Why SINAD doesn't matter

Discussion in 'Blind Testing and Psychoacoustics' started by purr1n, Aug 28, 2021.

  1. Vtory

    Vtory Audiophileâ„¢

    Pyrate MZR
    Joined:
    Sep 12, 2016
    Likes Received:
    10,831
    Trophy Points:
    113
    Location:
    East Coast
    Just wondering. If I understood correctly, your results and Marv's chart are about loopback performance rather than pure ADC capability only. Right? Curious about each signal generator's contribution.
     
  2. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    Analyzer generator performance is measured by connecting generator output to analyzer input, which we call loopback. Important to note that many analyzers have ability to loopback internally without any external connections. This internal loopback potentially results in better performance than real world with external cables. Thus my measurements used external loopback with real cables.
     
  3. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    To add greater perspective consider the Fletcher-Munson curves which have evolved with further empirically collected data into the ISO contours of equal loudness.
    iso226-2003 Equal Loudness Contours v2 - annotated - cropped.png
    The lower boundary of 20 dB SPL is a limit for residential listening spaces unless careful attention has been paid to controlling ambient noise sources. 40 dB SPL (not dBA) is more typical.

    Upper boundary is set at 100 dB SPL for very short listening sessions unless wishing to need hearing aids earlier in life.

    This makes for an 80 dB dynamic range. Allowing 10 dB S/N for the lower boundary and 10 dB headroom at the upper boundary, a 100 dB dynamic range would seem a practical limit above which little more is to be meaningfully gained.

    Headphone amplifier tests at
    0, -10 and -20 dBu
    0.775 Vrms, 0.245 Vrms, 0.075 Vrms
    100, 90 and 80 dB SPL respectively
    are much more useful than either 2 or 4 Vrms when many headphones have sensitivities around 97 dB / mW.

    Further, distortion sweeps from 20 to 10 KHz expand the information about an amplifier's performance enormously and are part of my standard test suite which may be seen in my technical measurements.
     
    • Like Like x 2
    • Epic Epic x 2
    • List
    Last edited: Sep 20, 2021
  4. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    AP is probably being conservative. I know that there exists variation with each unit. Some measure better than others. Amir's and mine actually measure particularly well. It's just the nature of these things since -120db down really is so insane. 120db is 1,000,000 to 1.

    Here's 500mVrms, SE loopback, 100kohms, using the high-performance analog generator (yes, analog)
    upload_2021-9-20_18-27-11.png

    With the cheat-mode (performance analyzer) turned on along with the analog generator
    upload_2021-9-20_18-28-45.png

    This is the built-in DAC (not analog) generator
    upload_2021-9-20_18-30-18.png

    Note how the skirt of the lower-distortion of the analog generator is wider and that the digital generator doesn't even really have a skirt. The cheat mode only works with a single sine stimulus. Multi-tone is not supported. Comes in handy for e-peening. Tradeoffs.
     
    Last edited: Sep 20, 2021
  5. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I could not resist:

    The area marked in orange is what typically is required for ASR to give a product a recommendation. This is assuming someone listens to music where 0dbFS output equates to 100db SPL.
    judge.jpg
     
    • Like Like x 5
    • Epic Epic x 2
    • List
  6. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    I think you've been too generous with the judgement area. Shouldn't that be a vertical line at 1 KHz from 0 to -30 dB SPL?
     
    • Like Like x 4
    • Epic Epic x 1
    • List

Share This Page