Schiit Yggdrasil Less is More (and MIL and OG) Impressions + Measurements

Discussion in 'Digital: DACs, USB converters, decrapifiers' started by purr1n, Sep 9, 2021.

?

Should SBAF get a loaner Yggdrasil Less Is More?

  1. Yes, please!

    75.4%
  2. Only if there isn't anything else more interesting

    11.5%
  3. No, I would prefer for a loaner an overpriced planar that looks like it's from House Harkonnen

    6.2%
  4. Save up for something better

    6.9%
  1. Azimuth

    Azimuth FKA rtaylor76, Friend

    Pyrate
    Joined:
    Nov 9, 2015
    Likes Received:
    6,838
    Trophy Points:
    113
    Location:
    Nashville, TN
    Home Page:
    You might be finding the limits of the DSP filter. I wonder of any of the Theta stuff would show something similar.
     
  2. Empyah

    Empyah Facebook Friend

    Contributor
    Joined:
    Apr 22, 2020
    Likes Received:
    156
    Trophy Points:
    43
    Location:
    Earth
    • Like Like x 4
    • Epic Epic x 2
    • List
  3. Andre Y

    Andre Y Friend

    Pyrate
    Joined:
    Oct 4, 2015
    Likes Received:
    220
    Trophy Points:
    43
    Location:
    Southern California
    Yeah, for sure. With ultrasonics, there are all sorts of unintended consequences that are really system dependent, and it changes depending on which component in the chain too. For example, a DAC that doesn't filter aliases well like a NOS DAC or a DS DAC with tons of noise shaping will always be putting energy out in the ultrasonic range, possibly affecting anything downstream, but an amp that has a peak due to slightly hairy stability at the top of its bandpass (like some of the later Mark Levinson designs) might act pretty normally until it gets fed signal near that peak, and then what it does depends on its particular design. A lot of it depends on the nature of the nonlinearity on how it might be provoked into audibility.

    Also, I misspoke earlier: if THD+N above 10k is caused by nonlinearities, like in an analog amp, then there could be audible consequences because it can modulate stuff down frequency all by itself.
     
  4. 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's the actual signal.
     
  5. jexby

    jexby Posole Prince

    Staff Member Pyrate Contributor
    Joined:
    Sep 28, 2015
    Likes Received:
    8,145
    Trophy Points:
    113
    Location:
    Exit stage left....
    I mean, uh yah. almost posted the same reply an hour ago, but waited to see if a finger pointing or noob firestorm broke out. good on ya SBAF peoples for not.
     
  6. Empyah

    Empyah Facebook Friend

    Contributor
    Joined:
    Apr 22, 2020
    Likes Received:
    156
    Trophy Points:
    43
    Location:
    Earth
    T'was the hill I picked to die on.
     
    Last edited: Sep 21, 2021
  7. Josh83

    Josh83 Friend

    Pyrate
    Joined:
    Mar 1, 2017
    Likes Received:
    1,375
    Trophy Points:
    93
    Location:
    Charlottesville
    I had to do the mental calculations as to whether it was dry humor and concluded it was.
     
  8. Scott Kramer

    Scott Kramer Friend

    Pyrate
    Joined:
    May 3, 2016
    Likes Received:
    1,446
    Trophy Points:
    93
    Great stuff @GoldenOne! You mention the hold-and-sample is baked into the chip... however this can be adjusted, and even turned off! See the diyaudio thread post 12. Probably part of the tuning/experimenting Schiit has already done.

    source: https://www.diyaudio.com/forums/vendor-s-bazaar/349809-dac11001-20-bit-2r-precision-dac.html

    Cool how you can inject jitter for testing... ever get the "shit source" LED to light?

    Also surprising MiL might be x4 oversampling... maybe they could not tame the beast?

    PS. lol #17 heroic de-glitcher; and see #30,40 for rising distortion with freq.
     
    Last edited: Sep 21, 2021
  9. GoldenOne

    GoldenOne Friend

    Pyrate
    Joined:
    Jul 17, 2020
    Likes Received:
    1,049
    Trophy Points:
    93
    Location:
    England
    10ns 1khz jitter did not get the 'shit source' light to turn on. I'll try with some other options later though. Typically random/noise jitter is much harder for a PLL to retain lock than something with purely deterministic jitter so I'll see what happens with some further testing
     
  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
    Could be that more than x4 oversampling presents issues for the track and hold. Less time to sample and wait for glitch to settle.

    We know this track and hold can be turned off on the DAC11001, but I'm betting much worse AmirNADS dash results, which would defeat the point of the MIL entry into the "Thunderdome".

