MiniDSP EARs measurement rig

Discussion in 'Measurement Techniques Discussion' started by dBel84, Nov 9, 2017.

Thread Status:
Not open for further replies.
  1. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    66,073
    Dislikes Received:
    108
    Trophy Points:
    113
    Location:
    Antarctica
    @Daveheart sent me his miniDSP EARS for calibration. Here are the results for the HD58X. (Running a calibration file for one channel of my miniDSP EARS.

    GRN = Dave's EARS L
    RED = Dave's EARS R
    BLU = Marv's EARS L
    ORA = Marv's EARS R
    mDSP.png
     
    ultrabike, monacelli and spoony like this.
  2. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    66,073
    Dislikes Received:
    108
    Trophy Points:
    113
    Location:
    Antarctica
    Uncompensated.

    GRN = Dave's EARS L
    RED = Dave's EARS R
    BLU = Marv's EARS L
    ORA = Marv's EARS R
    EARS HD58X.png
     
  3. Eggmansr71

    Eggmansr71 Rando

    Joined:
    Nov 6, 2017
    Likes Received:
    0
    Dislikes Received:
    1
    Trophy Points:
    1
    Location:
    Benton AR
    So a couple of questions for the thread. So the EARS are accurate up to what frequency? It seems they have an issue around 4.5k so maybe 3.5k. I’m looking to use eq on my headphones so I need to understand where I can trust the measurements and not. I also notice and see the results here a dip and then a spike at 8k. My EARS seem to match Dave’s EARS L shown above.

    The one situation it has helped for me is the Focal Clears. They have a nice rise of 2db around 1.4k that causes issues with some songs. I have seen this in most measurements shown on the web. So it allowed me to pull that down to a good level. A perfect song to hear this is Ronnie Dunn – Cost of Livin. His voice shouldn’t cause issue in the song yet is a touch to strong.

    A separate question - I also wonder why in looking at a lot of headphone graphs the goal isn’t a straight line. I know in loudspeakers that is the goal, so I’ve been wondering why we don’t try to make the measurements match that. It would make it much easier to eq and for me to read. Thinking of innerfidelity here.
     
  4. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    66,073
    Dislikes Received:
    108
    Trophy Points:
    113
    Location:
    Antarctica
    With proper compensation to account for production variances, the EARS appears so far to be as accurate as anything else out there for the entire range of human hearing.

    The spikes and dips specific to the EARS is different from that of the GRAS, but the behaviors seem linear - that is correctable with simple FR compensation.

    Your last question is a good one.

    One answer involves mulitple factors: headphone measurements are in their infancy, different measurement systems produce different results, compensations and targets are all over the place (Sean Olive did not help here with his market research target).

    @Serious did some excellent work in providing compensation curves for Tyll's measurements where the target is a straight line. I forget the link but it's somewhere on this site.

    I attempted to do the same with the preliminary EARS compensation I proposed in this thread. I still need more data though.

    The V2 flat coupler measurements expect a target with a slight downward slope and maybe a slight depression around 2-3kHz depending upon the headphone.

    Finally, even the "flat is neutral speaker" target only applies to a specific case: a nearfield on-axis measurement in an anechoic chamber.

    In reality, it's probably better to measure a speaker in its intended room at the listening position and expect a 1db per octave roll-off (or B+K curve) to be able to say it sounds neutral.

    It comes down to how we measure, what compensation is applied, what is the target, and what that target means.
     
  5. q2klepto

    q2klepto Rando

    Joined:
    Mar 16, 2016
    Likes Received:
    1
    Dislikes Received:
    0
    Trophy Points:
    3
    @Marvey - would you need us to send your our own EAR rigs for calibration - or can we simply use take our supplied calibration files, subtract your supplied calibration files (so now we have the difference between our mics) and add on your custom calibration files

    Or something along those lines.

    For example my Left calibration has this for 4khz:

    Freq SPL Phase
    4000.0 22.1 67.5

    And your supplied cal might have

    Freq SPL Phase
    4000.0 20.1 67.5

    Would this mean that your L mic needs less compensation (of 2db) than my L mic? And if i were to use your curves i would use the difference accordingly?

