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. Psalmanazar

    Psalmanazar Most improved member; A+

    Pyrate Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    5,345
    Trophy Points:
    113
    That's the reason I bought Objective combo. I didn't feel like carrying the Schiits around the house so I was willing to pay another hundred bucks for a JDSLabs product with RCA outs for speakers and a real jack.
     
  2. m8o

    m8o Acquaintance

    Joined:
    Nov 2, 2015
    Likes Received:
    21
    Trophy Points:
    18
    Thank you for these test.

    It's the M9XX that interests me. I don't mind the harmonics in most graphs, but the first. What was the input voltage?

    Any chance a 32 or 50 ohm load would affect anything on the M9XX?
     
  3. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    Wait a minute, those 60Hz THD plots are misleading because it looks like we're looking at some ripple instead of a THD component.

    BTW I was looking for a THD loopback, not just the noise floor.

    Notice the second order on the 1k and the 60Hz FFTs are pretty close in level.

    [​IMG]
    [​IMG]

    The third harmonic does not look to be the device, but rather that 180Hz ripple component that matches up with what you posted.

    [​IMG]
    [​IMG]

    I am not familiar with the mechanism that makes solid-state electronics less linear at low-frequency, but I'm open to be wrong. THD+N might very well increase because of flicker noise, but the harmonic components shouldn't be any higher.
     
  4. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Yeah, something may be up with at 180 Hz when using 60 Hz. However, notice that at least in the iFi Nano measurements a large part of the 180 Hz distortion seems to come from the UUT itself. For the others, it may have to do with either the measurement DAC or ADC.

    And yes, M1 seems to be pretty close in level for single ended measurements (with the exceptionof the iFi stuff).

    I'm confident howerver that the QA400 kicks the living crap out of my 2i2 interface on single ended measurements.

    Question is, what' is better than the QA400 for single ended audio band data captures for less than $300? (which interface through USB). I'm looking around but most stuff starts to climb into the kilobuck range. And pretty soon one starts getting into AP, Prism, or SR1 price ranges.
     
  5. fuzzybaffy

    fuzzybaffy New

    Joined:
    Nov 1, 2015
    Likes Received:
    0
    Trophy Points:
    1
    Ojneg: what are the measurements you just posted? I'm such a noob, sorry...

    Are those measurements of the m9xx? What was the overall meaning of your post? To say that the m9xx didn't measure as badly on your machine? Or were those measurements for something else?
     
  6. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    Same as what Marv posted, all the Magni/Modi. Just wanted to spotlight that.
     
  7. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    As I said, not really comparable because the loopback is using the SE outputs and inputs. With the dac/amp measurements, I used the balanced probes which have a little bit higher noise, but much less environmental spurious signals.

    I'll rig up something to do the differential loopback. Ultimately, the exercise is academic since the iFi, and the Grace to a lesser extent, still clearly do not measure as well as the others for the 60Hz tone. Just draw at line at -100db and look up from there.
     
    Last edited: Nov 2, 2015
  8. m8o

    m8o Acquaintance

    Joined:
    Nov 2, 2015
    Likes Received:
    21
    Trophy Points:
    18
    The M9XX 1K harmonic spectrum has me perplexed.

    What are those tones just below the 2K and 4K harmonic? Is that IM distortion from harmonics interacting?

    If it wasn't for that, those 7th and 9th order harmonics, and that almost 'spread spectrum' between 10K and 20K, I'd be thrilled having only even order harmonics. Just tough to look past all those high order harmonics.

    I'm still letting the order go through. Yes, the levels are still low. And I need something 'better' than the FiiO E17 to travel with me.
     
    Last edited: Nov 2, 2015
  9. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    It appears the 240Hz spike (probably 4th harmonic from AC) is interacting with the 2kHz.
     
  10. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Ortho load tests (33 ohms) have been posted back here: http://www.superbestaudiofriends.or...ts-grace-schiit-ifi-lh-jds-etc.427/#post-9854

    Can't wait for the JDS Element to come in. Will update the charts when it does.

    And BTW, I know people tend to complain that we are big Schiit and LH fans. Well, maybe, just maybe the measurements might correlate somewhat to what we hear. Keep in mind that the folks who have said good things about Schiit or LH gear never even saw these measurements before.
     