    I'm giving some thought as to why some DAC designers back in the day eschewed sample and hold. Makes more sense now. How the heck does one determine what is the "true" output value of a glitchy signal?
     
    Last edited: Sep 22, 2021
  11. schneller

    schneller Acquaintance

    Joined:
    Nov 19, 2016
    Likes Received:
    52
    Trophy Points:
    18
    Location:
    Washington, DC USA
    Can anyone who owns BF2 and LIM offer in-depth commentary about the differences between the two? I seem to have read a few comments so far that would lead me to believe that BF2, at least in some respects, holds its own against LIM.
     
  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
    From a pragmatic point of view, BF2 and LIM are close enough in technicalities where if the BF2 synergies well with your existing setup, then stick with it. This is what I have recommended to people in PM or who have called me or texted me. I still stand by this recommendation.

    The BF will be overall a bit thicker sounding. There's a just a trace of prickly highs, part of the lineage of parts trickle down from Yggdrasil A2/OG, that only the most sensitive nitpicky people will notice. I also think the BF2 may slam a bit more in the subbass, but just by a tiny bit. LIM a bit more midbass. This is all relative. Overall dynamics goes to LIM - stronger attacks - but only when needed as opposed to being on almost all the time compared to A2/OG. The BF in contrast is a bit more polite. The BF along with Modi-Multibit are a step towards the Classic / Vintage R2R sound because of this.

    Now if you are an obsessed audiophile truly wanting the next level of performance knowing that diminishing returns will be in play, but also having an uncanny ability to notice the smallest imperfections much like a kimchi connoisseur*, then no, the BF2 does not hold its own against the LIM. The only reason it may is because of system synergy issues - bad synergy with LIM. Keep in mind that engagement/microdynamics/expressiveness and microdetail are good step up with the LIM.

    Also, Schiit, and especially Mike, won't release a $2k+ DAC where his $700 DAC "holds its own in some respects". He'd rather leave it on the rack as one of the things that he tried that didn't work. A $2k+ DAC better decisively beat that $700 DAC.

    On the other hand, we have the MIL. I blame Amir for that.

    P.S. That Mike suggested that he could not make a better DAC; that anything else would be a sidegrade to Yggdrasil AG/OG (confirmed by Jason with the "synergy" and "flavors" statement). This is quite revolutionary in the world of HiFi where it's upgrade until the DAC is $109,888. I bet a bunch of folks are going to be pissed off.

    --

    BTW, it's my understanding that the LIM is out of stock already. I'm gonna say it: I told ya. I will also bet that there's a good possibility it could be more than 4-6 weeks. It's not GPUs. If places like TMSC can't make enough Apple CPUs or Ndivia CPUs, then they sure aren't gonna make enough the lower end shit that goes into stuff like cars. Anyway, the point which I wanted to get to is this: there seems to be lots and lots of MILs available.

    C'mon, no one wants an high AmirNADs R2R DAC?

    *I'm stuck in South Texas. Can someone recommend me a good kimchi that I can buy on my next trip to the Houston H-mart? I like the stuff with anchovies. There was a tub like this that a Korean lady brought to us when my wife just had her first kid when we lived in Irvine. Best stuff ever outside of a Korean grandma's special collections. But I could not remember what it was. It could have been directly imported from Korea. Currently ordering from Mama Kim's which is solid B+. But it's not A or S tier stuff. I've made A tier stuff myself, but PIA.
     
    • Like Like x 14
    • Epic Epic x 5
    • List
    Last edited: Sep 22, 2021
  13. Soundmancan

    Soundmancan New

    Joined:
    Jan 17, 2021
    Likes Received:
    26
    Trophy Points:
    3
    Location:
    united states
    I would defer to Purr1n's description of the Bifrost 2 vs Lim on page one of this thread. Personally, I had the Bifrost 2 just prior to the Lim and can tell you that the Lim is far superior (as it should be for 2k+). The stage on the Yggdrasil is bigger, bass is superior, textures, clarity and timbre etc is just better and more natural sounding. The Bifrost 2 is a great Dac as well but it is just one of those things where I am aware of and have to accept its shortcomings while listening to it. The Yggdrasil, to me, just isn't lacking for my preferences. Hope this helps. Apologies, Purr1n responded before I clicked on save so there you go.
     
  14. sp33ls

    sp33ls Friend

    Pyrate Contributor
    Joined:
    Oct 24, 2018
    Likes Received:
    169
    Trophy Points:
    33
    Location:
    PNW
    I've not A/B my LiM and BF2, but my impressions align completely with Marv's ^

    That being said, the BF2 is a great value if you prefer the R2R sound, IMO. If buying an Yggdrasil means only Ramen for a month, then go with the BF2.

