DAC Measurements Posting Standards

Discussion in 'Measurement Setups, Systems, and Standards' started by purr1n, Sep 26, 2015.

  1. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Posts on DAC measurements should adhere to the following guidelines. (Note: Atomicbob can do whatever he wants on the dScope)

    Basic Information
    1. Recommended: Include picture of amp
    2. Recommended: Description of switches, options
    3. Recommended: Picture of inputs and outputs
    Baseline Measurements
    1. Required: 1kHz spectrum -1dBR (or full scale) 44.1/48k sampling, linear 0-24k
    2. Required:1kHz spectrum -90 dBR 44.1/48ksampling, linear 0-24k
    3. Recommended: 1kHz spectrum -60 dBR 44.1/48ksampling, linear 0-24k
    4. Required:1kHz spectrum -1dBr 44.1/48ksampling, linear 0-1.1k
    5. Required:19k + 20k (1:1 -1dbR) -1dBr 44.1/48k sampling, linear 0-24k
    6. Required:19.1kHz -1dBr 192k sampling, linear 0-85k (44.1k playback)
    7. Recommended: PN white noise. Wide band view to note extension and nature of filter slope. -4 dBR 192k sampling, linear 0-85k (44.1k playback)
    Optional Measurements
    1. 700 and 1700Hz two tone -1dBr 48k sampling, log 0-24k
    2. Impulse response
    3. 1khz -90dbfs waveform 16 bits make sure software is not dithering
    4. 1khz -90dbfs waveform 24 bits
    5. Jitter JTEST 44.1k. Please describe output (Windows settings, software used, PCM, nature of signal, sampling rate, etc.) and audio input settings.
     
    Last edited: Nov 4, 2020
  2. firev1

    firev1 Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    44
    Trophy Points:
    18
    Location:
    Singapore
    Why not 0dbFS? some filters don't take to full scale well so its nice to expose it. Also, that would make it more comparable to John Atkinson's tests. Make -1dbFS optional.
     
  3. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    More likely to clip the D/A chip's internal. Can actually vary from unit to unit based on tolerances. IMO if an audio DAC is being told to play back full-scale signals it's not the hardware's fault
     
  4. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Also ran into some issues specifically with Windows DirectSound. ARTA and some other siggen programs don't give us exclusive mode or WASPI. I didn't want to be a position where I said something wasn't right when it was right. DAC measurements make me nervous.
     
  5. firev1

    firev1 Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    44
    Trophy Points:
    18
    Location:
    Singapore
    Seems fair enough. Calibrating the ADC side of things can be a real bitch sometimes. In real world

    Usually for generation I would use foobar with signals I have already saved. The signals are inspected on FFT with using Wavespectra(allows you to look at the 'perfect' signal without playback). Been doing so since that issue with ARTA and 24bit or 16bit files.

    Of course generation and acquisition are done on separate systems so I can avoid the problems where Windows/whatever can really hate on you doing both on one machine.

    Just realised that 6 and 7 maybe interpreted incorrectly,

    For both 19.1khz tone and PN white noise
    playback 44.1Ksps ADC sampling 192Ksps
     
    Last edited: Sep 28, 2015

Share This Page