Are 48 kHz enough to non-bat humans?

[antcollinet]

My concern is more about resampling, I agree with you that 48 kHz are enough

I don't think you need to worry about Neumanns resampling. While as pointed out above it is possible to get it wrong - a company that fails in that is not going to become as successful as Neumann, nor have such good measurements.

And it makes little sense to do the resampling on the Wiim. While I have little doubt that the Wiim can also do it properly, between the two companies it is Neumann that has the more impressive long term track record until now.

 

[Miguelón]

Curiously the seller told me the opposite: better do the resample on the streamer
The only way is to listen by myself, I will share the impressions to other members, as is not very frequent to use studio monitors for home listening.
I agree with you that Neumann has a solid reputation but I have some doubts about their experience in software domain.

 

[BDWoody]

There's Chinese saying, remove pants before farting.

That's hilarious.

We might need a thread to find the best 'sayings' from around the world.

 

[Miguelón]

That's hilarious.

We might need a thread to find the best 'sayings' from around the world.

Farting is another acoustic topic that merits its own thread.

 

[MaxwellsEq]

When I play Apple Music or Tidal with the limitation of 44.1/ 48 kHz in the app settings, I can perceive a degradation in sound on tracks that are 44.1 kHz or 48 kHz native

The problem here is you have no way of guaranteeing that you are comparing exactly the same version of the master. It's more likely that you are comparing masterG on 44.1 and masterU on 48.

 

[Miguelón]

The problem here is you have no way of guaranteeing that you are comparing exactly the same version of the master. It's more likely that you are comparing masterG on 44.1 and masterU on 48.

Interesting, I didn’t know the possibility of different audio masters in the same streaming platform…

 

[MRC01]

I don't think you need to worry about Neumanns resampling. While as pointed out above it is possible to get it wrong - a company that fails in that is not going to become as successful as Neumann, nor have such good measurements. ...

One of the devices that had improper resampling was an Oppo BDP-83 disc player. It had settings to set the max sample rate emitted from its digital outputs. If the source being played exceeds this limit, the BDP-83 downsamples it accordingly. The downsampling is always at integer multiples so you'd think they would get it right, especially from such a reputable company. But I found a particular 192k track that when the BDP-83 downsampled to 96k, it created audible hiss. I don't know what it was about that track that triggered this bug, but it was consistent and repeatable. The track was this album - music, not a test signal: https://play.qobuz.com/album/s8n6e4n2jmtuc
PS: downsampling the same track on my Linux PC with Sox was clean & transparent, no hiss or other artifacts. It was definitely a problem specific to the BDP-83.

 

[MaxwellsEq]

Interesting, I didn’t know the possibility of different audio masters in the same streaming platform…

It's a mess. Generally streamers only publish the most recent package from the Rights holders. This might well be remastered (good), but highly compressed (bad). I have some albums where remasters/remixes are brilliant (Steven Wilson's work) and others where it's dead and non-dynamic. Different masters get released at different times, so a streamer's 44kHz version may be the 2016 remaster, but 88kHz is the 2009 remaster.

 

[MRC01]

Interesting, I didn’t know the possibility of different audio masters in the same streaming platform…

It's a mess. Generally streamers only publish the most recent package from the Rights holders. This might well be remastered (good), but highly compressed (bad). I have some albums where remasters/remixes are brilliant (Steven Wilson's work) and others where it's dead and non-dynamic. Different masters get released at different times, so a streamer's 44kHz version may be the 2016 remaster, but 88kHz is the 2009 remaster.

Yep. Qobuz has so many it's confusing because they're not always clearly labeled. For example they have 3 versions of the Moody Blues Days of Future Passed, at least 2-3 versions of various Yes, Rush and Jethro Tull albums, some of which are original, others Steven Wilson remixes.

 

[Miguelón]

It's a mess. Generally streamers only publish the most recent package from the Rights holders. This might well be remastered (good), but highly compressed (bad). I have some albums where remasters/remixes are brilliant (Steven Wilson's work) and others where it's dead and non-dynamic. Different masters get released at different times, so a streamer's 44kHz version may be the 2016 remaster, but 88kHz is the 2009 remaster.

Very interesting! But after your answer I checked again the phenomenon: if you carefully listen on headphones, and change the settings from 44.1 kHz to max 192 kHz during the reproduction you will notice a difference in quality.

This cannot be a different master: the app has stocked the current track on the caché memory, they only execute change quality when going to next track (as their own information tells).

And you can hear the subtle change back and forth no matter the track!

Maybe a commercial strategy to encourage the premium master subscription?

 

[antcollinet]

Very interesting! But after your answer I checked again the phenomenon: if you carefully listen on headphones, and change the settings from 44.1 kHz to max 192 kHz during the reproduction you will notice a difference in quality.

Are you sure you are actually hearing that? You can't be sure if you haven't done the comparison blind.

If you are hearing a differece, it is most likely coming from the filter at 44.1kHz rolling off slightly in the audible band. However this is at such a high frequency and at a pretty low level (If you have a (correct) sharp filter selected) then you are pretty unlikely to be able to hear that unless you are a teenager.

Possibly you are more likely to hear it if you have selected a slow roll off filter in your DAC - if it has selectable filters.

 

[valerianf]

“If you are hearing a differece, it is most likely coming from the filter at 44.1kHz rolling off slightly in the audible band."
Exactly, here is one of the issue. Do you believe in the marketing arguments that the manufacturer use to prove that his design is perfect?
Only Amir can measure the filter attenuation.
In the doubt I select the 96khz track: I am sure that the filter will behave properly.

 

[Miguelón]

Are you sure you are actually hearing that? You can't be sure if you haven't done the comparison blind.

