Schiit Gungnir Multibit impressions

Discussion in 'Digital: DACs, USB converters, decrapifiers' started by Bill-P, Oct 7, 2015.

  1. Mr.Sneis

    Mr.Sneis Friend

    Pyrate
    Joined:
    Oct 2, 2015
    Likes Received:
    1,321
    Trophy Points:
    93
    Location:
    Phoenix, AZ
    I can definitely sympathize where you are coming from here, I have the "sounds like digital" problem after being heavy into vinyl for going on about 5 years now.

    As a sanity check, this weekend I used a Modi Multibit to compare to my remaining vintage DACs and to be honest it made me realize that the schiit MB sound it not without flaws but at the end of the day it's not an easy task to match or better the overall level of performance. All of it sounds fatiguing to me in one form or another so I am wondering if it is this kind of distortion I am sensitive to. Was never an issue for me 10 years ago!
     
  2. rlow

    rlow A happy woofer

    Pyrate Contributor
    Joined:
    Jul 18, 2017
    Likes Received:
    7,736
    Trophy Points:
    113
    Location:
    Canada
    Yeah you’re talking about the infamous crossover glitch. This was corrected some time ago.

    I actually appreciate your opinion here - I love decenting opinions. We need them. And maybe in this case, if you are listening to older Schiit DACs, you are hearing this glitch and this is what’s causing your issue.

    However I would have liked for you to try a different/better transport before coming to this conclusion. USB direct from computer IMO is bad for ‘etch’. I couldn’t handle Gen 5 USB on Yggdrasil A2 - just couldn’t. And this was using an expensive USB transport. Instead I had to run my transport through Eitr/SPDIF to calm it down.

    I would also be very interested for you to hear one of the newer Schiit’s to compare. I suspect that even with the corrected glitch, you will find something similar in your dislike of them - and that’s fine, they’re not for everyone. However the glitch may be a bit of a red herring in this case - it would be interesting to know.
     
  3. Biodegraded

    Biodegraded Friend

    Pyrate Contributor
    Joined:
    May 28, 2017
    Likes Received:
    7,979
    Trophy Points:
    113
    Location:
    Vancouver BC
    See the first graph in the Yggdrasil V2 measurement comparing the waveforms of V1 & V2 @ -90dB, and see the THD & THD+N graphs. Also worth looking at the Modi Multibit v2 firmware measurements.

    It seems from these that the 'glitch' was indeed fixed. But that's not to contradict @rlow 's comments above.
     
  4. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    This thread makes no sense (EDIT: Thread moved here, so maybe it makes a little more sense now).

    What is described here is a problem that is common to most multi-bit designs. Vendors of these types of DACs deal with these issues in different ways. @schiit indeed has continued to evolve their MB products to further address this problem.

    But the reason I find this thread senseless is that:
    1. All these issues have been discussed here before.
    2. The title "Why I don't like Schiit DACs" is misleading, because @schiit offers plenty of delta sigma based products that have none of these issues.
    3. Nelson Pass is full of shit. In that paper he says all distortion is bad (which I fully agree), and then says odd order distortion is fine (which is bullshit).
    @skem, just avoid multi-bit and stick to DS if you feel so strongly about it.
     
    Last edited: Jan 27, 2019
  5. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    (1) glad to hear the problem is real. Note that I acknowledged the problem didn’t appear in DS measurements.

    (2) feel free to change the thread title to “Why I don’t like Schiit Multibit DACs... but you might.”

    (3) since the problem is/was real, and others above have said this helps explain their feelings about these DACs, I’d say this thread has already been useful.
     
  6. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Don't be afraid to express your opinion or findings.

    However, this probably belongs to Gungnir thread. Specially if we have a multibit one already. I'll move it later.

    Indeed MB might not be for everyone.

    Note that we don't have single personal threads on: Why ultrabike likes porta pros but you might not, Why ultrabike likes the 2i2 but you might not, Why ultrabike does not like Orb speakers but you might, Why ultrabike does not like this but you might.

