2016 July/August NY area measurements by Serious

Discussion in 'Audio Science' started by Serious, Jul 30, 2016.

  1. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    2016 July NY Mini Meet in ear mic measurements

    I think I lost the text that I wrote over a week ago which annoys me. Basically I wrote about in ear measurements vs the more traditional coupler measurements. There is no such thing as a "best" measurement. Like with all things, it's a matter of tradeoffs.
    These measurements also follow a new posting standard for me:
    • All FR measurements will have 35db range, 20Hz to 20kHz with a resolution of 1400x600
    • Colors will be R000G255B000 for Left (Bright Green) and R255G000B000 for Right (Bright Red).
    If someone bitches, I can change it. Tell me if the colors look bad, or if you preferred the larger 1080p images. I wanted them to resemble the Changstar era measurements of Marv and Ultrabike.

    I measured the following headphones:
    • HD650
    • HD800(S)
    • PM2 modded Maxx134
    • HFM Edition S
    • HE1000 modded Maxx134 (L only)
    • THX00 (R only)
    • Code-X
    Keep in mind these are very noisy. Also, with my own IEMic measurements you will most likely see a peak at 8kHz. Don't worry about this. This is actually my ear and I don't think I can hear it. Also the area above 10kHz in general seem to be very inaccurate. I would take everything above maybe 6kHz with a grain of salt. Coupler measurements would be better for checking for resonances in this area. I don't really want to post the CSDs to prevent confusion. Many measurements have ringing at 8kHz, but that's just my ear. Where the IEMic measurements excel are (IMO) a good look at overall FR without a need for compensation and realistic bass measurements because of a realistic seal by design.
     
    Last edited: Aug 24, 2016
  2. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    HD6X0s:

    Modded HD650 (SteelCannon):
    FR IEM HD650 SteelCannon.png

    Modded HD650 (thegunner100):
    FR IEM HD650 thegunner100.png

    I believe both of these were modded by thegunner. The only real differences are the condition of the pads and headband clamping pressure, with (I believe) thegunner's being the looser ones, which I preferred. Overall both HD650s sounded slightly midbassy with slightly rolled off treble energy to and moderately too prominent upper mids to me. Stil, the HD6X0 headphones have a fantastic midrange tone.

    One thing that I found very interesting was the comparision against my modded HD600. My HD600 is coin modded and has dynamat mods but I didn't remove the spider (yet). I didn't bring my HD600s to 'Murica with me, so I had to rely on older measurements (1 or 2 months old). Still, they should be comparable.

    HD600 (Serious)
    FR IEM HD600 home 1.png
    As you can see, the bass extension is very similar. The region between 2-4kHz is actually slightly more prominent on the HD650s, but at the same time the resonance around 5kHz is less strong and subsequent treble energy is lower in level.
    I wouldn't worry too much about the dip in the treble in the HD650 measurements. I believe a lower headband position lifts this area on my head.
    I also attached a comparison of the L/R average of the 3 curves. Green is my HD600, orange is thegunner's HD650 and blue is SteelCannon's.
     

    Attached Files:

    Last edited: Jul 30, 2016
  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
    HD800

    This was interesting. We had 3 HD800s, none of which were stock and none of which were silver, and a stock HD800S. Like I said in the impressions thread, I preferred mine.

    First up: Craig's HD800, modded by Maxx134:
    FR IEM HD800 Black.png

    Despite the measurements, these didn't sound all that bright. They were hardwired and it could be that a lot of the subjective differences come down to the cable.

    Next up: Maxx134's personal pair of modded HD800s:
    FR IEM HD800 Maxx.png
    These were too bright for me. I would have to compare against a stock HD800, but they were probably still a lot better than the stock HD800.

    Time for my own HD800:
    FR IEM HD800 Serious.png

    These were the measurements I got at the meet. Usually the area above 6kHz is in line with the level at 500Hz. This depends a lot on how deep you insert the mic into your ears. The midrange looks very flat in these measurements and I actually preferred the midrange against the HD650. You can't do anything about the bass extension.


