the most of any one service.  But, there are a lot of services running.  Skype and other crap were going in the background.  I shut them down.  I thought AO would have taken care of that.  

Link to comment
On ‎2‎/‎7‎/‎2019 at 2:43 AM, KingRex said:

Qobuz high resolution say its FLAC.  Does this mean my server is having to decompress the file to get back to lossless.  I was doing a critical listen last night and my internal WAV 16 44.1 files seem to have more life and realness than the 24 96 FLAC files.

 

I have observed  this before with high resolution files I own.  WAV is better than FLAC.  I attribute it to processing noise in my server.  Who knows the real reason.  

 Yes, system noise is a major part of the answer, but Admin will not accept that .wav can sound better than .flac !

He thinks that those who prefer .wav over .flac are far more likely to prefer stupid MQA too.:D

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment

If there were no such thing as system noise, I would think a 24 bit flac file would sound better than a 16 bit wav file as the  dynamic range and signal to noise ratio should be superior.   

 

Are you saying there is something inherent between wav vs flac that processing noise aside, one sounds better than the other. 

Link to comment
7 hours ago, KingRex said:

Are you saying there is something inherent between wav vs flac that processing noise aside, one sounds better than the other. 

All that I am saying here, is that I have found that the quieter electrically the PC is, the smaller the differences between .flac and the original .wav file.

 Note also the quote from Cookie Marenco in my signature.

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...