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. Soundmancan

    Soundmancan New

    Joined:
    Jan 17, 2021
    Likes Received:
    26
    Trophy Points:
    3
    Location:
    united states
    Well, I went out on a limb and bought the LIM! Should be here tomorrow. After sufficient warm up time, I will post some thoughts on it. I will be using my Jotunheim 2 for amplification (Vali 2+ as well). I currently have the bifrost 2 (which I am quite fond of) and am quite curious how it stacks up against the Lim. I have previously had the Gungnir Multibit w/Unison which was nice but found it to be wider and more upfront sounding than the Bifrost. The Bifrost I felt staged better for my preferences than the Gungnir and (from what I have read) was similar (but not on the same level) as the Yggdrasil. I have always wanted an Yggdrasil but the fact that the SE outputs were gimped was off putting to me because I use a mixture of gear and would (obviously) prefer everything being as equal as possible. And heck, I can always buy the OG boards down the line should I feel the need.
     
  2. Soundmancan

    Soundmancan New

    Joined:
    Jan 17, 2021
    Likes Received:
    26
    Trophy Points:
    3
    Location:
    united states
    This is a great question and one I would defer to Marv to answer correctly. My limited understanding is that the Yggdrasil main board for all Yggys are the same with the difference being the actual DAC cards. So if this is indeed correct, then I would assume that the Yggdrasil OG still has the gimped SE outputs because it still the same Yggdrasil of old. Only the newer Yggys with the new DAC cards would have the better SE implementation.
     
  3. roshambo123

    roshambo123 Friend

    Pyrate Contributor
    Joined:
    May 26, 2018
    Likes Received:
    2,763
    Trophy Points:
    93
    Location:
    Los Angeles
    Just a thought on the Yggdrasil LIM loaner. You'd probably need a longer than average loan period per person due to the warm-up time.
     
  4. blubliss

    blubliss Friend

    Pyrate
    Joined:
    Oct 4, 2015
    Likes Received:
    128
    Trophy Points:
    33
    Location:
    Los Angeles, CA
    I've got the LIM on hand, going to install, just a bit confused. I assume there is no actual flashing of firmware needed, that changing the eprom on the DSP board is the firmware update? Can someone confirm?
     
  5. internethandle

    internethandle Almost "Made"

    Contributor
    Joined:
    Sep 23, 2017
    Likes Received:
    785
    Trophy Points:
    93
    Location:
    long beach, ca
    Yes, that should be correct, at least per my understanding. When the GS came out, Schiit shipped it with the A2 firmware, and offered it in the form of a shippable EPROM to any A1 owners who did not want to spring for the full A2 upgrade but wanted the updated firmware. There's a guy on YouTube with a video detailing his doing such a EPROM swap on an A1, should be the same for the LIM EPROM:



    You can also just use Marv's (probably better) guide to updating to Gen 5 on here if you need help with the screws.

    Not too big of a deal if you're used to this kind of work (Marv is right that the actual alignment of the pins on the two LIM boards should be more difficult/requires more caution). The BF2, as part of Autonomy, introduced a SD card-based flashing system, since I don't think Gungnir or any of the other DACs in the line-up have an EPROM that is swappable in the DSP section, only Yggdrasil, so any firmware update would have had to have been done by sending it back to Schiit.

    Edit: I may be wrong about the EPROM on previous multibit boards for the other DACs in the lineup. After having a look at a picture of the Gungnir MB board, it looks like there may be one there in the DSP section near the SHARC chip. Regardless, never has had a user-servicable (or otherwise) update, to my knowledge.
     
    Last edited: Sep 13, 2021
  6. blubliss

    blubliss Friend

    Pyrate
    Joined:
    Oct 4, 2015
    Likes Received:
    128
    Trophy Points:
    33
    Location:
    Los Angeles, CA
    Thanks, got it all installed and it's making music.
     
