Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track? (Read 3718 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Hi,

Apologies if this has been nailed to death a billion times in this forum already, but I did a thread search and can't find the answer I'm looking for. How easy is it to tell a difference between 44.1 vs 48 hz sample rate on a 16 bit track? 99% of my library are 44.1.


Thank you.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #1
How easy is it to tell a difference between 44.1 vs 48 hz sample rate on a 16 bit track? 99% of my library are 44.1.


Extraordinarily difficult if not impossible.  They're virtually the same sampling rate, and both are sufficiently high that they span the range of human hearing.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #2
Upsampling should introduce almost very few changes.  Downsampling will theoretically damage the audio more but I doubt you'll be able to detect it just with your ears.

It is worth noting that while your library is stored at 44.1, it is entirely possible that it gets upsampled to 48 when you play it.  Look into your Alsa or Pulse settings to see what resampling is happening behind the scenes.  That's on a computer, much harder to tell what a hardware device is doing.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #3
Upsampling should introduce almost very few changes.  Downsampling will theoretically damage the audio more but I doubt you'll be able to detect it just with your ears.

It is worth noting that while your library is stored at 44.1, it is entirely possible that it gets upsampled to 48 when you play it.  Look into your Alsa or Pulse settings to see what resampling is happening behind the scenes.  That's on a computer, much harder to tell what a hardware device is doing.

What is Pulse? I'm using Windows 7 w/ Objective Dac.


How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #5
Dark_wizzie, http://people.xiph.org/~xiphmont/demo/neil-young.html may be of some interest to you. Basically, the sample rate only determines the upper end of the frequency range that can be captured. At a 48 KHz sample rate, all frequences between 0 and 24 KHz can be represented at full volume, although in practice, the uppermost few KHz of that range is rolled off gently in a "transition band". At a 44.1 KHz sample rate, the top limit is 22.05 KHz.

If you're an adult, you probably can't hear absurdly strong test tones above 18 KHz, and as you age, that number just goes down. It's even less likely you'll notice that high end when it's being drowned out by music that dominates the lower frequencies. In any case you're never hearing above 20 KHz, nor was any of the music you listen to recorded, mixed and mastered on equipment with any regard for content above that limit.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #6
Dark_wizzie, http://people.xiph.org/~xiphmont/demo/neil-young.html may be of some interest to you. Basically, the sample rate only determines the upper end of the frequency range that can be captured. At a 48 KHz sample rate, all frequences between 0 and 24 KHz can be represented at full volume, although in practice, the uppermost few KHz of that range is rolled off gently in a "transition band". At a 44.1 KHz sample rate, the top limit is 22.05 KHz.

If you're an adult, you probably can't hear absurdly strong test tones above 18 KHz, and as you age, that number just goes down. It's even less likely you'll notice that high end when it's being drowned out by music that dominates the lower frequencies. In any case you're never hearing above 20 KHz, nor was any of the music you listen to recorded, mixed and mastered on equipment with any regard for content above that limit.

Thanks. I'm 20 and ever since I was a little kid I didn't do so well with hearing tests for high frequencies. I can't really hear above 16, 17khz.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #7
What is Pulse? I'm using Windows 7 w/ Objective Dac.


I assume some Linux thing, Alsa definitely is. I don't why the poster assumed you'd be on Linux.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #8
Does using 48kHz instead of 44.1kHz for lower bitrate (128kbps-196kbps) mp3's result hearable difference (when no SRC involved)?

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #9
99% of my library are 44.1.


Leave it like that. That is, if you have one of the old quirky soundcards that couldn't really deal with anything but 48, then ... then you are most likely better off replacing the soundcard.

On a general note, upsampling will not add any information that wasn't already there. Downsampling will remove information, so then - generally speaking - one should ask whether the "to" format is good enough for the purpose (e.g. do not resample to 16k and expect it to sound the same).

Do not confuse "upsampling" with "oversampling". Oversampling is a technique used in DACs for a couple of purposes, the most important (at least when it was introduced) was to keep the noise from the DAC so far out of the audible range that one would not need steeper filters on the analogue signal.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #10
They're virtually the same sampling rate, and both are sufficiently high that they span the range of human hearing.


Just as a meta-discussion:

The latter is well-known, and that is also the point that is disputed by the selfproclaimed golden ears that are so cocksure they hear more than the rest of us, that they insist they need not test it.

The "virtually the same" part is well-known only among those who can calculate logarithms I think. The coders around here of course know this, but for anyone else: From 22.05 to 24 there are only three quartertones. That is, on a piano, one-and-a-half the distance from a tangent to the next (black or white).
The 20 kHz bar is about a d#, six octaves and a little above the middle-c (if I have punched the calculator right!). 22050 is nearly the f above that d#. 24000 is slightly above the f#. I.e. if you were able to actually hear the 24 mark, the 22050 would cut off only three quartertones - to the not-so-math-savvy, google can calculate the following for you: (log2(2400)-log2(2205))*12, the "*12" because there are twelve halftones on an octave.

But "two kiloherz!" sounds way more than three quartertones, until one is fluent in logs.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #11
I.e. if you were able to actually hear the 24 mark, the 22050 would cut off only three quartertones

What you are forgetting is that those 2kHz can contain quite a few overtones. Sure, music works in logarithms, but that does not mean that ears in general do too. I therefore think your explanation is a bit oversimplified. If I remember correctly, because of the structure of the basilar membrane the 'granularity' of frequency hearing is higher for higher frequencies. Can't find a reference now though.
Music: sounds arranged such that they construct feelings.

How easy is it to discern between 44.1 vs. 48 kHz for a 16-bit track?

Reply #12
Sure, music works in logarithms, but that does not mean that ears in general do too.
They do, to a rough approximation, in both frequency and intensity.

Cheers,
David.