CD audio - a general question

That is a VERY Good question. I can't wait to read a good response

Yeah. I think it's a redbook spec Larry that it has to be a stereo track, so the burner will take the mono file and basically make a two track mono file out of it just by copying the mono file and placing one copy on each side of the stereo field. You should be able to just do as Slugbaby suggested, ripping the CD, and then taking eaither side of the stereo file, discarding the other, and have the same mono file you started with before burning to CD.

BTW, did you get any of our ise storm up there last week? Man it's been terrible here. Folks are just now getting back to work and there's still a few people without power. I heard so many tree branches cracking under the weight of all that ice one night that I thought we were having a thunderstorm.

If you want to know for sure, then test it using your own setup. Shouldn't take more than about 40 minutes or so. For me, I'd probably be more confident with those results than be reading it in a forum.

1. Take a mono audio file, burn it to CD as CDDA, rip it to a stereo file.
2a. Use Sonar to export the stereo file as split mono (each should be identical, use phase inversion to do a null test)
2b. use Sonar to export the stereo file into a mono format (which should sum L and R channels)
3. Compare the new mono files to the original file using a null test

My guess is that one of these new mono files will have a higher level than the original (in particular the stereo summed results since you are a mixing a signal with an exact copy of itself).

Also, digital audio extraction (CD ripping) is not an exact process. You can never be absolutely sure that what you ripped is an exact copy of what was burned unless you have the original file used to create the CD to compare it to. The results of DAE can only be described using statistical confidence which is increased by reading, re-reading, re-reading, ... the same sector during extraction to verify that results match every time and the ability of the hardware/software to piece together data from overlapping reads. Although with today's knowledge and technology I'd say you can probably be 99.5% sure or more of the data being the same with most setups. You can google "digital audio extraction" and Principles of Digital Audio, by Pohlmann (2000) has some good info.

In terms of maintaining integrity, the only problems I can foresee are bad extraction or somehow altering audio level. Everything is staying in the digital realm and there should be no real sample-altering processing except extraction and the summing in step 2b above assuming correctly written software.

Larry,

I can give you a definitive answer regarding this, but it will have to be later on in the day or sometime tonight.

Fwiw..... I have auditioned the vocal plugin in Sonar 5 and thought it was very over-rated. Hopefully, it is better in S6. I would personally stick with Melodyne or something that will not induce phasiness or phase issues.

ORIGINAL: krizrox

I admit to being a little confused by this. Let's say I have a mono WAV file and I burn it to a CD-R as CD audio (not a data CD). Am I to understand that my mono WAV file becomes stereo once it is burned to the CD-R (I assume this is a Red Book specification). Now let's say I want to rip that file back to my computer. It is ripped as a stereo WAV file right? Is getting it back to mono simply a matter of the ripping process or do you need to use something like... Sonar and bounce to a mono track? I guess my question is how do you maintain mono integrity?

How does this work Mr. Wizard?

Red Book standard CD audio specifies two channels PCM @44.1/16, so a properly recorded mono CD is actually "dual mono," with two identical channels. The L and R tracks are supposed to be identical.

If the CD was properly authored (and was truly a mono source to begin with), and your ripping software is doing its job correctly, you should be ripping two identical WAV files (I presume you're ripping to WAV?) as channels L and R. If you bounce them to mono in a program like Sonar, it should be essentially identical to taking either the L or R channel and simply adjusting the level to compensate for any pan law (minus any rounding errors, dither, or other minor artifacts from the render process).

Now, when you talk about "the ripping process," vs "bouncing to a mono track," the devil is always of course in the details. I don't know what the issue was with your client's tracks, but it shouldn't have been caused by the fact of of a properly recorded CD of monaural audio being sensibly ripped and/or rendered to mono, if that makes any sense.

Of course, if there are alignment issues, or if the two L and R channels were summed to a new mono level that caused clipping, or if the CD wasn't truly mono, or if the Zoom device uses some compressed file format that was then rendered to CD as PCM with artifacts from data compression, or if the ripping software mishandled a stereo PCM to mono WAV conversion, or if some other digital gremlin was working against you, then anything goes.

Hope that helps.

Cheers.

Yep is correct the Red Book spec does not support mono files at all. The data must be 44.1 / 16bit stereo (two channel)