BLIND TEST INVITE: Do digital audio players sound different?

[jabbr]

4 minutes ago, Miska said:

 

Of course you have output filter like with analog class-D too. Higher the frequency, easier it is to remove it with simple filter without effects near audio band...

 

Yes!

 

This type of design interests me quite a bit. The issue is controlling transmitted EMI from before the filter!

 

Doable but not entirely trivial.

[psjug]

4 hours ago, Miska said:

 

There are some systems that work like that, Mola-Mola and NAD for example. I don't know exact reasons, maybe because it requires some development effort that spans from DSP domain to electronics, not a small, cheap or easy task. And quite different from traditional designs for fairly conservative market.

 

The Zetex/CSR/Qualcomm DDFA technology is in a fair number of products.  Seems to be mostly toward the budget end - stuff made by NAD, Bluesound, Denon, and now Sonos.

[jabbr]

11 minutes ago, psjug said:

The Zetex/CSR/Qualcomm DDFA technology is in a fair number of products.  Seems to be mostly toward the budget end - stuff made by NAD, Bluesound, Denon, and now Sonos.

 

All of the Class-D type amplifiers use a feedback loop, analog for Class D or digital for DDFA, but consider a DAC itself: there is no such feedback loop prior to the I-V stage. Our DACs prove that it’s entirely possible to produce a low distortion current amplifier with zero feedback  

[psjug]

20 minutes ago, jabbr said:

 

All of the Class-D type amplifiers use a feedback loop, analog for Class D or digital for DDFA, but consider a DAC itself: there is no such feedback loop prior to the I-V stage. Our DACs prove that it’s entirely possible to produce a low distortion current amplifier with zero feedback  

Ha - if everything in the world would just behave the way we want then we'd never need feedback!

[fas42]

21 hours ago, bibo01 said:

Why do you think this technology is not more widely implemented in hi-fi systems?

 

One can just be shifting around the area that requires great precision - and you end up no closer: if the power supply voltage becomes everything, then you have to do a great deal to ensure that the magnitude of that is 'perfect' at all times - the slightest glitch or noise on it is directly transferred through to the speaker; highly effective distortion generation ... so, you now may need to have some corrective feedback - and it all starts getting complicated again ...

[Qhwoeprktiyns]

I listened to all four versions of the four songs and could not tell them apart - but it did seem to me that the sound quality was quite poor. Complete lack of dynamics. To check, I downloaded from Qobuz a single track - the McLoren Salvant (a waste of one euro) - and played that back on my system, against the other versions. Even playing the Qobuz download (WAV, 16/44) at lower volume it is immediately obvious that the quality is far superior than the ADC conversion that has been done.

 

The test only reveals that DAC + ADC + DAC <> DAC ! Waste of time...

[esldude]

3 hours ago, hopkins said:

I listened to all four versions of the four songs and could not tell them apart - but it did seem to me that the sound quality was quite poor. Complete lack of dynamics. To check, I downloaded from Qobuz a single track - the McLoren Salvant (a waste of one euro) - and played that back on my system, against the other versions. Even playing the Qobuz download (WAV, 16/44) at lower volume it is immediately obvious that the quality is far superior than the ADC conversion that has been done.

 

The test only reveals that DAC + ADC + DAC <> DAC ! Waste of time...

I would like to have seen the inclusion of the original file without it being labeled as such.  Might have been surprising if it had been included.  

[Qhwoeprktiyns]

2 minutes ago, esldude said:

I would like to have seen the inclusion of the original file without it being labeled as such.  Might have been surprising if it had been included.  

 

Good idea.

 

If it actually turns out that one of them is the original file then kudos to Archimago and I will look like a complete fool - but my ears tell me that is not the case and I would be very surprised if it were.

[esldude]

9 hours ago, hopkins said:

 

Good idea.

 

If it actually turns out that one of them is the original file then kudos to Archimago and I will look like a complete fool - but my ears tell me that is not the case and I would be very surprised if it were.

Here is a reprise of a short simple listening session. 

 

You can download these two zip files.  Each will open into three  30 second snippets of music.  Two are the digital originals.  One is a copy sent thru a DA to AD stage.  Tell me which one is the copy.  The odd man out.  It should be easy as the ADC is a very inexpensive unit.  Or you can say you cannot hear a difference.  

 

https://www.dropbox.com/s/sswuh8boaj6x9ze/Fiona Apple.zip?dl=0

 

https://www.dropbox.com/s/0jx0nv44ms0ddbo/Hyperion Knight.zip?dl=0

 

I don't want to pollute this thread so please PM me with your results. 

[esldude]

Never mind the above files.  There is a mistake.  I'll fix it and repost later. Sorry. 

