Measurements: Mid-Fi DAC/Amp Comb Units (Grace, Schiit, iFi, LH, JDS, etc.)

Discussion in 'Portable and Other Gear Measurements' started by purr1n, Nov 1, 2015.

  1. 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
    Discussion thread is here: http://www.superbestaudiofriends.or...p-comb-units-grace-schiit-ifi-lh-jds-etc.411/

    Just a reminder:
    1. Grace m9XX (with extra USB power plugged in)
    2. LH Labs Geek Out V2 Infinity
    3. JDS The Element (just ordered)
    4. iFi iDSD Nano
    5. Schiit Modi 2U / Magni 2U
    BTW, my wife and I have already listened to most of them. We probably have the least time with the iFi iDSD Nano. We are still going back and forth a little bit doing some 1vs1 to nail down some impressions. We also need to wait for the JDS unit to come in. Give us another week for complete subjective impressions. Subjectively, there are some clear obvious winners and losers. Some are close.

    DSC_2015.jpg

    I think there is some pretty decent correlation with the measurements and our subjective impressions so far. We are going to start with a 330ohm load which is nominally the HD650s impedance (but actually easier since we are using a straight resistor with no impedance bump in the bass). The voltage out will be set to 0.550Vrms or close to that, which is quite loud on the HD650. Note that's it impossible to get it to a perfect 0.550V with volume controls / potentiometers on the amps. The screen captures will indicate around 55mVrms or 0.050Vrms because I am using /10 differential probes to avoid blowing the QA400. Noise level is good tonight, no spurious crap.

    The measurements will be as follows:
    1. 1kHz distortion spectrum
    2. 60Hz single tone
    3. 600Hz and 1700Hz 1:1
    4. 60Hz and 7000Hz 2:1
     
    Last edited: Nov 1, 2015
  2. 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
    High Impedance Dynamic Headphone Loads

    1kHz 0.550Vrms into 330 ohms

    The Schiit Uber stack and the LH Geek Out V2 are the runaway winners. iFi Nano iDSD is the loser, although the Grace has some minor garbage high up. All measurements are via USB input.

    Grace m9XX (see also links below)
    http://www.superbestaudiofriends.or...e-schiit-ifi-lh-jds-etc.427/page-2#post-10702
    http://www.superbestaudiofriends.or...e-schiit-ifi-lh-jds-etc.427/page-3#post-10891
    IMPORTANT UPDATE - SOLVED: http://www.superbestaudiofriends.or...e-schiit-ifi-lh-jds-etc.427/page-3#post-12307
    Grace m9xx p.png

    iFi iDSD Nano
    iFi iDSD Nano.png

    LH Geek Out V2 Infinity Balanced
    LH GOV2b.png

    LH Geek Out V2 Infinity SE
    LH GOV2se.png

    Schiit Magni 2U / Modi 2U Stack
    Schiit mm2u.png

    JDS The Element

    JDS The Element.png
     
    Last edited: Nov 10, 2015
  3. 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
    60Hz 0.550Vrms into 330 ohms

    Grace m9XX
    60 Grace m9xx.png

    iFi iDSD Nano
    60 iFi nano.png

    LH Geek Out V2 Infinity Balanced
    60 GOV2b.png

    Schiit Magni 2U / Modi 2U Stack
    60 Schiit mm2u.png

    JDS The Element
    60 JDS The Element.png


    (Note: Better conditions tonight for JDS measurement. No small spike at 60Hz and 240Hz for any of the measurements like two nights ago, including the other gear, so keep that in mind. It's the nature of measurements.)
     
    Last edited: Nov 5, 2015
  4. 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
    Last edited: Nov 14, 2015
  5. 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
    Last edited: Nov 14, 2015
  6. 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
    Low Impedance Ortho Headphone Loads

    The following set of measurements are intended to simulate typical ortho load. This measurement will suck more current from the amps. I didn't do a full set, just the 1kHz and 60Hz+7000Hz 2:1 tones. It's pretty clear the pattern repeats itself, although the Grace and iFi units seem to struggle more relative to the LH and Schiit gear under the more difficult load.

    1kHz 0.550Vrms into 33 ohms


    LH Labs Geek Out V2 Infinity
    GOV2i.png

    Grace m9XX
    m9xx.png

    Schiit Modi 2 Uber / Magni 2 Uber

    mm2u.png

    iFi Nano iDSD
    nano.png

    JDS The Element
    JDS.png
     
    Last edited: Nov 5, 2015
  7. 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
    60Hz and 7000Hz 2:1 (same total power level) into 330 ohms

    LH Labs Geek Out V2 Infinity
    6-70 GOV2i.png

    Grace m9XX

    grace gn.png

    Schiit Modi 2 Uber / Magni 2 Uber

    schiit good night.png

    iFi Nano iDSD
    6-70 nano.png

    JDS The Element
    upload_2015-11-5_22-46-21.png
     
    Last edited: Nov 5, 2015
  8. 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
    RESERVED
     
  9. 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
    OK, you guys can comment now. I'm uploading one more two tone measurement.

