Are high sample rates making your music sound worse?

Since no one >10 can hear over 20kHz, there is no point to listening to something like 192kHz in the first place, if your amp can even properly playback such high frequencies, all you are doing is increasing the chances of driving your tweeter into distortion.

@goofyfoot  
 
To put it blunty, yes. Everything I found shoes that if people don’t know, they don’t even get up to 70% accuracy (picking CD or better over 320Kbps MP3), it’s usually 40%-60%.  
  
320Kbps was chosen for a reason, as that’s what they found was good enough.

@goofyfoot

It’s no different for vinyl unless you want the possible distortion that may arise from driving your tweeters higher than they can handle, same distortion as would be for playing the actual vinyl. Not saying all/most vinyl does this, but some do. Disregarding this caveat, it makes no difference.

@lalitk
@tatyana69

Find the error in my logic:

Sampling rate: Due to Nyquist, as long as we are 2x more than what we want to capture, it captures it 100% identically. Impossible for transient response for instance to be better with higher sampling rates, as that by definition means the 44.1 version didn’t sample it 100% identically. One caveat would be how good the filter in your DAC is, but even most cheap DACs don’t attenuate below 19kHz (and especially not more than 0.1dB), and since I doubt you can hear that high, it’s all good.

Bit-depth: All this is means is how large of dynamic range you can have, it describes the noise floor. 16Bit (undithered) has an average noise floor of better than 96dB, dithered brings it up to 110-120dB, 24-Bit is 144dB. Your average treated room has a noise floor of no lower than 30dBC, so that’s limiting you to less than 16Bit anyway. 

Also, the people I’ve heard from who are in the industry, all agree that digitized vinyl sounds identical. You finding this different either means you listened to garbage conversions or your brain is telling you what you want to hear, which since you hear a benefit from going to 192kHz leads me to think it’s the latter. Open to discussion, so please find the error in my logic.

@goofyfoot  
 
Just saying your findings differ from most human trials I’ve seen. 

@cleeds

Well, I said that if you load up the high-rez digital file and subtracted the differences between a 16/44.1, you’ll see the differences. You replied that I think listening doesn’t matter. As I’ve said, what you hear is the most important, measurements don’t mean anything compared to what you hear, measurements just are able to tell us wether you are hearing things that actually exist. If I played the same digital file to 100 people and then played the same file again and asked if they heard a difference, how many do you think will say yes?

@geoffkait

Maybe if you bought a CD player at a dollar store. A $75 Blu-Ray player will output the same bits as a $10,000 CD player, only differences are jitter, which even a “cheap” DAC will reduce to below -100dBFS. 
 
Not sure if you know, you can literally drill a hole in a CD and it will not change anything. CD has a lot of protection against faults.

@lalitk

Measurements are far more accurate and far more reliable than our ears. Stating your ears are better and thus that’s why you hear a difference is such a typical response.

I’d like one explanation for why 192kHz is better than 44.1kHz, let’s say using a Chord Qutest as it’s filter is pretty much the best I’ve seen (well, let’s say the Chord Dave to get ahead of comments saying the Qutest is too cheap to be transparent).

Sampling rate only deals with how high of a frequency you want to capture, there is 0% difference within frequencies captured from a lower sampling rate, this is all proven by Nyquist, it’s a 100% capture, or else digital music wouldn’t Moro in the first place, as every sample relates to one specific waveform.

I’m willing to bet you can’t hear above 18kHz (if you are over 50, probably 12kHz), so stating that the inclusion of frequencies at 80kHz makes a difference is just nonsensical (unless you have an explanation).

I’ve asked why you think/know it makes a difference and all I got was “my ears are better”. You don’t even need measurement gear, all you need is a program that loads the two digital files and shows you the difference between them.

@goofyfoot 
 
Telling them anything about what the files are, and the fact that you know which file is being played invalidates any findings from a scientific point.  If another person did the same and the people said it was no different, what then?  
 
As I’ve asked, what benefits does 192kHz have over 44.1kHz? Do you believe we can hear higher than 22kHz or that those inaudible frequencies still influence our hearing?

@geoffkait  
 
Not sure if you were joking, but to cover bases: 
https://www.snopes.com/fact-check/bewaring-of-the-green/

@cleeds

So now you are saying even if the bits are the same it sounds different?