    I would just post on the appropriate product page, depending if we are talking a mix of impressions and/or new contributed measurements or other relevant information.
     
    Last edited: Jan 27, 2019
  7. Psalmanazar

    Psalmanazar Most improved member; A+

    Pyrate Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    5,345
    Trophy Points:
    113
    Benchmark and Crane Song products were made for you @skem

    This kinda shit is up there with the Lynx Hilo vs Prism Sound vs Bic Cristal loopback tests and distortion measurements in idiocy-filled threads that were so popular on gearslutz years ago. The final conclusion is always “Don’t buy blind and buy the one you like the sound of better.” All those tiny bits of distortion may or may not be audible and certainly very well may not be causing what you don’t like about the sound. I highly doubt the old fixed crossover glitch is causing what you dislike about Schiit Multibit products.

    Nelson Pass while experienced is far from an authority on the cleanest and clearest possible gear. He doesn’t make it himself.

    The rest of us can enjoy our lofi distorted garbage that can barely match or exceed CD Redbook specs for SNR and THD yet stomp the tape machines used on all older recordings and the mics used to record everything.
     
    Last edited: Jan 27, 2019
  8. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    Uggh. Benchmark sounded like boring undetailed crap. I haven’t tried Song Solaris, but will.

    Edit: Crane Song Solaris is excellent. My #2 favorite DAC now.
     
    Last edited: Mar 30, 2019
  9. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Alright. Merged the "Why I don’t like Schiit DACs... but you might" thread here since @skem is discussing the Gungnir MB DAC.

    As far as Benchmark being boring undetail crap, it seems @skem is still searching for the DAC that will make him stain his pants. Hope you find it mang.
     
  10. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    So many issues with the post https://superbestaudiofriends.org/i...-multibit-impressions.158/page-40#post-242979.

    First, the glitch is gone. Here is an example for Gungnir MB A2:
    FFT_0_L1T18_1.png
    This graph has been scaled differently with the y-axis in dBu for the FFT which is an absolute value. -75 dBu is 90 dB below +15 dBu which corresponds to 0 dBFS for the Gungnir MB A2. The oscilloscope sine is at -90 dBFS.

    Second, the 65 dB below the fundamental statement in the OP ignores that the fundamental was at -60 dBFS to begin with. So that 3rd harmonic is at -125 dBFS. That will translate to -110 dBu. If you can hear -110 dBu on any set of headphones with an amp running unity gain, you have the best ears on the planet.

    Here is another updated graph to consider:
    20190115 Gungnir MB A2 A12 THD+N FFT 1 KHz settings.png
    FFT_2_L1T13_1_A.png
    The fundamental is at -3 dBFS which is +11.7 dBu in this measurement as shown in the settings above the graph and thus the peak is off the chart above the y-axis which begins at 0 dBu. Again, this has been scaled for absolutes. With such a loud fundamental, the 3rd harmonic is at -90 dBu. For Focal Utopia with sensitivity of 104 dB SPL / 1 mW at 80 ohms the fundamental would produce +125 dB SPL while the 3rd harmonic output would be at +35 dB SPL.

    Again, if you can perceive 3 KHz at 35 dB SPL while being subjected to 125 dB SPL at 1 KHz you have amazing auditory acuity.

    Preference is one thing and I get that Schiit MB DACs aren't for everyone. But mis-interpreting measurements and making broad statements based on out-dated data is quite another issue.

    Sorry about the graph visibility issues which hopefully have been corrected. The threads were merged as I was editing a response.
     
    Last edited: Jan 27, 2019
  11. Psalmanazar

    Psalmanazar Most improved member; A+

    Pyrate Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    5,345
    Trophy Points:
    113
    Your bright British speakers will influence the treble way more than anything you stick behind them. Same with your room’s reflection points. Blame those for the treble if you want to complain, not a relatively tame converter. More unhinged or drier conversion would destroy you.
     
  12. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    Thanks @atomicbob. I can't believe I tripped up on the dB subtraction, but you're right! I've annotated the original post to point out the mistake.

