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: What do standard redbook players do with 48Khz+24Bits? (Read 6093 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

What do standard redbook players do with 48Khz+24Bits?

I thought that a standard CD player will just puke with this? I accidentally burned some tracks with these settings yet they played in my car CD? Do they just ignore the header info and play at 44/16 (assumedly will play audibly faster?) Thanks!

What do standard redbook players do with 48Khz+24Bits?

Reply #1 –
Quote from: nonsub on
I thought that a standard CD player will just puke with this? I accidentally burned some tracks with these settings yet they played in my car CD? Do they just ignore the header info and play at 44/16 (assumedly will play audibly faster?) Thanks!

Probably your CD burning program converted them to 16/44, because there's no way for a redbook CD to have any other format. If your CD burning program just converted to 16/48 and burned that, they would play slower than normal (and it would be obvious).

What do standard redbook players do with 48Khz+24Bits?

Reply #2 –
Quote
If your CD burning program just converted to 16/48 and burned that, they would play slower than normal
don't confuse what a software program will do using a computer soundcard with what an audio CD player does. 16/48 would produce nothing but very unpleasant noise or complete silence. No audio CD player could make any kind of music out of it.

It is true that many CD-R writing applications will convert other formats to stereo PCM, 16/44, as they write the CD. If you don't start with exactly the correct format, this is the only way you can get a CD that plays in a hardware CD player.

What do standard redbook players do with 48Khz+24Bits?

Reply #3 –
Quote from: AndyH-ha on
16/48 would produce nothing but very unpleasant noise or complete silence. No audio CD player could make any kind of music out of it.

Not so.  bryant's quote is absolutely correct.  If you don't believe it, try it yourself.

What do standard redbook players do with 48Khz+24Bits?

Reply #4 –
You are claiming that an incorrect sample rate written to a CD-R will playback audio, just at a different rate? I admit that isn't one of the errors I've made with any of mine, but it certainly is not consistent with anything I've ever read on the subject.

What do standard redbook players do with 48Khz+24Bits?

Reply #5 –
I'm (we're) claiming that if you burn 48/16 to CD-R the playback will sound slower since your player reconstructs the data at 44.1k samples per second.

Try it yourself.

What do standard redbook players do with 48Khz+24Bits?

Reply #6 –
Quote from: AndyH-ha on
You are claiming that an incorrect sample rate written to a CD-R will playback audio, just at a different rate? I admit that isn't one of the errors I've made with any of mine, but it certainly is not consistent with anything I've ever read on the subject.


Since theres no way for the CD player to know you gave it the wrong sample rate, it'll just play the samples you give it, except the Dt between each sample will be too large, and so it'll take longer to play the file, and the frequencies will be shifted.  Basically, its just a byproduct of how PCM works.

Of course, if you try this with 24 bit audio, your samples won't line up with where the CD player's samples are, so you'll probably get garbage.

What do standard redbook players do with 48Khz+24Bits?

Reply #7 –
Never having heard this before, it seems an interesting proposition of the ‘wile away some idle time' variety, so I decided to. I played with a few different songs on-computer, it being very easy to make the desired manipulations that way, until I found one that would be very obvious.

For the first attempt I definitely achieved something that sounded different, but only because a resampling algorithm is so bad. The writing program converted the file to 44.1k without so much as a by-your-leave.

The next program said, "no way, my father raised me to be better than that." That exhausted my list of on-board writing apps.

I have an older machine with Nero 5 which has produced a few screechers because I forget to resample to 16 bit before writing the CD. The program just wrote "audio" CDs with the data, and my players attempted to play them. Results were never pleasant.

I transferred my test file to this machine and made another audio CD. I printed out Nero's detailed log at the end. The log makes no mentioned of the resampling in any manner I can detect, but resample it did. The quality is better than that other program, but this got me no closer to verifying or disproving the claim.

I hate adding new software I don't really need when I'm not sure how much it will attempt to get its tentacles around, but if you can recommend something in particular as function and safe here, perhaps I will give it another try.

What do standard redbook players do with 48Khz+24Bits?

Reply #8 –
I'd just make a 44.1KHz wave that actually contains 48khz samples.  That way the resampler won't be used.  I doubt much of any software will actually allow you to burn a wave that has the wrong sampling rate knowing.

Do you have matlab handy?  The wavwrite command is a great way to do this.

What do standard redbook players do with 48Khz+24Bits?

Reply #9 –
No matlab. Perhaps I can think about the problem a bit in my spare time.

What do standard redbook players do with 48Khz+24Bits?

Reply #10 –
What is the point of this discussion?

Whether the vector of numbers:
X48 = [x1, x2, x3,...,xn]

maps to a new vector:

X44 = [x1, x2, x3,...,xn]

or is samplerateconverted such that m ~ 44.1/48:

X44 = [x'1, x'2, x'3,...,x'm]

Or if 24-bit words leads to a strange 16-bit patter:

x1=0xAAAAAA
x2=0xBBBBBB
x3=0xCCCCCC

x'1=0xAAAA
x'2=0xAABB
x'3=0xBBBB


Would be totally up to the software doing this operation and any philosophical discussion on what is "right" serves little purpose...

-k

What do standard redbook players do with 48Khz+24Bits?

Reply #11 –
Quote from: AndyH-ha on
Perhaps I can think about the problem a bit in my spare time.


You can burn any data as audio by following the directions:
  • convert to raw (e.g., with "sox.exe input.wav -t raw output.bin" or your favourite audio editor)
  • create a cuesheet file in the same directory as OUTPUT.BIN
    Code: [Select]
    FILE "OUTPUT.BIN" BINARY
      TRACK 01 AUDIO
        INDEX 01 00:00:00
  • burn this cue/bin pair with an appropriate program (Nero, Roxio's EMC, Goldenhawk CDRwin)

By the way, I found this solution when I tried to create a 4ch/22050 redbook Audio CD
 

What do standard redbook players do with 48Khz+24Bits?

Reply #12 –
A WAV file is a string of sample values plus some (separate) control information. The samples are merely level values, they contain no timing or frequency information.

When doing A to D or D to A, the software application normally controls the soundcard clock. For playback, the applications gets the sample rate from the WAV file control information, sets the soundcard clock to the correct oscillation frequency, and proceeds to feed it sample values in time with that clock.

In an audio editor it is a simple, one click, choice to change that WAV control information. One can tell the program that the sample rate is really 48K or 96K or 8K. Instantaneously, in human term, a change is made to a byte or two in the WAV control information. There is no resampling; nothing at all is done to the audio samples.

Now this application, or any other, will play back the audio either fast or slow, according to the "false" sample rate info you've provided. This might be fun to amuse your friends, to listen to chipmunk voices or undersea monsters. I've found it useful on a few occasions when dealing with difficult audio cleanup.

A CD writer doesn't use a clock. It takes the WAV file and makes its tokens and redundancies, and interleaves the data, and does all necessary standard reformatting. It creates a mechanical clock on the disk. This isn't meant to be the precision clock needed for faithful, jitter free playback, but it provides the basic starting mechanism. Basically, the writer lays down a fixed number of samples per linear  measurement along the audio track.

This is an absolute requirement that unconditionally translates to 44,100 samples per second when the disk is spinning at the correct rate. If you manipulate the input and provide false information, the audio CD still ends up with 44,100 samples in the amount of linear space that will be processed in each second. If the original audio was made up of more or fewer than 44,100 samples per second, then, just as with that simple control header change in the audio editor, the 44, 100 samples you get back each second will be either more of fewer than the original audio requires; the music will play faster or slower than it "should."