    Finally, the HD800S:
    FR IEM HD800S.png
    WTF. A big peak at 5kHz and a big dip at 6kHz. This is because the "6kHz peak" is actually at 5kHz on my head. It's possible that the HD800S may work better for other people, but as I see it, the HD800S is not a fixed HD800 at all. If anything it may actually be worse. The peak at 5kHz and dip at 6kHz make for a huge resonance at 5kHz.
     
    Last edited: Jul 30, 2016
  4. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    PM2 modded by @Maxx134
    FR IEM Oppo PM2 Maxx.png
    No subjective impressions since I didn't listen to them. Supposedly these are pretty good. FR looks too warm for me, but otherwise quite good.
     
    Last edited: Jul 30, 2016
  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
    HFM Edition S
    FR IEM HFM Edition S.png

    Again, I didn't listen to these. The differences between R and L reflect a different seal. Seems very peaky at 4kHz, especially with a bad seal. Generally not too great, unless this peak could be reduced.
     
    Last edited: Jul 30, 2016
  6. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    HE1000 modded by @Maxx134
    FR IEM HE1000L.png

    These still have the dip at 2kHz. Bad ortho wall at 5kHz. These were too bright for me.
     
    Last edited: Jul 30, 2016
  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
    THX00 (R only)
    FR IEM THX00 R.png

    Right channel only because I think I got a bad seal with the L channel, which lowered the whole region below 500Hz. Generally I thought these were U shaped with more bass than Marvey's and Tyll's measurements indicated and with a trouble spot at 10kHz that I didn't like. The dip at 8Khz is an IEMM™ artifact.
     
    Last edited: Jul 30, 2016
  8. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Code-X
    FR IEM Code-X.png

    These had the Audeze Vegan pads.
    Wow, these look weird. Let's go over them: Fantastic bass extension down to less than 10Hz, followed by a weird bump centered around 200Hz. Overall most likely warm midrange tone to 2kHz, where it starts rising again. Ortho wall between 5 and 6kHz. Instead of a peak at 8kHz, we see a dip with no ringing. Weird. Most likely the pads.
    The whole region above 3kHz is above the 500-100Hz midrange level which is a bad thing in my book.

    Overall these measurements don't look all that great despite the great coupler measurements that Ultrabike showed us.

    I really wish I listened to these. these are supposed to be one of the best modern planars and I don't doubt that, but at the same time I would like to see how a good modded HE6 compares.
     
    Last edited: Jul 30, 2016
  9. thegunner100

    thegunner100 Hentai Master Chief

    Pyrate
    Joined:
    Sep 25, 2015
    Likes Received:
    2,461
    Trophy Points:
    113
    Location:
    NYC
    Thanks for posting the measurements! My pair was the one with the looser clamp and more worn pads. I prefer the more relaxed sound that I get with the looser clamp, plus it plays nicer with my glasses.
     
  10. Maxx134

    Maxx134 Dunning–Kruger effect poster boy

    Pyrate
    Joined:
    Oct 27, 2015
    Likes Received:
    818
    Trophy Points:
    93
    Location:
    NYC
    It was a nice surprise too see you were able to take measurements at the meet and that you posted them.
    Thanks for the mic advise as well.

    Also of interesting note is that upon listening / comparing all our HD800 ,
    I belive mine sound darker to me than the others in overall signature.

    Craig's hd800 which I mod for him was using the French mod and my "bass tape" mod,, then it had the SAA Endorphin cable hard wiring, and the nice gloss black paint.

    My HD800 had a Sennheiser earfoam insert in place of the French mod, and bass tape mod, but I believe the whole frequency peak shift to higher range came from the dynamat extreme damping mod ...

    In general I feel that most attempts to fix the stock resonance,

    also diminishes sound staging,
    except for the French mod which seems to preserve it better.

    The Code-x had an Audeze vegan pads and an internal cleaning.

    The HEK had two layers of organza internally, as it is sensitive to modding. .
     