If you are hearing a differece, it is most likely coming from the filter at 44.1kHz rolling off slightly in the audible band. However this is at such a high frequency and at a pretty low level (If you have a (correct) sharp filter selected) then you are pretty unlikely to be able to hear that unless you are a teenager.

Possibly you are more likely to hear it if you have selected a slow roll off filter in your DAC - if it has selectable filters.

I have tried with various DACs, actually hearing with the headphone output of my Focusrite Scarlett 2i2 4th gen. and Sennheiser hd 600 and i notice the same difference.

As you said is more audible in the highs, is subtle but as I mentioned above I have a huge congenital visual impairment and I trust a lot on my hearing.

Perhaps this explains why I’m able to perceive those differences: I don’t think I can hear above 15.000 Hz because of my age (this relays on physical aging process on the inner ear and not on the acoustic cortex of the brain) but I’m quite sensitive to minimum color changes of the sound.

Love the combination of Scarlett and Sennheiser, by the way: my Ifi Zen Signature colors a lot more and costed double the price, actually I sell it on Wallapop

P.S: I’m not saying that 44.1 kHz settings sound bad, just my representation of the sound scene is not as satisfying as the one I listen on the max. quality setting. I have a neuronal phenomenon called synesthesia: for me frequencies are identified with visual positions, lows are perceived as near the ground and highs goes more to the celling. In stereo imaging (despite I’m actually listening with headphones with one single driver) I can see a rectangle if the representation is complete and flat. Sometimes, depending on the system used I see an elipse stretched to the sides, without the high and low corners of my reference rectangle. For example with humble apple earbuds (cabled, bluetooth I cannot support it).
When coloration happens, as with my Zen Dac, I can see the rectangle stretched in some sides (midhighs and low bass), the vertical sides corresponding to the frequency response curve I guess…

People usually have horizontal representation of the stereo image, wider or narrower, center focused etc. Mine is bidimensional as a cinema screen and frequencies play the role of vertical axis.

 

[antcollinet]

as I mentioned above I have a huge congenital visual impairment and I trust a lot on my hearing.

This may well make you more sensitive to marginally audible differences. I've no real knowledge here - but I guess it is possible your synesthesia might also reveal differences that are not perceived as sound differences by others.

But just so you are aware, it doesn't make you immune to the very human processes of cognitive biases - which can cause us to perceive differences that don't exist in the sound waves. I would also expect that your synestheseia would also be prone to be altered by such biases. (We are just as prone to optical illusion as audible illusion - and synesthesia seems to be a mixing of audio and optical processing)

This is why there is so much emphasis here on blind testing when we think we are hearing differences that the measurements suggest are likely to be inaudible.

Again - I'm not stating that you are NOT hearing a real difference here. But it is not possible to be sure you ARE if you are doing uncontrolled listening.

 

[Miguelón]

“If you are hearing a differece, it is most likely coming from the filter at 44.1kHz rolling off slightly in the audible band."
Exactly, here is one of the issue. Do you believe in the marketing arguments that the manufacturer use to prove that his design is perfect?
Only Amir can measure the filter attenuation.
In the doubt I select the 96khz track: I am sure that the filter will behave properly.

This may well make you more sensitive to marginally audible differences. I've no real knowledge here - but I guess it is possible your synesthesia might also reveal differences that are not perceived as sound differences by others.

But just so you are aware, it doesn't make you immune to the very human processes of cognitive biases - which can cause us to perceive differences that don't exist in the sound waves. I would also expect that your synestheseia would also be prone to be altered by such biases. (We are just as prone to optical illusion as audible illusion - and synesthesia seems to be a mixing of audio and optical processing)

This is why there is so much emphasis here on blind testing when we think we are hearing differences that the measurements suggest are likely to be inaudible.

Again - I'm not stating that you are NOT hearing a difference here. But it is not possible to be sure you ARE if you are doing uncontrolled listening.

Oh, I absolutely agree! No one is free from purely psychological (non peripheral based) artifacts, even the pros can be affected by expectations. Amir said in a review from Neumann KH 80 DSP that he liked the deep of the sound but he subjectively perceived “small as computer speakers sound” based on its little size and not on measurements!

In this case my possible bias has not a huge importance, first because Tidal change its prizes since April and allows full access for 10.99€/month. Surely they cannot resisted the Apple Music competence in price.

By the way, what I compared blind was the difference in EQ between Qobuz, Apple Music and Tidal, and they really sound quite different.

For example Qobuz emphasizes mid highs and highs and sound more clear and forward but lack in bass and rhythm. I found it boring.

Apple Music are more emphasized in low frequencies but lack in highs, I found Tidal the most satisfying and precise, with the exception of MQA files.

Why is this possible? They equalize a little bit their tracks in order to produce their own sound signature or may be due by encoding? This look bizarre as they all utilize loosless algorithms (appart from MQA).
Maybe they use different masters?

 

[antcollinet]

Maybe they use different masters?

This is most likely.

Or perhaps they have different "loudness" standards (relating to dynamic range compression) which can dramatically alter the perceived quality of the music.

In this case my possible bias has not a huge importance, first because Tidal change its prizes since April and allows full access for 10.99€/month.

Cognitive biases operate at the subconcious level. They might have nothing to do with your conscious bias or expectations. You might, for example, hear a difference just because you are listening for one - which creates a subconscious expectation that there is one.

 

[Miguelón]

Cognitive biases operate at the subconcious level. They might have nothing to do with your conscious bias or expectations. You might, for example, hear a difference just because you are listening for one - which creates a subconscious expectation that there is one.

Yep, I know them well. I’m physical therapist and cognitive bias made part of my job related to the patient perception. Also when we do clinical research we should take into account psycho physical interaction and do blind, double blind (patient and performer) and even triple blind (patient, performers and the expert in bio statistics if we have one).

Thank you for your answers!