D/A and A/D | Digital Show and Tell (Monty Montgomery @ xiph.org)

Discussion in 'Education' started by Alpha0ne, Nov 27, 2015.

  1. Alpha0ne

    Alpha0ne Producer

    Joined:
    Jan 26, 2012
    Messages:
    671
    Likes Received:
    142
    Post on Tom Cosm's Facebook - page. Just want to share with the community.

    "Why analog audio within the limits of human hearing (20hz - 20 kHz) can be reproduced with PERFECT fidelity using a 44.1 kHz 16 bit DIGITAL signal." - Tom Cosm


     
    • Like Like x 3
    • Interesting Interesting x 1
    • List
  2.  
  3. Alpha0ne

    Alpha0ne Producer

    Joined:
    Jan 26, 2012
    Messages:
    671
    Likes Received:
    142
    Just only one person read this ?? lol.
     
  4. Rhodes

    Rhodes Audiosexual

    Joined:
    Feb 4, 2015
    Messages:
    928
    Likes Received:
    558
    Very informative.
    Thanks a lot!
    I enjoyed that 1/2 hour :chilling:
     
    • Like Like x 1
    • Agree Agree x 1
    • List
  5. smoothripple

    smoothripple Kapellmeister

    Joined:
    Dec 9, 2013
    Messages:
    102
    Likes Received:
    68
    Location:
    Northen Continental Divide, N. America
    HP-3625 Signal Generator & HP-3585 Spectrum Analyzer. Cool gear. High dollar stuff once. Now owned by hobbyists & Ham Radio home-brewers.
    Sure do wish I had them on my bench. :winker:

    Monty Montgomery presents well. I think I'll seek more of his presentations.

    But... all he really said was that the D/A conversion RESULTED in a fairly pure waveform, not that the exact same information was transferred. For most of us the information transfer is what counts. The information that would have been contained in the spaces between each sample point is essentially lost, and in no way recoverable on down the signal chain. [because it is just not there] He says that digital bit-depth does not matter. I disagree.

    Possibly it is because his perspective is from the mind of an engineer, concerned mostly with the physics of a given scrutiny, rather than that of an artist.

    All that said, why not sample a signal multiple times simultaneously [obviously the more the better] but shift time in micro-second increments in the A/D converter so as many gaps as feasible are covered. This would require a companion D/A converter that would recombine the segments in a similar fashion to give a close approximation of the original signal. Somewhat like spread-spectrum but in a time framework rather than a relative position framework. This would be somewhat like what 24/96 does compared to 16/44.1 but it could do it in an infinitely precise manner. I predict that this tech will be in production very soon. It was always limited by hardware's capabilities.
    [imagine nitrogen-cooled A/D & D/A converters :woot:]

    Disclaimer:
    I have no excuse for this post as I am NOT drunk nor superbly herbed.
     
    • Like Like x 2
    • Funny Funny x 1
    • List
Loading...
Similar Threads - Digital Show Tell Forum Date
Selling E-mu and Digital Sound Factory samples Selling / Buying Yesterday at 3:41 PM
NEW: Schwabe Digital - Orange Clip Software News Mar 21, 2024
Getting to the bottom of the high end ITB digital eq argument.. Mixing and Mastering Mar 20, 2024
Why Modern Digital Synthesis Is More Analog Than Analog Software Mar 2, 2024
Any Digital DJ Tips viewers here? Anyone else with same thoughts on recent content? general discussion Feb 13, 2024
Loading...