Quality of sound between Deezer streaming and Flacs stored on NAS

What media server are you running on the Synology? Worth checking that it isn’t set to transcode to MP3. 

It’s possible that you are just accustomed to the sound of the lossy Deezer stream. Try only listening to the NAS for a couple of days, then go back to Deezer and see what you think. 

Thanks Chrissu for your fast response. Appreciate.

The Synology uses the standard package Media server. All the boxes are unticked. Is that okay?

Your suggestion of listening to the NAS for a longer time makes sense. I will certainly do this.

The Naim app should display the bitrate for the currently playing track. What does it report?

It shows Flac 192 when playing a Flac from the NAS. That file is indeed a Flac 192.

And is your Deezer playing MP3 at (320kbps). If so you are probably noticing the limitations of your replay equipment... and it might be more suited to playing regular FLAC at 44.1/16 or MP3... and the extra information and energy in the hidef media is making the streamers sound less comfortable. MP3 tends to contain less HF information and so may well sound warmer on some equipment.. at the expense of insight, neutrality, realism and presence... but if your replay equipment is unable to show these attributes anyway you are perhaps better off just playing what sounds nicest.. and if MP3 so be it.

@Simon in Suffolk; Do you think the Mu-so is not capable enough to reproduce 24/192?

Following on from Simon’s comment, I would certainly try comparing Deezer to 16/44 versions as well as hi-res and see how that compares. You may be able to set the server to transcode 24/192 to 16/44, or just try some CD rips. 

Also, I think it’s important not to jump to conclusions. To me, sometimes something can sound worse at first, when actually it is just different to what I’m used to, and my slow brain just needs a bit of time to adjust and appreciate what I’m hearing. 

I will certainly do some tests with the various formats played from the Nas and find the differences. Soon as I have finished this I will let you know. Thank you all for helping!

Matthijs posted:

@Simon in Suffolk; Do you think the Mu-so is not capable enough to reproduce 24/192?

I understand it is entirely capable of rendering that resolution, whether it’s optimal for the Muso is another matter.

I see.

Matthijs posted:

I will certainly do some tests with the various formats played from the Nas and find the differences. Soon as I have finished this I will let you know. Thank you all for helping!

While you're experimenting,  don't forget to test transcoding,  this means the Synology sends a WAV stream to the Muso.     You set this in the Synology Media Server DMA Compatibility tab.    Tick both the enable transcoding & the FLAC boxes,  tick them all if you want.  

We have a couple of Qbs and they are certainly less critical of poorer material than the main system. For example, Spotify is listenable on the Qb but horrible on the main system. That said, the mu-so is capable of great sound and does sound a lot better with steamed music from the nas than on 320k Spotify. I imagine Deezer streams CD quality and that might be worth trying. Maybe they do free trials. 

Thank you for your useful tips and remarks!

While you're experimenting,  don't forget to test transcoding,  this means the Synology sends a WAV stream to the Muso. 

@Mike-N: does wav offer a better quality than flac?

If you send WAV, there is less work for the streamer to do to convert it into what it needs to play. That’s why it should sound better. 

It was my understanding that the Synology Media Server was unable to transcode gapless files properly, but maybe that has now changed. Most people with Synology use MinimServer. I think Asset is now available for Synology, and it’s easier to use than Minim. 

Aha, I see.

Matthijs posted: 

While you're experimenting,  don't forget to test transcoding,  this means the Synology sends a WAV stream to the Muso. 

@Mike-N  B: does wav offer a better quality than flac?

No not better quality, although some would argue that point.  Naim streamers are reputed to sound better on WAV, that is true with my NDX although the difference is minimal    Its just that while you are experimenting,  for the sake of a couple of tick boxes,  its worth a go.

Hungry Halibut can add some of his thought on this as he has Muso's & transcodes from FLAC   (I have only WAV & DSD)

hungryhalibut posted:

..................   I think Asset is now available for Synology, and it’s easier to use than Minim. 

Best not go there yet.   It's a Raspberry adaption & is a bit tricky to set up.    However Illustrate are developing an Asset UPnP for use with Synology & its at the public beta stage at the moment.   It's a very simple download & install & there are no extras such as Java as needed with Minimserver.  It works really well,  As I say its still in development but it appears to be not so far away from full release. 

In this modern age where cpus are very capable in most devices I am very doubtful someone can hear any  audible difference between flac and wav as I have never heard none in 10 years of streaming audio. Another audiophile myth if you ask me. You could argue that wav is using more bandwidth and putting more stress on the network circuit of your streamer and adding noise. noi be this is the reason  why so many audiophiles buy these expensive network cables

SimonPeterArnold posted:

In this modern age where cpus are very capable in most devices I am very doubtful someone can hear any  audible difference between flac and wav 

Have to disagree on that one Simon,  I can hear a difference with my NDX,   I am not saying WAV is better, but I do hear a difference.    

