24bit vs 16bit, the myth exploded!

[stv014]

  It's not possible to record a DC signal onto a discs. What you can record is a blank space of 0 frequency. This is actually quite common on test discs and is used to measure any signal to noise ratio.

 
I can easily render DC, or a 0.1 Hz (or other very low frequency) sine wave to a 44.1 kHz stereo 16-bit PCM format sound file. Is there anything that would make it impossible to burn the data to a CD-R as an audio (Red Book) track ?

 

[Digitalchkn]

Can you give me an example of a disc that has such an audio file on it? I have loads of engineering test and set up discs from CD player manufacturers. But none of them has a signal like you describe. I do have 1Hz as the lowest recorded frequency on a disc. That's as low as it goes.

Take any old waveform editor and deliberately introduce a constant offset to the waveform. Save it in 16bit/44.1K format. There you have it. The fact that you haven't seen a disc with such an offset is I would say is positive as the DC offset effectively limits the dynamic range of the recording (amongst other things.)

 

[silverharbinger]

Not to get too far off the subject, but this thread is an incredible resource. I know it helped me to understand that more isn't really better in audio file formats in a much better way than I was able to piece together over the years. I mean, I could never hear a difference past 16-bit in playback quality, but I just wasn't sure if it mattered. Now I know it seems we simply reached the limits of human hearing years ago, so now we're pushing formats for dogs and bats to enjoy.
 
I wonder when we will finally get there with picture resolutions? Imagine when there's no need for another generation of TV, game system, or whatnot because you won't be able to see any difference? I guess electronics will get really boring.

 

[UltMusicSnob]

   
I can easily render DC, or a 0.1 Hz (or other very low frequency) sine wave to a 44.1 kHz stereo 16-bit PCM format sound file. Is there anything that would make it impossible to burn the data to a CD-R as an audio (Red Book) track ?

An unbroken sequence of zeroes in the samples would be 0 Hz at 0 voltage. How about an unbroken sequence of, say, 0xc7f1's? That would also be 0 Hz, but at some non-zero voltage (what would the physical driver do, shift forward a millimeter and hang there???).

 

[stv014]

  Can you give me an example of a disc that has such an audio file on it?

 
No claim of its commercial availability was made. There might not be such disks sold, but if that is the case, it is not because it would be technically impossible for limitations of the data format, but rather because it would not be practically useful.

 

[stv014]

  what would the physical driver do, shift forward a millimeter and hang there???

 
Yes. A high DC level could damage the driver, though, so it is best to only try it with some cheap speakers.

 

[UltMusicSnob]

  Not to get too far off the subject, but this thread is an incredible resource. I know it helped me to understand that more isn't really better in audio file formats in a much better way than I was able to piece together over the years. I mean, I could never hear a difference past 16-bit in playback quality, but I just wasn't sure if it mattered. Now I know it seems we simply reached the limits of human hearing years ago, so now we're pushing formats for dogs and bats to enjoy.
 
I wonder when we will finally get there with picture resolutions? Imagine when there's no need for another generation of TV, game system, or whatnot because you won't be able to see any difference? I guess electronics will get really boring.

In the future we'll get fitted with bionic cochlear implants that have effective response from 0.001 Hz (elephants, pigeons) to 200 kHz (bats, dolphins). Then we can start having these arguments all over again.

 

[Digitalchkn]

An unbroken sequence of zeroes in the samples would be 0 Hz at 0 voltage. How about an unbroken sequence of, say, 0xc7f1's? That would also be 0 Hz, but at some non-zero voltage (what would the physical driver do, shift forward a millimeter and hang there???).

And that's by definition DC. Your driver (assuming your amplification is also DC couple) would simply recenter itself. You can apply your music on top of that, but now you'd limite the range that the speaker could swing . Hasn't anyone tried to connect a AA battery to your speakers to test continuity before?

 

[UltMusicSnob]

   
Yes. A high DC level could damage the driver, though, so it is best to only try it with some cheap speakers.

It's going to heat up fast, yes? The magnet will be pushing against the physical resistance of the driver material itself continuously.

 

[Digitalchkn]

It's going to heat up fast, yes? The magnet will be pushing against the physical resistance of the driver material itself continuously.

To that effect. You're getting into the topic of constant power vs RMS power. Let's just say for a given peak voltage or /current the RMS power of your music is going to be less than (or equal to) constant power at that peak.

 

[xnor]

  Far from it. There are loads of tunes with audio information below 20Hz.

 
I didn't say there are no CDs with information below 20 Hz, just that you won't find much content (maybe should have written energy) below 20 Hz.
 
 
Quote:

And that's by definition DC. Your driver (assuming your amplification is also DC couple) would simply recenter itself. You can apply your music on top of that, but now you'd limite the range that the speaker could swing . Hasn't anyone tried to connect a AA battery to your speakers to test continuity before?

A DC coupled amp would potentially amplify the DC. Maybe you meant AC coupled? Such an amp (like the O2) doesn't output DC (well in reality every amp outputs a tiny bit of DC...).
With such an amp your headphones are safe even if the source suddenly decides to output DC and the amp has a high gain factor.

 

[silverharbinger]

  In the future we'll get fitted with bionic cochlear implants that have effective response from 0.001 Hz (elephants, pigeons) to 200 kHz (bats, dolphins). Then we can start having these arguments all over again.

 
I had that thought too, and for people with tinnitus and all the other nasty conditions it would be amazing, but I hope it doesn't happen in my lifetime. Having to constantly update my gear *AND* my eyes, ears, tongue, etc.? Plus, just imagine all the aftermarket ear mods that will be offered on these forums? OH GAWD!

 

[xnor]

You better hope those implants have configurable frequency response or else you might go crazy...

 

[silverharbinger]

  You better hope those implants have configurable frequency response or else you might go crazy...

 
There's an app for that.

 

[Digitalchkn]

I didn't say there are no CDs with information below 20 Hz, just that you won't find much content (maybe should have written energy) below 20 Hz.

A DC coupled amp would potentially amplify the DC. Maybe you meant AC coupled? Such an amp (like the O2) doesn't output DC (well in reality every amp outputs a tiny bit of DC...).
With such an amp your headphones are safe even if the source suddenly decides to output DC and the amp has a high gain factor.

I meant DC coupled, like headphone amps you mentioned. It's really same argument with DC coupled headphones amps damaging your headphones.