    In summary, the GOV2 and Schiit Uber Stack are the best measuring. The Grace m9XX tends to have crap in the treble, higher order distortions. The iFi has the most crap, and it tends to be lower order distortions. We'll wait for the JDS to come in.
     
    Last edited: Nov 1, 2015
  10. BioniclePhile

    BioniclePhile The Terminal Man - Friend

    Pyrate
    Joined:
    Oct 25, 2015
    Likes Received:
    386
    Trophy Points:
    63
    Location:
    Admudnson-Scott Antarctic Facility (Seattle WA)
    Home Page:
    The Geek Out V2 and Magni2/Modi2 seem to have the least amount of irregular squiggly lines, which appears to be a good thing if I were to say so without any previous knowledge.
     
  11. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    This is good Schiit. :piratemug:
     
  12. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    So it seems the GOv2 balanced surpasses your system's measurement resolution with the 1k tone, but at low frequencies is not quite as linear. Surprised to see that TBH especially given the large difference. Are we missing something here?

    Other than that it seems to be all typical. IMD plots are always interesting to analyze, but by-and-large, we can predict the IMD performance (and overall linearity) just by looking at THD. I'm not familiar with any cases where THD is low but IMD is orders of magnitude higher (or vice versa).
     
  13. fuzzybaffy

    fuzzybaffy New

    Joined:
    Nov 1, 2015
    Likes Received:
    0
    Trophy Points:
    1
    Are the measurements for the M9xx with the second USB cord plugged into a power outlet? Or just the primary USB cord into the computer?
     
  14. 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
    Aye. Second cord on the M9xx was plugged into a power outlet.
     
  15. 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
    That's normal. Bass is harder to reproduce.
     
  16. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    Harder to reproduce for transducers with limited excursion, but we're talking electronics here.

    Did you ever post a loopback for your analyzer?
     
  17. 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
    I've observed the same for electronics. The 1kHz is always great. Bass sometimes not so much.

    Loopback using same sensitivity / calibration @0dBr as the 0.550Vrms tests. Keep in mind this is SE loopback. I'd have to make cables for balanced probe loopback, which I know has less spurious signals.
    loopback.png
     
    Last edited: Nov 2, 2015
  18. fuzzybaffy

    fuzzybaffy New

    Joined:
    Nov 1, 2015
    Likes Received:
    0
    Trophy Points:
    1
    I was about to pull the trigger on the m9xx, but now the measurements are kind of getting me concerned. :(

    So does this mean the Schiit stack is measuring better? I wonder if this translates to real-world, audible differences...

    Also, when you receive the JDS Element, will you be doing measurements on that as well?
     
  19. Psalmanazar

    Psalmanazar Most improved member; A+

    Pyrate Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    5,345
    Trophy Points:
    113
    The Schiit stack is also the cheapest and among the better built.

    I'm not entirely surprised by the results of the test. Where would you place the ODAC+O2 combo? Right below the Schiit stack as it's more expensive? My friend and I both couldn't pick out his Modi vs my ODAC volume matched on equalized near fields in his home studio in ABX tests.
     
  20. 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
    Yes, will do measurements on JDS when it comes in. Will also do 33ohm loads to simulate typical planars.

    It's true that the measurements on the m9XX don't look great compared to the Schiit stack or LH; but honestly, the distortions and spikes are theoretically below the threshold of hearing. We are looking at junk 60-70db below the fundamental tones. That being said, the Schiit Uber stack is the best sounding overall. The one caveat is that the Modi 2 is a bit more grainy / digital sounding than the Grace m9XX or the Geek Out v2; but it's nothing out of the ordinary and well within my tolerance level, but I thought it should be mentioned. Shockingly, the Sabre based GOV2 has the least grainy / digital characteristics, but then again I am running it in FRM mode which has an early treble roll-off. The Schiit stack is the most neutral. The Grace is warmer with the most delicate treble. The GOV2 is between them. It will be interesting to throw the JDS into the mix when it arrives.

    I wouldn't discount the Grace yet. I've tried travelling with the Schiit stack, and it's not easy carrying around two boxes, RCA cables, and two wall transformers. The Grace is pretty easy to pack up and throw into the bag, and besides it does have the best volume control knob.
     

Share This Page