    Last edited: Nov 2, 2015
  11. fuzzybaffy

    fuzzybaffy New

    Joined:
    Nov 1, 2015
    Likes Received:
    0
    Trophy Points:
    1
    Sorry for so many questions, but was the filter setting on the m9xx set to default? Because, IIRC, there are four different filter setting, and I seem to recall something about your cat accidentally messing with the settings...
     
    Last edited: Nov 3, 2015
  12. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    No problem at all. Please ask questions. The more Q&A while I still have this unit, the better. I tend to go through stuff very quickly and sometimes I miss little things. You won't be able to ask questions or make specific requests later.

    I used F1, which I believe is the linear phase filter with sharp cutoff. The linear phase filters have always been my favorite. I also prefer the sharp cutoff, but on occasion with some DACs, I've preferred the slow rolloff. The minimum phase filters I've never liked - rather fatiguing over time. I tried F3 (minimum phase sharp roll-off) briefly and did not like it. The filters on Grace do not sound all that different. The effect is very subtle.

    On the other hand, the different filters on the GOV2 sound quite different. I prefer the FRM (linear phase roll-off) to the TCM which screams "Sabre Digititus!". There is a third digital filter the GOV2 for streaming sources. That one basically sounds like ass (streaming or not) and I have no idea why LH Labs included it.
     
    Last edited: Nov 3, 2015
  13. mickeyvortex

    mickeyvortex Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    36
    Trophy Points:
    18
    Maybe
    Maybe USPS needs to get one of these babies to deliver the Element.
    image.jpeg
     
  14. fuzzybaffy

    fuzzybaffy New

    Joined:
    Nov 1, 2015
    Likes Received:
    0
    Trophy Points:
    1
    Thanks for being so open and welcoming of questions, Purrin.

    Are these measurements for the DAC portion of the m9xx? Or just the amp portion? Or for the overall sound, and therefore both?
     
  15. USAudio

    USAudio New

    Joined:
    Nov 4, 2015
    Likes Received:
    6
    Trophy Points:
    3
    Location:
    Seattle, USA
    Thanks for the terrific work here.

    An iFi Micro iDSD would be interesting as well, as it is closer in price ($499) to the Grace and "should" measure significantly better than the Nano.
     
  16. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Measurements are all of DAC/amp combinations. The intended and assumed use.
     
  17. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    IMPORTANT UPDATE ON m9XX MEASUREMENT:

    I wanted to post an important update regarding the Grace m9XX measurements. I've been speaking with Michael Grace the past two days. Here are his results (using an AP) for the 1kHz tone:

    IMG_2913.jpg

    Compared to this from me using the QA400/QA190:
    [​IMG]

    We are trying to figure out why I get the high frequency spurs on the m9XX and not the others. Some possibilities:
    1. Monoprice power adapter which has thrown in the box temporarily (the deliveries going out are using a better quieter adapter, not the Monoprice one, so no worries!)
    2. Something specific to my environment and the unit. Maybe need to try another place that's not my house.
    3. Something specific to my PC. Maybe need to try another laptop.
    4. Interactions between the QA400/QA190 and the m9XX. Maybe need to try a soundcard.
    5. Possible defective unit.
    I thought I should mention that Michael Grace has been gracious (pun intended), super cool, and very professional. We are both working together to solve this mystery. I'd like to give Grace the benefit of the doubt, as OJ (who has access to AP gear) mentioned to me that the Grace stuff always measured superbly.
     
    Last edited: Nov 4, 2015
  18. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Awesome you guys are getting to the bottom of this. Hopefully I can learn one or two things :)
     
  19. JK47

    JK47 Guest

    Ya, rather than talking crap, you guys are searching for other variables and not jumping to conclusions.
     
  20. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Measurements on DACs and amps are always super tricky. I've had the capability to do these kinds of measurements for several years now, but have never made a lot of stuff public or official. Lots of pitfalls, lots of tricks, lots of environmental factors, lots of opportunity for user error. I've ran into too many "WTH" scenarios. I've always tried not to take a cowboy approach.

    People probably don't recall that I only started to offer headphone frequency response measurements six months or more after my CSD methodology (which in turn was maybe two years after I joined HF). It's only until now that I feel somewhat comfortable, but still not super comfortable offering these kinds of measurements. (Jason's tip on the QA400/QA190 accelerated this process.) Now if only AP had the wisdom to give me free measurement gear instead of Jude!

    Anyway, I'm pretty sure the methodology used here and these sets of measurements aren't totally broken, otherwise the Schiit and LH Labs gear would not have measured so well.
     

Share This Page