Audeze LCD2 rev 2 (pre Fazor) Unicorn Edition Measurements

Discussion in 'Headphone Measurements' started by purr1n, Apr 8, 2016.

  1. JohnM

    JohnM Author of REW - Rando

    Joined:
    Dec 13, 2015
    Likes Received:
    134
    Trophy Points:
    33
    To see the levels of "stuff that shouldn't be there" it is probably better to plot the harmonic dB SPL traces, then the absolute levels are on show and there is no %-relative-to-what question to answer. The distinction between what is actually distortion and what is boosted by the shape of the linear response can be important, however - the linear response can be straightforwardly compensated for by linear alteration of the signal, not the case for the distortion components.
     
  2. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Equalization of the signal may also alter distortion components in absolute levels. However, it makes perfect sense to me to use absolute levels when relative measurements may lead to confusion.

    I'll try publishing distortion in absolute SPL for some of the Stax I measured in the past.
     
  3. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Distortion in db plots also make it really easy to tell what's noise and what is actually distortion. The thing is that it also makes comparing two measurements harder. With distortion in % graphs you can easily say "D2 at 300Hz is 1.1%" whereas with the FR and distortion in db plots you often find yourself saying "well, distortion looks pretty low, looks to be about -55db throughout". Easier to be "exact" with the distortion in %, but also easier to be confused by the exact numbers.
    I like the FR and distortion in db plots but the % plots would be pretty useful with very low noise microphones.
     
  4. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    They are both useful. Distortion in dB (and %) is probably more meaningful if you make all your measurements at a particular SPL level, at a particular frequency. For example, if I run an HD600 @ 100 dB @ 1kHz, and compare results when running KSC75s @75 dB @ 1kHz, then likely distortion values on the HD600 may look worse than the ones from the KSC75. Things might be a little more apples to apples if both measurements are done @90 dB @ 1kHz.

    It is not always perfect though. I had a Beats headphone measurement where the cans hit a null around 1 kHz and then things seemed unfair for the Beats headphone.

    An alternative is to use total power or so. But for the most part I do 90 dB @ 1 kHz. I think that's what Tyll does also.
     
  5. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Level matching is of course extremely important and as you said you may get a problem with the varying frequency response curves of different headphones (I still remember when you measured the Beats Studio and ended up using white noise for setting the level). In this case I actually sometimes use my ears: I set a very neutral headphone to a reference level (say 90db at 500Hz or 1kHz) and then try to match the headphones. (I actually remember once doing this with my UERMs and HD800 to see if my listening volume changed with IEMs (because of the ambrose ear lense thing) and the level ended up being within 0.2db between 500Hz and 1kHz.) Also, with bright or peaky headphones you would probably not listen as loud anyway (this is of course no excuse for bad distortion, just that other problems may be worse in this case).
     
  6. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    You can also calibrate with pink noise or other, and using some standard filter the way they calibrate and measure SPL meters. That approach may be more robust.

    However, I did the 90dB @ 1kHz mainly because that's what Tyll seemed to be doing (he also does 100dB @ 1kHz).
     
  7. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    I thought Tyll used 500Hz:

    Headphone Measurement Procedures - Frequency Response
     
  8. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Shit, you are right. Dunno why I thought it was 1kHz all this time. Getting senile.
     
  9. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    BTW John, just wanted to say thanks for all your contributions and awesome work with REW.
     
  10. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,777
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Too much work. There are always better ways to manipulate and present raw data. I'd rather stick with things as they are for the sake of consistency and comparability with other existing measurements. It's not as if what we have now isn't understandable or all that confusing. There is nothing wrong with the "conventional" distortion measurements.
     
    Last edited: Apr 30, 2016
  11. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Agree. I'll keep things as they are.

    I'll add other measurements or stuff if things become unclear or if folks are interested.
     
  12. pila405

    pila405 New

    Joined:
    Aug 15, 2016
    Likes Received:
    1
    Trophy Points:
    3
    May I ask what's the S/N?

    And why everyone seem to hate the fazor LCD-2? From Tyll's measurements they seem pretty decent. Look at this measurement, for example (I assume this is fazor because FR and IR is closer to the fazor measurement rather than 2.2):
    http://www.innerfidelity.com/images/AudezeLCD2SN5325928.pdf

    And the "Unicorn Edition" is because of the low THD? The FR?
     
    Last edited: Aug 15, 2016
  13. Smitty

    Smitty Too good for bad vodka - Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    562
    Trophy Points:
    93
    Location:
    Ca$hburn, NoVA
    The only thing that matters with the LCD-2 discussion is the sound. Fazor models don't sound as good, veiled, somewhat flabby bass, check the main LCD-2 thread for detailed impressions. This unit is the Unicorn because the sound was one of the few things that approached HD800 levels of clarity and detail. The FR is really good, but the CSD plot shows amazing damping, and a very tame ortho wall. That's what sets it apart from all other LCD-2s that have been measured/heard but the SoCal group.
     
  14. pila405

    pila405 New

    Joined:
    Aug 15, 2016
    Likes Received:
    1
    Trophy Points:
    3
    What is the "ortho wall"? And if the sound signature isn't for someone's liking, he can always use equalizer.
    I try to understand what would cause exactly less clarity in a pair as the one I linked to. You can always get more treble, if you wish.
    And what about the 2016 models? have any one measured some of these?
     
  15. Bobcat

    Bobcat Friend

    Pyrate
    Joined:
    Oct 3, 2015
    Likes Received:
    400
    Trophy Points:
    63
    Location:
    Texas
    Home Page:
    Try doing a search on the site?
     
  16. pila405

    pila405 New

    Joined:
    Aug 15, 2016
    Likes Received:
    1
    Trophy Points:
    3
    I did, and from all the questions I raised only one was answered - the "ortho wall"
     
  17. Smitty

    Smitty Too good for bad vodka - Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    562
    Trophy Points:
    93
    Location:
    Ca$hburn, NoVA
  18. pila405

    pila405 New

    Joined:
    Aug 15, 2016
    Likes Received:
    1
    Trophy Points:
    3
    Thanks, but I that's the thing I read. I can't understand why there is so much hate for the LCD-2. Doesn't seem to be logical.
     
  19. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,777
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    The "hate" comes from several things:
    1. Earlier LCD2s had the potential to sound better. Unknown if 2016 models are better, but I wouldn't believe HF. See related item #6 below.
    2. Inconsistent quality with varying transient response. The veiled LCD3 issues upon initial release also affected LCD2s. Units that degraded in transient response over time.
    3. Driver failures. How many failed in my very hands? How many new ones continue to fail?
    4. Ergonomic nightmare.
    5. Quixotic frequency response - sudden dump after 2kHz. Uneven treble, in some cases a treble peak, or three.
    6. Ever increasing pricing for minimal or reverse gains in sound quality. Slower transient response, increased distortion, etc.
    There are two things to understand. The correlation of transient response to the ortho wall frequency. The correlation of clarity to distortion. Not everything is defined by frequency response. Errors in FR are considered linear distortion, which can be relatively easily corrected. The "distortion" aspect, at least the common use of such term, refers to non-linear distortion. We are kinda fucked with non-linear distortion because it can't be fixed.

    An honest and realistic assessment based on at least two dozen LCD units is not hate. If it were, most of us who heard the unicorn edition would have hated it too.
     
    Last edited: Aug 16, 2016
  20. pila405

    pila405 New

    Joined:
    Aug 15, 2016
    Likes Received:
    1
    Trophy Points:
    3
    So all one needs is to be lucky enough when buying a late pre fazor LCD-2.2
     

Share This Page