    Your GRAS compensation @4khz for the HD600 might be SPL 23.5, which would mean that i would add +2db or 25.5db.

    Or am i completely wrong on this.

    I'm in Canada so shipping my EARS rig to you wouldnt be super cheap heh.
     
  6. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    66,073
    Dislikes Received:
    108
    Trophy Points:
    113
    Location:
    Antarctica
    In theory, that would work. However the supplied calibration for my EARs was off. @Daveheart's EARS looked promising though. I think something might have gotten knocked loose on my EARS during shipping.

    I'll probably ask miniDSP for another unit just or be sure.
     
  7. q2klepto

    q2klepto Rando

    Joined:
    Mar 16, 2016
    Likes Received:
    1
    Dislikes Received:
    0
    Trophy Points:
    3
    Here are measurements of my THX00 stock pads - wondering if anyone else is getting these giant dips at around 3.5khz

    Plus not sure how you guys have your waterfall graphs go from 1-5ms ..mine goes to 300ms

    EDIT: Thanks @LSW updated my CSD to be more inline with headphone measurements. Will have to run it again @100db for better CSD

    WIll add my HD800SD, HD6XX and ZMF Auteur charts here or in the main measurements thread.

    Maybe ill play around with it being flush as well

    [​IMG]

    [​IMG]
     
    Last edited: Feb 9, 2018
    LSW likes this.
  8. LSW

    LSW Facebook Friend

    Contributor
    Joined:
    Jun 17, 2017
    Likes Received:
    102
    Dislikes Received:
    0
    Trophy Points:
    43
    Location:
    USA
    I would recommend you read this whole thread , especially the last 3-4 pages.

    Also this guide by @philipmorgan will get you started. Hope this helps.
     
    q2klepto likes this.
  9. zhouharvey

    zhouharvey PayPal’s $2 Sensible Enforcer

    Banned
    Joined:
    Feb 19, 2016
    Likes Received:
    19
    Dislikes Received:
    57
    Trophy Points:
    18
    May I ask if the microphone could be mounted from the bottom of the main body (does it have the thread opening for mounting it to a tripod)? I want to use it alternatively as a binaural microphone.

    Thanks!
    Harvey
     
  10. MF_Kitten

    MF_Kitten Banned per own request

    Banned
    Joined:
    Jan 1, 2016
    Likes Received:
    179
    Dislikes Received:
    28
    Trophy Points:
    43
    you can just drill a hole in the baseplate. I do wish it had that already too though.
     
    zhouharvey likes this.
  11. zhouharvey

    zhouharvey PayPal’s $2 Sensible Enforcer

    Banned
    Joined:
    Feb 19, 2016
    Likes Received:
    19
    Dislikes Received:
    57
    Trophy Points:
    18
    That's a bummer. I don't want to drill a hole myself as it might affect the performance a bit if the body were hollow. I have already sent the company an email making this suggestion and hope they could improve in future.
     
  12. LSW

    LSW Facebook Friend

    Contributor
    Joined:
    Jun 17, 2017
    Likes Received:
    102
    Dislikes Received:
    0
    Trophy Points:
    43
    Location:
    USA

    Looks good. Also read as much you can by @Hands and @Marvey on the measurement forums. They have a ton of useful information about CSD's, how to present them, and how to interpret them.

    Here is a ss of my CSD settings from just trying to copy the guys on here that really know what they're doing.
    [​IMG]

    I am considering an ears rig myself but have the usual early adopter wet feet. I should probably let the intelligent people who know what they're doing develop a sound methodology for it before I put my fat doofus hands on one.
     
  13. Hands

    Hands Overzealous Auto Flusher - Measurbator

    Staff Member Friend MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    11,212
    Dislikes Received:
    40
    Trophy Points:
    113
    Location:
    Colorado
    Home Page:
    99% of me "knowing" what I'm doing is emulating guys like Marv and Tyll. Trial and error until it looks right.

    Then again, that's how I figured out how to push the right computery buttons and be a good computery nerd.