    Last edited: Jul 30, 2016
  11. Maxx134

    Maxx134 Dunning–Kruger effect poster boy

    Pyrate
    Joined:
    Oct 27, 2015
    Likes Received:
    818
    Trophy Points:
    93
    Location:
    NYC
    I accidentally ha d typos and deleted parts of my post above due to using my phone browser.
    So I fixed it.
     
  12. Maxx134

    Maxx134 Dunning–Kruger effect poster boy

    Pyrate
    Joined:
    Oct 27, 2015
    Likes Received:
    818
    Trophy Points:
    93
    Location:
    NYC
    Most of the headphones exhibited some distortion(wiggly lines) from 400-1k and I suspect that this is the background noise leaking in from the open headphones.
    Since "Serious" HD800 had rug liner,
    I suspect the lack of wiggly lines as a result of the damping of external noise.


    Also, I forgot I also did the bass mod on Craig headphones and so you will notice from 30 to 40 Hertz both mine and his stays closer to the 80db mark than the other HD800 headphones...

    I suspect I will have to do some further testing to compare and find out the difference in my dynamat mod.
    Subjectively I noticed less resonance, less Soundstage and more definition.
    This is not necessarily better as I also like Craig HD800 as I find his sweeter and mine darker, probably due to the lower midrange increase on mine
    Remember more music will be in this area.
     
  13. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Yes, forgot to mention that this would translate into this. You could narrow the IR windows for the FR visualization but then the accuracy would drop below a few 100Hz, depending on how much you decide to limit it. I usually use 500ms left and 750ms (with BH7 window) which is plently to get good accuracy into way below audible frequencies. Longer was just noisier. You could use a shorter and different window left of the IR, I guess everyone will have their own settings in that case.
    In this case I experimented with shorter windows and the noise never went away, so I just left it at my usual settings.

    You will also sometimes see a very narrow peak or dip at either 50 or 60Hz. This is simply the power hum and can be ignored. I will try to get rid of it when I try to get better distortion measurements, but for now this isn't a high priority to me.

    Actually, when I did an overlay of all 3, the bass extension never changed. The HD800S might have had just a bit better bass extension, but that could also just be measurement variance

    Here's an overlay I made very quickly. I tried to match the colors of the 3 HD800s with their real color, but that didn't work with the HD800S. The HD800S is green for no reason. I matched them all at about 100Hz, but you can still see the real level (before matching) at 500Hz in the legend.
    FR IEM HD800all.png
     
  14. 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'm going to add Elear measurements when I figure out a compensation curve for the half broken WM61A that I had to use and transfer the files from my craptop to my desktop. That's just a matter of time. Maybe today, maybe not. Don't worry, I still have a bunch (of WM61As) and also ordered 50 which are on my way soon. Oh and I will most likely ditch the WM61A for a superior mic capsule in the future.
     
    Last edited: Aug 24, 2016
  15. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Will now post the coupler measurements I got. This means CSDs. Yay. Will not post distortion for now as I'm pretty sure all of them were too noisy.
    These coupler measurements are a little different from the few that I posted before. I didn't use a plate behind my foam coupler. This means a smoother 500Hz-3kHz region (no dip at 2kHz anymore). This also means that the "coupler target response" plot that I posted some time ago are outdated now. Now it's a smoother response from 500Hz to 3kHz, but still with a dip at 4-5kHz. I like this coupler for open headphones but I couldn't get a seal with the TH-X00 and Maxx134's closed headphones. In a sense this is my CSD-coupler as it doesn't add resonances that aren't there.
    The coupler measurements were taken with a minidsp UMIK-1. This doesn't really change the results much.

    CSD settings (wish I could add a kool green matrix or 4chan style text on the top left corner of my pics):
    100 slices
    13.37ms window (Hamming left/BH4 right)
    0.25ms rise time
    1/48th smoothing
    no use CSD mode
    36db range relative to 90db reference level at 500Hz. This will mean closer to 33-30db range in the upper mids and treble if the headphone follows the target.
    347Hz-20kHz frequency range (trying this now instead of 500Hz on the low end)

    The window length and rise time are super crucial. Window has to be sufficiently long, but you also have to avoid getting reflections. Fortunately these measurements don't seem to have very many reflections.
     