    And this is coming form someone who loves both the LiM and Ramen.

    Interesting note on the the Livestream today, Jason tried to briefly describe a bit more why he may have a preference for the MiL. TL;DR, maybe the more "disjointed" presentation may align with his music preferences. Also, mentioned when get gets back to CA, he hopes to explore the MiL a bit more with Mike.
     
  15. Wojciech Czupta

    Wojciech Czupta New

    Joined:
    Sep 22, 2021
    Likes Received:
    37
    Trophy Points:
    13
    Location:
    Poland
    Having Yggdrasil OG (Analog2) feed via EBU, outputting to Freya+ and then to very transparent, fast and full range studio monitors (ATC SCM100 active) would you consider an upgrade to LIM or MIL? Would that make sense at all? My benchmark is Dave&Mscaler which was crazy good but also crazy expensive (please don't mistake with Dave alone which I didn't like).
     
    Last edited: Sep 22, 2021
  16. sp33ls

    sp33ls Friend

    Pyrate Contributor
    Joined:
    Oct 24, 2018
    Likes Received:
    169
    Trophy Points:
    33
    Location:
    PNW
    Most would consider them to be "sidegrades." I think this is why Jason decided to use the term "flavors" to differentiate.

    I would browse through these impressions and see if one of the flavors would better suit you -- or just stick with the OG if you're happy with your setup.
     
  17. sp33ls

    sp33ls Friend

    Pyrate Contributor
    Joined:
    Oct 24, 2018
    Likes Received:
    169
    Trophy Points:
    33
    Location:
    PNW
    Someone mentioned the Holo Spring already, but I'd be curious to know how the LiM stacks up against the Holo May KTE. I was originally thinking I'd try and see if I could come across one second-hand someday, but after using the LiM, I'm left wondering why bother..?

    Not saying the LiM is superior to Holo May (probably not?), but that the LiM is making me wonder "what more do you really need to enjoy the the music..?"
     
  18. rhythmdevils

    rhythmdevils MOT: rhythmdevils audio

    Pyrate
    Joined:
    Apr 15, 2020
    Likes Received:
    12,239
    Trophy Points:
    113
    Location:
    Bay Area, CA
    Home Page:
    Update:

    AES sounds better than BNC out of the Pi2AES. When I switched the LIM sounded more resolving.

    I now have my Liquid Gold plugged straight into the XLR outs of the LIM and my 789 plugged into the RCA output with the switcher in between. It's hard to tell wtih the LGX because the XLR inputs actually had to burn in. No joke, I wasn't expecting this, but the amp sounded like it did when it was new. So it will be hard to comment here. But my og Solaris our of the 789 oiut of the SE outputs with no iso max transoformers in the path sound noticeably better. So I can verify what @purr1n said about the SE outputs on the LIM sounding better than XLR -> transformer.

    This is kind of amazing because it opens up 3 outputs on the Yggdrasil. I can buy so many unnecessary amps now. :eek:

    It's also nice knowing the Liquid gold is just straight XLR out of the LIM into the amp. All balanced and nothing in between. I could now technically have multiple balanced amps. Not that I.need or want them. One switcher for SE and another swifcher for XLR.
     
  19. earnmyturns

    earnmyturns Smartest friend

    Pyrate
    Joined:
    Sep 25, 2016
    Likes Received:
    3,218
    Trophy Points:
    113
    Location:
    Palo Alto
    Home Page:
    What follows is a bit of extrapolation from OG A2 and Spring 2 KTE: Holo OS mode is meh. NOS wins over OG in coherence, tonal density, lack of upper range harshness, blackground, but at the cost of some lower-mid smearing for 44.1 or 48 material. If you only listen to 88.2+ material, or are willing to play with external upsampling, the Holo DACs are worth exploring. Otherwise...
     
  20. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    Never use a hair dryer. It produces concentrated heat which is good for finding thermal problems of specific circuit components during prototype phase. Concentrated heat will unevenly heat a finished unit probably creating performance problems. There is a reason we use thermally controlled chambers to verify product performance. At my corporate job we have several thermally controlled chambers. Room heater might do a little better but unless you are willing to monitor an array of thermocouples attached to the component for even heat distribution this method again may create performance problems.

    technical measurements for 3 hrs and 408 hrs power on time found here
    Inferred jitter measurements were made for both sets. Repeated here for convenience:

    yggdrasil (very first iteration) inferred jitter 3 hrs power on time
    20150430 Yggdrasil inferred jitter.PNG

    yggdrasil (very first iteration) inferred jitter 408 hrs power on time
    20150517 Yggdrasil inferred jitter.PNG
     
    • Epic Epic x 13
    • Like Like x 11
    • List

Share This Page