Analyzer?

Discussion in 'General Audio Discussion' started by Beatnik, Jul 11, 2018.

  1. Beatnik

    Beatnik What's this ?

    Messages:
    4,430
    Location:
    DFW
    Out of curiosity I set everything as close to flat as possible using pink noise and with a 15 band per side eq, it made my ears fatigued. I set it by ear and I can run it it at 105db with no fatigue at all. What am I doing wrong ?

    This is by ear adjusted to the recording, music playing. Is it a crap ap ? Better ap ?

    Screenshot_2018-07-11-20-40-04.png
     

     

    Please register to disable this ad.

  2. mashaffer

    mashaffer Super Member

    Messages:
    3,341
    Location:
    North Central Indiana
    At 105dB you should be fatigued... defense mechanism. :rockon: But seriously at high SPLs your ears (if they are still working) will be rather sensitive to the high frequencies. In addition many recordings will emphasize the extremes for sizzle effect and to accommodate mediocre playback equipment so it doesn't really surprise me. Also consider that if there are some high frequency distortion products especially IMD or high order HD they will likely cause fatigue and a little roll off will help with that.
     
    E-Stat and gdmoore28 like this.
  3. Beatnik

    Beatnik What's this ?

    Messages:
    4,430
    Location:
    DFW
    This is pink noise after adjusted by ear.

    Screenshot_2018-07-11-21-24-15.png
     
  4. Beatnik

    Beatnik What's this ?

    Messages:
    4,430
    Location:
    DFW
    Normally it's not blasting at that level, but sometimes I do like to rock out. (feel it)

    When attempting to set this with pink noise do you shoot for completely flat across the board, or is it going to roll the highs down and boost some bass ? In order to get it close to flat I really had to boost the highs and lower the bass way down ? I know using a phone ap is not ideal, just working with what I have.
     
  5. gadget73

    gadget73 junk junkie Subscriber

    Messages:
    36,447
    Location:
    Southern NJ
    How flat is the response of the microphone you are using? Without knowing that, any results are going to be very suspect.

    Honestly if it sounds right, I would pay no mind to what your smartphone has to say on the matter.
     
    noogies, Bodyblue and Beatnik like this.
  6. turnitdown

    turnitdown Well-worn member Subscriber

    Messages:
    1,870
    Location:
    Tallahassee, FL
    Absolutely flat causes fatigue in most listeners. We (as humans) don't find it the most pleasing -it sounds shrill. A slight roll-off is preferred. Look up the Fletcher-Munson curve. You're probably tuning it, by ear, in accordance.
     
    TudorTurtle and Beatnik like this.

     

    Please register to disable this ad.

  7. ConradH

    ConradH Addicted Member

    Messages:
    7,838
    Location:
    Canandaigua, NY
    Flat is a great target for amplifiers- you can't design without a target of some sort. Flat systems, OTOH, aren't very pleasant. The worst speakers I ever designed were also the flattest. Somewhere, usually via room characteristics or added treatments, the response needs to slope a bit if you're going to be happy long term.
     
    musichal, gdmoore28 and Beatnik like this.
  8. Beatnik

    Beatnik What's this ?

    Messages:
    4,430
    Location:
    DFW
    This is really just an experiment, all the talk of these wonderful sounding flat systems, just toss on a recording and everything is perfect, got me over thinking things again. To my ear and preference I really don't think that could ever happen.
     
  9. drumbum

    drumbum Super Member

    Messages:
    1,191
    Phone apps (using internal mic) are nothing more than a novelty.
     
    Bill Ferris, John James and mashaffer like this.
  10. JoeESP9

    JoeESP9 ESL's & tubes since 83

    Messages:
    16,145
    Location:
    Philadelphia PA
    IME a true flat in room response is almost instantly fatiguing. A gentle roll off above ~12KHz and a slight boost below ~75Hz IMO/E makes for a more pleasant overall sound. The flatness is implemented between those two frequencies.
     