I have listened, didn’t hear a difference. Whose correct, me or you? Why? Because your ears are better or your system is better?

@tatyana69

I highly doubt it has the exact same measurements, meaning a full suite, not just frequency reponse and wattage. Even a high end brand like Mark Levinson cannot make the exact same amp, they all have minuscule deviations.

I listen, and I’ve never heard a difference with going to 24Bit or changing speaker wires (well, I admit nothing fancy like $5000 cables). Now, if you hear a difference, whose right? Well, since simple math and measurements shows that we shouldn’t hear a difference, I would argue I am.

All I get are responses like yours, and never any actual responses to my questions. If you think 192kHz is audibly different to 44.1kHz, I’d think you to explain why, when I bet you can’t even hear up to 19kHz.

@cleeds

Again, so are you saying when the audio in the digital file is the same they still sound different? As that’s what I’m talking about, compare the audio <20kHz for a 44.1 and a 192 file, and no differences will exist above -100dBFS.

@geoffkait

If someone says the midrange is more pronounced with new speaker wire, where an SPL meter or a measurement mic shows no volume differences, whose right?

Just how headphone burn-in isn’t real, a $500 USB cable won’t sound better than a $20 one (well, I’ll be careful, the sound coming out using the $500 cable won’t be different than the $20 one).

@tatyana69

Thats a really poor analogy. Of course it won’t sound the same, as you are changing the physical environment, so you now are dealing with altered reflections/reverberations/echo. 
 

@itsjustme

These make really critical mis-assumptions about what is going on in up sampling.
First, go learn about up-sampling and interpolation filters. Then learn about reconstruction filters and their issues. Then think about how much better you make things if you first interpolate and then feed it to the reconstruction filter. Lots of analog issues get much easier.
It need not change the original data one bit (both literally and figuratively)

Look at the measurements for any non-MQA compatible DAC, all their filters shouldn’t attenuate any frequencies you can hear, especially from companies such as Chord, and any phase errors are going to be in the treble where it won’t be audible on any decent DAC.

Also, if you are upsampling to a non-multiple (44.1 to 192) it no longer is bit-perfect, and the rounding errors would need to be minimized by the DAC/software.

I’ve yet to see any actual benefit to upsampling when disregarding any limitations by the DAC used.

@itsjustme 
 
I know the Chord upsamples, so does the Benchmark, and so do many others, I’m specifically referring to upsampling before the DAC. DACs do this as it helps with jitter, doing this before the DAC doesn’t help.

@itsjustme

Using a multiple to upsample is an easier job and hard to mess up. Using a non-multiple helps with jitter but is harder to implement.

As for up/over sampling before the DAC, it matters whether either side takes care of intersample overs, the Benchmark models for instance does take this into account and lowers the level before oversampling.

Interpolation does “destroy” the original data while synchronous upsampling preserves the original data. It matters how good the interpolation is. Going with your currency analogy, it would be like if I had a $100 USD but someone took it and gave me $100 in Bahamian currency, which averages a 1:1 conversion with USD, but sometimes it’s not, it has been worth 0.98:1 and sometimes 1.02:1. 

Stereophile’s measurements of DACs shows how much aliasing occurs by using a high frequency test tone, if you look at the AudioQuest DragonFly Red’s measurements for instance, the aliasing image at 25kHz is below -100dB.

I can’t comment on phase though as I haven’t seen that tested before, but we are very forgiving of phase errors (not talking phase mismatch) in the treble that I doubt any decent DAC will have an issue.

@lvrooman54 
 

the resulting dynamic range is usually better than CDs, even though the source is from LPs.

 
Vinyl is reported to have a noise floor of roughly -65dBFS, so I don’t see how any ADC or conversion technique can make it better than CD at -96dBFS, unless you are adding noise-shaped dither, in which case the same can be done for 16/44.1.

@rexp

CD already offers us better quality than we can hear in a room, so that argument doesn’t hold. 
 
I also only skimmed the article, but I didn’t see any mention of ADC/DAC, it’s merely saying your hardware likely can’t handle the higher sampling rate, the Benchmark DAC3 for instance acruslly downsamples 192kHz audio to 110kHz, as the chips used have poorer performance at 192kHz, and not that you’d hear the difference  of 110 vs 192.