Mike-B posted:

SimonPeterArnold posted:

In this modern age where cpus are very capable in most devices I am very doubtful someone can hear any  audible difference between flac and wav 

Have to disagree on that one Simon,  I can hear a difference with my NDX,   I am not saying WAV is better, but I do hear a difference.    

[@mention:73793444532682398]

Sorry Simon, I have to agree with Mike-B, same here : there is an audible difference between FLAC and WAV for the same track (very well recorded one). Made several tests, and even made the "wife test" 

We will agree to disagree then but thats the beauty of this daft hobby. 

Again Simon, I disagree, it’s not a myth at all.. and given the extent that Naim go to to mitigate but not eliminate entirely suggests they are very much aware of it and design around it too. (In fact I know at the engineering level they are as I have spoken with some of their design engineers on this ). This is all about processor execution ground plane and powerline modulation, as well as possibly some local EM field radiation although layered PCBs help reduce.

Now in the DSP for example code execution timing is critical, and Naim software engineering  can almost modify the execution timing to act as a digital tone control based on the noise profile from the processor... its a little hit and miss and sometimes has unpredictable results, but is why there are severeal beta firmwares that all sound different, as their code execution timing has been altered, until they finalise on the best sounding firmware, albeit the code operands for the SHARC processor are completely unchanged.

The same to principle applies to the media converter code.. and Naim have optimised WAV PCM rendering processing over FLAC.

Now if you decouple,  to all intents and purposes,  the renderer from the DAC, such as using the SPDIF output into a reclocked and isolated DAC, the differences between FLAC and WAV rendering become very much minimal to not apparent with a Naim streamer.. however switch to the more closely coupled DAC in built into the streamer and the difference between FLAC and WAV rendering becomes very much more apparent.

Mike-B posted:

hungryhalibut posted:

..................   I think Asset is now available for Synology, and it’s easier to use than Minim. 

Best not go there yet.   It's a Raspberry adaption & is a bit tricky to set up.    However Illustrate are developing an Asset UPnP for use with Synology & its at the public beta stage at the moment.   It's a very simple download & install & there are no extras such as Java as needed with Minimserver.  It works really well,  As I say its still in development but it appears to be not so far away from full release. 

Thanks Mike. I bought a Qnap in order to use Asset, as I got fed up with Minim. Both are fine and well specified servers though, but Minim is a bit more of a faff. 

Simon-in-Suffolk posted:

Again Simon, I disagree, it’s not a myth at all.. and given the extent that Naim go to to mitigate but not eliminate entirely suggests they are very much aware of it and design around it too. (In fact I know at the engineering level they are as I have spoken with some of their design engineers on this ). This is all about processor execution ground plane and powerline modulation, as well as possibly some local EM field radiation although layered PCBs help reduce.

Now in the DSP for example code execution timing is critical, and Naim software engineering  can almost modify the execution timing to act as a digital tone control based on the noise profile from the processor... its a little hit and miss and sometimes has unpredictable results, but is why there are severeal beta firmwares that all sound different, as their code execution timing has been altered, until they finalise on the best sounding firmware, albeit the code operands for the SHARC processor are completely unchanged.

The same to principle applies to the media converter code.. and Naim have optimised WAV PCM rendering processing over FLAC.

Now if you decouple,  to all intents and purposes,  the renderer from the DAC, such as using the SPDIF output into a reclocked and isolated DAC, the differences between FLAC and WAV rendering become very much minimal to not apparent with a Naim streamer.. however switch to the more closely coupled DAC in built into the streamer and the difference between FLAC and WAV rendering becomes very much more apparent.

Thanks for your opinion and technical input but I still maintain from all my experience I hear no audible difference across the systems I have had.  I went round this years ago when I used LMS and squeezebox products as there was one side saying do the transcoding to PCM on the server as it improved sq  over using the device to do it. Again heard no difference. I do also feel audiophiles look for things that are not important at all in an attempt for audio nirvana they will never really find. But hey that's another topic all together.

Simon ok fair enough, and yes any resultant noise profile will be system dependent. However  I am referring to the phenomenon of processor noise coupling in a closed system from an engineering perspective rather than any view of an audiophile which to my closest understanding  is simply a person who appreciates recorded audio in all its forms. From my personal perspective I effectively decouple my DAC from my Naim streamer renderer to isolate Ethernet interframe timing noise and media codec rendering noise from my playback so thess issues don’t affect me for the most part anymore. (But not removed entirely)