  7. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,780
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    @sphinxvc asked me about "engagement" which I took to mean as combination of microdynamics, immediacy, PRAT, liveliness, expressiveness, etc. It is one of the most important factors for me. Examples of DACs with good engagement? The MSB DACs, Rockna Wavedream, Soekris DAC2541, the Massdrop RDAC. It's the quality that keeps me up, not wanting to put the headphones down, or while listening to speakings, falling asleep on the couch drooling. This of course assuming that all other aspects are acceptable or worked around. People who audition gear will immediately notice tonal signature, timbre, and transients. That's what I call a technical listen for the outer qualities - it's trying too hard. I tried to listen for this engagement fact, past these outer qualities, because these outer qualities can always be worked around. Lack of engagement cannot. One cannot find synergizing components to restore what's lost in class D amps, 150db feedback in opamps, etc.

    This is why I love vinyl, especially stuff on a good pressing produced on an all analog chain, even on a record in mediocre shape. A $500 record player with such a record gets us more expressiveness than any DAC, albeit that at this price point there will be some serious issues in bass quality, hum, and maybe IGD. And crappy SNR, surface noise, ticks, pops, no matter what.
     
    Last edited: Sep 13, 2021
  8. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,780
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Again, not enough time so I didn't want to put it on the charts. (IFrom experience, I'm usually never wrong, but I do want to make extra sure - like maybe it needs a week of warm-up).

    More Is Less (MIL): I wouldn't say it rates very low on engagement. Maybe a tick below the rest, at Bifrost 2 level, which is no slouch. We are splitting hairs as we should because this is a $2k+ product. Strangely, I didn't hear any better retrieval of plankton. To hell with measurements. I blame it on the sample and hold thing. The 11001 is new chip, so perhaps it's secrets haven't been unlocked. Feedback has been given to the brain trust at Schiit. They have some ideas (could be implemented in firmware), but as with these things, we won't know until they are tried.

    Note DNL is better on the DAC11001A, but INL is much better (x4) on the AD5791. DNL is the error "drift" from bitcode to bitcode. INL is the overall error of the output to the bitcode.
     
    Last edited: Sep 13, 2021
  9. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,780
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    With a 24-bit input, the LIM doesn't truncate. It dithers. With x4 DAC8812 per channel (each DAC8812 has x2), it's potentially an 18-bit part. With dither, it's hitting 23-bits with ease (1 bit is 6db).
    [​IMG]

    However, starting around -110db, the output becomes less linear and not according to what it should be. Past -120db, we are off +/-0.5db.
    upload_2021-9-13_19-20-42.png

    In short, it will resolve to a bit more than 23-bits, only very accurately (with respect to output level) to just past 18-bits, and somewhat accurately (+/- 0.5db) to 20-bits. I doubt any hires content truly has any content past 20-bits. (If there is a microphone involved, for sure no way).
     
    Last edited: Sep 13, 2021
  10. Beefy

    Beefy Friend

    Pyrate
    Joined:
    Apr 10, 2021
    Likes Received:
    1,738
    Trophy Points:
    93
    Location:
    Canada
    I certainly agree. My only theoretical concern with less 'real' bits, is that it does somewhat limit headroom if you are used to using digital volume control. But even then, if you can get down to -110dB before problems hit, it really is completely theoretical. I can't imagine any real scenario where you might attenuate -60dB, and still be listening loud enough that you expect to hear 60dB of dynamic range below that.
     
  11. Justin S

    Justin S Friend

    Pyrate Contributor
    Joined:
    Mar 14, 2018
    Likes Received:
    1,506
    Trophy Points:
    93
    Location:
    Toronto
    Thanks. I do have some 48/24 stuff, but am mostly streaming redbook FLACs from Tidal. I do occasionally play with HQPlayer and the developer recommends dithering to the bit depth of the DAC - your explanation and charts are helpful. Some of the HQPlayer filters can work very nicely with the 2541. I don't imagine I will do much of the with the LIM, on account of the awesome Schiit filter.
     