 

[asdf1000]

On 2/15/2019 at 1:03 AM, Miska said:

There are some systems that work like that, Mola-Mola and NAD for example.

 

The NAD direct digital tech was apparently invented by Rob Watts...

 

https://www.head-fi.org/threads/chord-electronics-qutest-dac-official-thread.869417/page-6#post-13967870

[Qhwoeprktiyns]

4 hours ago, esldude said:

Never mind the above files.  There is a mistake.  I'll fix it and repost later. Sorry. 

 

Will try when you upload the files. FYI i can only play 16/44 so had to convert Archimago's files (but i dpubt that explains the difference). 

[Ralf11]

14 hours ago, esldude said:

Never mind the above files.  There is a mistake.  I'll fix it and repost later. Sorry. 

 

 

did you leave out the virus?

[esldude]

10 hours ago, Ralf11 said:

 

did you leave out the virus?

In audio circles it is known as an earworm.  

[esldude]

On 2/16/2019 at 4:03 AM, hopkins said:

Will try when you upload the files. FYI i can only play 16/44 so had to convert Archimago's files (but i dpubt that explains the difference). 

@hopkins  Here are the files.  In two polls I just posted.  You can listen and see what you think.  The originals were 16/44 and so are the files part of the poll.  So no need for you to convert. 

 

 

[Qhwoeprktiyns]

Thanks - as I mentionned in the poll, they are much harder to distinguish than in the case of the Salvant file in Archimago's test. Will try again tomorrow night (need some rest ...).

 

[Archimago]

On 2/15/2019 at 6:36 AM, hopkins said:

I listened to all four versions of the four songs and could not tell them apart - but it did seem to me that the sound quality was quite poor. Complete lack of dynamics. To check, I downloaded from Qobuz a single track - the McLoren Salvant (a waste of one euro) - and played that back on my system, against the other versions. Even playing the Qobuz download (WAV, 16/44) at lower volume it is immediately obvious that the quality is far superior than the ADC conversion that has been done.

 

The test only reveals that DAC + ADC + DAC <> DAC ! Waste of time...

 

Strange experience ("complete lack of dynamics") since the dynamic range of the recordings remain the same as the original... Plus I can run the test samples through digital subtraction with the original and demonstrate excellent null with the high quality device in the 4 samples.

 

Perhaps you should double check your gear with 24/96 playback. Another possibility is that the Qobuz track has a different mastering from the CD I used and you prefer that version?

 

[Archimago]

BTW, I know there is an "official" 24/96 version of the McLorin Salvant out there...

 

Also, the 24-bit test samples are of lower level than the original by about -2dB. Make sure to volume correct if trying to compare with an original CD / download.

[Qhwoeprktiyns]

Applying the conversion from 24->16 may be an explanation. I used dbpoweramp music converter: https://dbpoweramp.com/dmc.htm

Unfortunately, I have another DAC which plays up to 24/192 (also an ECDesigns model), but from USB, and with a little less fidelity. I will try.

 

BTW, the differences were apparent "to my ears" even when playing the original file at a lower volume.

 

Until then, perhaps my comments can be taken with a "grain of salt".

 

I did try the samples provided by esldude, and there the difference between original and converted files where much smaller.

[Archimago]

As promised, some discussions about the digital filter effects with real music and specifically with the 16/44.1 test and a couple of the devices I used:

 

https://archimago.blogspot.com/2019/02/musings-measurements-on-why-2496.html

 

 

[emcdade]

On 2/13/2019 at 5:04 PM, fas42 said:

 

People don't appreciate how powerful lower resolution captures are for revealing obvious problems - you don't listen to YouTube to find out how good it sounds; rather, it's how bad it sounds! If there are obvious deficiencies in the SQ heard over a 'poor' communications link, how much worse will these sound in the flesh?!

 

People believe that there is some magic happening when you listen to a system in the 'right place', that this will solve its clear issues ... ummm, crap!! If the sound is audibly wrong, then it's wrong - simple as that. Listening sitting with your head between two metal clamps, in a 'perfect' room, is a fudge to get your brain to ignore the obvious ...

 

Most gear isn't obviously deficient.  It's the subtleties that make or break a piece of gear over the long haul, and those will be difficult to parse out when it's been put through the processing ringer multiple times and colored by your own system chain.

[Archimago]

Reminder... Blind test closes tomorrow!

 

Enter any last-24-hour results now :-).

 

[Archimago]

Big thank-you to the participants!

 

Unblinded:

http://archimago.blogspot.com/2019/05/blind-test-results-part-1-do-digital.html

 

Discussions to follow...