    I'm still an idiot. :)
     
    AllanMarcus, Elnrik and LSW like this.
  14. samvafaei

    samvafaei Rando

    Joined:
    Jul 15, 2016
    Likes Received:
    34
    Dislikes Received:
    0
    Trophy Points:
    13
    We just received ours. We're going to do some tests to see how it performs compared to the HMS. Let me know if there's something specific you guys want to know.
     
  15. spoony

    spoony Spooky

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    370
    Dislikes Received:
    3
    Trophy Points:
    63
    Can it hold a paper roll in its mouth?
     
    zhouharvey likes this.
  16. zhouharvey

    zhouharvey PayPal’s $2 Sensible Enforcer

    Banned
    Joined:
    Feb 19, 2016
    Likes Received:
    19
    Dislikes Received:
    57
    Trophy Points:
    18
    Have you tried using it as a binaural microphone? How good does that work?
     
  17. bartzky

    bartzky Acquaintance

    Joined:
    Dec 22, 2016
    Likes Received:
    62
    Dislikes Received:
    0
    Trophy Points:
    18
    Location:
    Germany
    Home Page:
    A lot has already been said about measuring full size cans. I like to add something about IEM measurements and provide a calibration curve.

    First of all: These ears are small! It's unbelievably hard to get a good fit with most IEMs. Many IEMs have to be fitted upside down as the shell won't fit into the concha. With some IEMs I wasn't able to obtain a good measurement at all - like with the PSB M4U4.

    To get a calibration curve I measured 10 individual IEMs with the Ears and my IEC 60318-4 compliant setup. To reduce the effect of positioning I did 4 individual measurements per earpiece. So in total the calibration curve was obtained by taking 160 individual measurements. I tried to cover different driver types and numbers.

    Measured IEMs:
    • Brainwavz B100
    • Campfire Audio Andromeda
    • Campfire Audio Vega
    • InEar StageDiver 2
    • KZ HDS3
    • KZ ZST
    • Westone W40
    • Westone W60
    • Westone W80
    • Xiaomi Hybrid Pro HD

    The following graph shows the difference curves between IEC and Ears for each IEM in gray aligned @ 1 kHz (80 dB -> 0 dB). Red is the average difference across all ten IEMs. Read: "The Ears need X dB to fit the IEC standard."

    [​IMG]

    The above curves were measured with miniDSP's RAW calibration applied. It's apparent that there is some significant variance between the individual curves and the average. So even with applied calibration there will be an inaccuracy that needs to be considered. The Ears is no IEC compliant rig and calibration won't change that.

    The obtained calibration curve can be downloaded here: Ears Calibration for IEMs
    Use REW's arithmetic function (A*B) to apply your individual RAW calibration.
     
  18. Eudis Fernandes

    Eudis Fernandes Aurorus Audio - MoT

    Joined:
    Mar 23, 2016
    Likes Received:
    110
    Dislikes Received:
    4
    Trophy Points:
    33

    I've measured enough thx00 across different interations of my diy rig to know that indeed 3.5khz is typical on thx00 and similar models.

    ----
     
  19. Maxx134

    Maxx134 Dunning–Kruger effect poster boy

    Friend
    Joined:
    Oct 27, 2015
    Likes Received:
    818
    Dislikes Received:
    176
    Trophy Points:
    93
    Location:
    NYC
    No need to blast your headphones at 100db lol
    Most measurement at 90db at 1k, plus 25db range is fine no need to see 40db down
     
  20. MF_Kitten

    MF_Kitten Banned per own request

    Banned
    Joined:
    Jan 1, 2016
    Likes Received:
    179
    Dislikes Received:
    28
    Trophy Points:
    43
    Has anyone tried to simply remove the ears and replace it with a foam coupler instead? It'd give you a handy flat plate coupler stand. I think I'm gunna try it out some day. The neat thing is that you can just make a compensation curve by comparing the compensated measurement with the ear to the flat coupler measurement (measure a speaker, so the position remains the same).
     
    zhouharvey likes this.
Thread Status:
Not open for further replies.

Share This Page