  12. Dzerh

    Dzerh Friend

    Pyrate
    Joined:
    Apr 1, 2018
    Likes Received:
    425
    Trophy Points:
    63
    Location:
    San Ramon, CA, US
    I'd add that technically proficient DAC with low engagement factor is not necessary an evil thing, - think about office setup for background listening.
     
  13. sp33ls

    sp33ls Friend

    Pyrate Contributor
    Joined:
    Oct 24, 2018
    Likes Received:
    169
    Trophy Points:
    33
    Location:
    PNW
    True, but a high end DAC for background listening? That's some FU money right there, and I love it... ;)

    Makes me wonder how long before some folks start opamps rolling the LiM... throwing the likes of an OPA828 or discrete opamps for the I/V.
     
  14. leafy

    leafy Facebook Friend

    Contributor
    Joined:
    Jul 18, 2017
    Likes Received:
    105
    Trophy Points:
    33
    Location:
    Taiwan
    I have a system with Yggdrasil GS/Freya S/Vidar/LS50. I use the gain control on Freya S to according the level of engagement. When sitting on the couch with the intent to watch TV or listen to music, it's the 4x switch. Turning it to 1x or passive really puts the system into a background listening mode that doesn't attract that much attention.
     
  15. MetalStef84

    MetalStef84 New

    Joined:
    Nov 2, 2018
    Likes Received:
    64
    Trophy Points:
    13
    Location:
    Italy
    So yesterday I took the little experiment to hook both SE and BAL out of my Yggdrasil A2 to the respective inputs of my bryston . The question in my mind ofc was: can I hear this "gimped SE thing?" (and if yes it could have yield to "is it better for me an Yggdrasil like the LIM with no gimped SE?"). Reason is I would like to use one Yggdrasil for the BHA-1 (XLR) + tube amp (SE) + speaker monitor (SE... that due to paranoia now runs out of jot pre).

    Hypothesis are:
    1) this connection does not draw away anything to the BAL configuration and with the volume matching done with the "low/high" gain switch (that's 6dB and should be the 20Log(2)=6dB missing from SE to BAL x2 effect, but I felt I had to very slightly adjust volume still, using windows for better precision since numbers are numbers, but knob angle are not easy to replicate consistently)
    2) (that is more or less a reiteration of 1) Yggdrasil A2 can drive 10k // 15k input of bha-1 fine. I don't know if the Yggdrasil output stages are independent/buffered or not (sry). I didn't want to connect and disconnect many times leaving floating RCAs...
    3) there are no f****d up crosstalk between the bha-1 input stage to screw things up.

    Results: both with quick A/Bs and with longer listenings, I could not find any difference. I listened to pure acoustic music focusing on background, timbre and reverberation/plankton/details and some pop/rock focusing on slam/tone/positioning... Sometime I did felt the SE "rounder", but only when I forgot to do the fine volume matching and could have been placebo.

    So question is: when you talk about "gimped SE", while I got the "hardware / summing" part clear (at least in my imagination), I don't know where to focus to spot the difference. In the convert-2 thread @purr1n talked about "blanket on" and in Yggdrasil A2 I could not find a better description. So when you say gimped, the effect is what? Sennheiser veil? Rounder transient (slew rate of the Yggdrasil output driver?!)? Slower? Do you know some tracks that exhibit this more explicitly? Some people seems to clearly hear this, as @rlow (page 30 on Yggdrasil A2 thread) so I suspect I am missing something... or maybe I am lucky enough that my gear is now better than my hearing and yeah, I am done being worried xD (as mentioned anyway by @atomicbob and @purr1n, saying it is the last % of performance addressed but still, with all my effort, I cannot have a grip on it).
    PS: hope this question makes some sense in this thread. Yggdrasil A2 is closed but since the LIM has this feature of non gimped SE, I hope any answer to this can make sense here. Sorry if this is not the case.
     
