This discussion is locked
Matthew Devenish

Q: Checking a CDR master for errors using phase cancellation

Hi,

Can anybody shed some light on this conundrum?
Scenario. I have mastered an album in waveburner ( 1.5.2), I bounce the project which produces a waveburner session with one audio file with all the pq codes etc. This file is a 16 bit aiff file. Burn a cdr master from this bounced session. Copy the aiff files from the audio cdr
Take this audio file ( open the waveburner session package, goto audio folder and copy it out), put this audio file into logic 8 ( or another sample accurate daw), onto a stereo track. Now on a second stereo track in logic, lay up the tracks taken from the audio cdr against the bounced file from waveburner, you can zoom in to logic down to samples and make sure the audio is entirely in sync. Put an phase invert plugin on one of the stereo tracks. Theoretically I would have thought there should be 100% cancellation. all music disappears but there is a type of residual noise. If you look closely at the waveforms they are ever so slightly different, almost not visible.
Does anybody have a suitably technical explanation as to why this should be the case?
I am still to do a sine wave sweep check to see what happens with this.

quad g5 ppc, Mac OS X (10.4.11)

Posted on Sep 6, 2010 5:23 PM

Close

Q: Checking a CDR master for errors using phase cancellation

  • All replies
  • Helpful answers

  • by Jazzmaniac,

    Jazzmaniac Jazzmaniac Sep 7, 2010 1:59 AM in response to Matthew Devenish
    Level 2 (493 points)
    Audio
    Sep 7, 2010 1:59 AM in response to Matthew Devenish
    Uhm, dithering? Or maybe just quantization noise if you don't dither?

    Cheers,

    Jazz
  • by Matthew Devenish,

    Matthew Devenish Matthew Devenish Sep 7, 2010 8:54 AM in response to Jazzmaniac
    Level 1 (110 points)
    Sep 7, 2010 8:54 AM in response to Jazzmaniac
    Hi it can't be the dithering as I am comparing an audio cdr burnt with the bounced waveburner session ( dither added in the conversion from 24bit to 16bit). If the bounces were done after the cdr had been burnt, then you would expect this and worse. As far as i understand it, there should be total cancellation occurring. The exact same noise waveform (dither) that was added to the audio is being burnt to cdr. ( not just the same settings!)
  • by Jazzmaniac,

    Jazzmaniac Jazzmaniac Sep 8, 2010 8:26 AM in response to Matthew Devenish
    Level 2 (493 points)
    Audio
    Sep 8, 2010 8:26 AM in response to Matthew Devenish
    I can't follow your explanation of which file you're comparing to the reimported CD audio. Try to explain it again please. In any case: Two different exports/bounces/burns are not identical because dithering actually is random.

    Cheers,

    Jazz
  • by Matthew Devenish,

    Matthew Devenish Matthew Devenish Sep 8, 2010 4:54 PM in response to Jazzmaniac
    Level 1 (110 points)
    Sep 8, 2010 4:54 PM in response to Jazzmaniac
    As I explained, I am comparing the bounced waveburner project audio file, with the imported audio file from cdr. Therefore, theoretically the data should be 100% the same, so a phase invert should totally cancel. As far as I am aware the burn process should not send the data through any phase or level change processes.

    Absolutely! 2 separate bounce and burn processes do produce a much bigger difference file, which is what you would expect, the same as 2 bounces with dither will not totally phase cancel, in fact this is also the case with some reverbs.