    EDIT: Turns out I didn't screw up on the subtraction. The distortion is 65dB below, which his what matters, and "unity gain" conversion to dBu is irrelevant.
     
    Last edited: Mar 30, 2019
  13. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Because "people" already knew the fixes were in place with Yggdrasil A2 and Modi MB A2 (unannounced). The Gungnir Multibit A2 also technically doesn't exist, even we suspect it's been out for at least a year now.

    Yggdrasil A2 measurements are here:
    https://www.superbestaudiofriends.o...rasil-v2-upgrade-technical-measurements.5820/
    https://www.superbestaudiofriends.o...ments-deconstructing-asr-amirs-hack-job.6442/
     
  14. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    I removed the tweeters and it still sounded funny, as it said in the original post.
     
  15. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    I do not agree as there is insufficient information presented.
    The graph to which I think you are referring has a y-axis of dBr. What is the reference for that graph? If the fundamental normalized to 0 dBr is at -60 dBFS then it would be consistent my measurements. Otherwise I have no idea what that measurement represents as the reference for dBr hasn't been given.
     
  16. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    The glitch was fixed around the time JA released his measurements, way before the redone A2 boards (which not only incorporated glitch fixes, but a tweaked output stage capable of driving "studio" loads at 600ohms, and possibly board quality and circuit layout improvements). No difference in sound.

    If the A2 gear sounds more palatable to you, it's more likely the result of the other more significant changes. Otherwise, the Yggdrasil and Gungnir A2 still remain moderately incisive sounding DACs. In the words of TAS head-honcho Robert Harley:

    ...without etch, clarity of musical line, whole-body involvement—the Yggdrasil is as good as digital gets. Yet the Yggdrasil’s bold incisiveness may not resonate with listeners who prefer a more relaxed and easygoing sound. I, however, have no such reservation...
    Don't be like the Madam Sandler guy on Computer Audiophile who was too quick to find extremely minor non-audible issues with measurements and equate them with a certain quality of sound that he didn't like. He claims he heard all this harshness and shit from his new Yggdrasil, posted all the Yggdrasil measurements from JA, when I already knew the the glitch was fixed. All these DAC measurements are so good that they remain more a curiosity and less an evaluative tool. This is the curse of measurement gear that has gotten better and better every year in the last 30 years. Listening to DACs remains supreme. So does listening to headphones where errors are gross (measurements however still remain a good tool to "weed" things out so we can better direct our audition efforts).

    It is what it is. There are other DACs that are more polite and forgiving from Metrum, PSA, Crane Song, etc. I don't remember where I posted this, but I wrote up a quick history of DACs and concluded that the Schiit higher-DAC DACs have a sound more in common with typical D-S DACs than vintage R2R or discrete ladder. MASH, DSD, PWM is a whole another approach.
     
    Last edited: Jan 27, 2019
  17. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    It would be hilarious and tragic if all the Schiit DACs I've tested were just old and glitchy–though @purr1n says no difference in sound. I somehow never picked up on the crossover glitch in all my reading here. I suppose I should try Schiit once again before deciding.
     
  18. Psalmanazar

    Psalmanazar Most improved member; A+

    Pyrate Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    5,345
    Trophy Points:
    113
    You complained about brightness and etchedness. Putting a blanket over bad tweeters won’t get rid of bad tweeters. NS10M with toilet paper still doesn’t come close to the 1990s NS10M Studio or contemporary HS5. B&W pain won’t go away until you get rid of B&W speakers.
     
  19. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    @atomicbob
    If fundamental is at -60dB and harmonic is at -125dB (and assuming a power scale)
    Then the "dB below" is the ratio:
    10 Log (absolute-harmonic-level / absolute-fundamental-level)
    where the absolute-harmonic-level is -125 dB = 10^(-125/10)
    and the absolute-fundamental-level is -60 dB = 10^(-60/10)
    Is that correct?
     
  20. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,652
    Trophy Points:
    113
    Location:
    On planet
    No, SPL, dBu, dBV, dBr, volts etc all follow 20 log (ratio). 10 log (ratio) is for power. None of the measurements discussed are power.
     

Share This Page