  16. crenca

    crenca Friend

    Pyrate
    Joined:
    May 26, 2017
    Likes Received:
    3,830
    Trophy Points:
    113
    Location:
    Southern New Mexico
    It's a tough thing to hear in my experience. It's been a while since I A/B'd it, and I have the Gungnir A2, but I heard it as a slight (and I do mean slight) loss of the DAC's incisiveness and "liveliness". Perhaps with a different DAC or a different set of preferences/priorities it would not matter, but since this is the very quality I am after with these Schiit multibit DACs it does. Since I run SE amp and preamp, what I did get a balanced>SE transformer. Thing is that these transformers color the sound (slightly) in there own way...but the trade off is one that works for me in my system.

    I would say don't worry about it - your struggling to hear it in your system which makes it "academic", whereas a switch to an LIM and the subsequent change in sound would not be theoretical.
     
  17. rlow

    rlow A happy woofer

    Pyrate Contributor
    Joined:
    Jul 18, 2017
    Likes Received:
    7,738
    Trophy Points:
    113
    Location:
    Canada
    I came to realize later that some of the differences I was hearing was the diffs between preamps, not the outputs from Yggdrasil. It’s a very difficult thing to compare both sets of outputs in an apples to apples way unless you have an amp with both inputs that are equally transparent and don’t do any conversion that ends up homogenizing or shifting things on the way out to your your speakers/headphones. While I trust @purr1n’s ears (definitely) my personal experience subsequent to this was that I don’t detect much difference between the two outputs, at least in the chains I have tried it with. This may simply be a limitation of the setups I’ve tried them with however.

    I had said somewhere later on that I feel this diff between SE and Bal is overblown wrt Yggdrasil. IMO if you can’t hear it, you’re good man, honestly don’t worry about, I stopped a while ago.
     
    Last edited: Sep 14, 2021
  18. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,780
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Yup. I wouldn't worry about it then. Like what I said. The difference between SE and BAL on the A2/OG is less than what people say or more than what people say. It will also be dependent upon the SE and BAL circuits of the integrated amp, preamp, or headamp that you are using. Lots of variables.

    BHA-1 I believe has a summing circuit at the XLR inputs. The pot isn't a four gang pot, so I'm assuming the summing happens before the volume control. Bryston is very careful not to say "fully balanced through" and only mentions balanced in the context of "balanced output".

    Jotunhiem and BHA-1 are convenient as preamps for a secondary function, but they will not be as transparent as dedicated preamps as their primary function is to drive headphone loads which are more demanding than line loads.
     
  19. M3NTAL

    M3NTAL Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    1,697
    Trophy Points:
    113
    Location:
    Arizona
    My Sonic Frontiers isn't shaking in its boots, but I am! Exciting having a modular DAC in this stage of the world. I've never taken up a Schiit DAC because every system I've heard has been their DAC+ their amp and it is always too in my face and seems to go for a macro vs. micro type of presentation.

    My SFD-1 in-between the Jotunheim-A might be why I don't find the LCD-R too forward, but for sure leaning towards that Schiit sound I've heard before.

    I prefer an effortless type of engagement. It is up to me to decide how deep I want to dive, but also not miss out on the macro scale of the event that was recorded.

    I like the idea of both the LiM and the Soekeris collab. I doubt I'd be able to play out the gate, but hope to eventually get a taste.
     
  20. roshambo123

    roshambo123 Friend

    Pyrate Contributor
    Joined:
    May 26, 2018
    Likes Received:
    2,763
    Trophy Points:
    93
    Location:
    Los Angeles
    On the GS, I heard the difference between the SE and BAL (at least using BAL+Jensen, or SE vs BAL on MJ2, both of which have some confound) is the SE was harmonically richer, more grey, flatter dynamics, and less plankton than pure BAL or Jensen'd BAL. I felt like SE made GS closer to a G1 sound and I actually liked it better than the BAL+Jensen combo for SE amps. GS is quite detailed and incisive and I like having the option to furry it up a little sometimes using the SE out.
     

Share This Page