  16. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    HE1000 modded by @Maxx134 coupler measurements:

    Frequency response

    HE1000 FR.png

    CSD
    (in bright, flashy colors of cancer)
    HE1000L CSD.png

    HE1000R CSD.png


    This was with the same mods that it had before which I think were simply two layers of organza behind the driver. Subjectively I didn't like the 2kHz dip and it was too bright for me from 3-4kHz on. I hate a rising response from 2kHz on. Apart from the ortho wall the CSDs look pretty clean. What's weird is that the little wiggles in the midrange weren't at the same frequency for L and R and that the right channel always measured with more subbass than the right. This is not measurement error in this regard.
    Ultrasonic extension up to around 42kHz, where it starts dropping off rapidly.
     
    Last edited: Aug 26, 2016
  17. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    @Maxx134's Code-X with Audeze Vegan pads coupler measurements:

    Frequency response

    Code-X FR.png

    CSD
    (in bright, flashy colors of cancer)
    Code-X L CSD.png

    Code-X R CSD.png

    You knew they were going to be super clean. Left channel looks a little worse than right. These had the Audeze Vegan pads (which I think was the only mod over the regular LFF Code-X). I would ignore the small hashy spikes in the treble. FR makes sense given the IEM measurements. The CSDs are nearly perfect. Close to the cleanest headphone I've ever measured.
    Ultrasonic extension seemed to be better than the HE1000. Extension up to around 50kHz. I don't know how much my mic rolls off.

    Subjectively they didn't sound as dynamic as my HD800 on Maxx134's amp. Also the FR was more wonky. I wonder what they sound like with the stock pads.
     
    Last edited: Aug 26, 2016
  18. Maxx134

    Maxx134 Dunning–Kruger effect poster boy

    Pyrate
    Joined:
    Oct 27, 2015
    Likes Received:
    818
    Trophy Points:
    93
    Location:
    NYC
    About the HE1000:
    I noted to Serious that I believe my drivers in this replacement unit are not matched.
    Although I did not note any sub bass difference ,
    I did notice one side seemed slightly brighter in mids than the other,
    yet they both had same tonality..
    So I decided to let them play in a drawer, as they were new replacements and had less than a few (2maybe?) hours use.

    I have since modded them with the addition of the "acoustic coupler" mod which also address the treble response.

    About the Code-x:
    Please take note I DID NOT mod them in any way.
    I only :
    cleaned them internally
    and washed the felt by grill,
    Painted the grill,
    Filled in a crack in the wood.
    Apply new Audeze pads, only because they sold this pair to me with a POS bullsh*t Alpha pad which has a small oval inner diameter,
    So it was a totally incorrect pad to even use.
    When I purchased them I had no idea it has all these issues before I received them.
    EDIT***
    I found out that yes indeed the Code-x comes stock with the Alpha Pads,
    So definitely the Audeze pads were an upgrade as it gave a much more open & less congested sound.


    Anyways the point is they are great, and I only pad rolled them..
     
    Last edited: May 10, 2017
  19. Maxx134

    Maxx134 Dunning–Kruger effect poster boy

    Pyrate
    Joined:
    Oct 27, 2015
    Likes Received:
    818
    Trophy Points:
    93
    Location:
    NYC
    Also take note we were using a higher output impedance tube amp which I modded.
    The Code-x do sound more dynamic out of my friend Craig's EC2a3.
    The HD 800 will sound more dynamic than Code-x out of my tube amp.
     
    Last edited: Aug 26, 2016
  20. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Yep, wished I could have measured this. To me it seems like a good concept and I wouldn't be surprised if it sounds and measures a lot better now.
    Yeah, that's what I thought. The only change is the pads. I should have made that more clear. Updated the title.

    I'm sure a more beefy amp will make them sound better, but I still find Orthos to lack the dynamics of great dynamic headphones like the HD800. Maybe you need a super beefy 400W speaker amp to truly kick their ass, but I still doubt that they can sound as dynamic.


    Also, to not confuse the others, I always measured with my Leckerton.
     

Share This Page