    TudorTurtle, Alobar and gdmoore28 like this.
  11. Lavane

    Lavane AK Subscriber Subscriber

    Messages:
    7,099
    Location:
    northern cal.
    I have a spectrum analyzer I've had for many years. Using the mic and pink sound generator to get an flat response always gave me ear fatigue. As stated in previous posts the roll off on the top end sounds better. That being said, most of my SS integrated amps and pre amps tend to have the tone controls set flat or close to it. Every one and brand seems to have it's own house sound, even with controls set flat or using the tone defeat control. None of my tube equipment has tone controls but I wouldn't call the response flat. Adding an EQ with an analyzer to try and get a flat response for the room never sounded right to my ears.
     
    Bodyblue likes this.

     

    Please register to disable this ad.

  12. Beatnik

    Beatnik What's this ?

    Messages:
    4,430
    Location:
    DFW
    So, when people say "flat" they're just talking midrange and bass/treble to their liking or whatever, no hard set rules there ?
     
  13. lini

    lini just me...

    Messages:
    5,865
    Location:
    Munich, Bavaria
    Beatnik: Did you consider that "flat response" in case of pink noise would actually mean a 3 dB drop per octave? So, provided the display is linear, a flat setting would actually result in frequency response with a 3 dB/octave rise.

    Greetings from Munich!

    Manfred / lini


    edit: I guess the above could easily be misunderstood - what I meant is: Who knows, whether the app actually does a proper sprectrum analysis over 1/3 (or whatever fraction of an) octave bands, in case of which pink noise plus flat response should result in columns of equal height. Whereas in case the columns would merely show the responses at certain frequencies, a flat response should show that 3 dB per octave drop.
     
    Last edited: Jul 12, 2018
  14. Beatnik

    Beatnik What's this ?

    Messages:
    4,430
    Location:
    DFW
    Flat to me suggests a straight line across the board left to right on an analyzer. All frequencies as equal db as possible. Anything else would be a curve of some sort.
     
  15. lini

    lini just me...

    Messages:
    5,865
    Location:
    Munich, Bavaria
    Beatnik: I've just modified my post above to make more clear, what I actually meant to express.

    And on a proper spectrum analyser fed with pink noise it would rather be "all certain fraction of an octave bands as equal dB as possible" than "all frequencies as equal dB as possible".

    Greetings from Munich!

    Manfred / lini
     
  16. Beatnik

    Beatnik What's this ?

    Messages:
    4,430
    Location:
    DFW
    I need a picture of what a "flat response" would look like on a proper analyzer.
     

     

    Please register to disable this ad.

  17. teal'c

    teal'c It's all moo

    Messages:
    15,393
    Location:
    Michigan, N of Detroit
    You're worrying about what's "right" instead of what sounds good.
    Flat is boring. Try driving through a flat state, hard to stay awake.
     
  18. JoeESP9

    JoeESP9 ESL's & tubes since 83

    Messages:
    16,145
    Location:
    Philadelphia PA
    I would hardly call 75Hz to 12KHz the "midrange". Yes many equalize for flat and then apply a "house curve". In my case it's a 6dB/Octave roll off above 12KHz and an upward bass tilt that starts around 70Hz and reaches +3dB at 20Hz. Please note, once the settings are made I no longer adjust the DSP.

    A flat response to 20KHz isn't boring it's annoying. A flat response to 20Hz sounds a little lean.
     
    musichal likes this.
  19. I LIKE MUSIC

    I LIKE MUSIC Super Member

    Some general information.

    Pink noise and white noise are not the same.

    upload_2018-7-12_16-40-17.png i

    If one uses pink noise as their source and then tries to make the display on their analyzer look flat, the actual response of their system will not be flat as previously mentioned.

    White noise would be the broadband noise source to use for that.
     
  20. JoeESP9

    JoeESP9 ESL's & tubes since 83

    Messages:
    16,145
    Location:
    Philadelphia PA
    pink noise
    noun
    Physics
    noun: pink noise
    1. random noise having equal energy per octave, and so having more low-frequency components than white noise.

    white noise

    white noise is a random signal having equal intensity at different frequencies, giving it a constant power spectral density.
     

Share This Page