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!

    75.6%
  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

    6.1%
  4. Save up for something better

    6.9%
  1. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    90,849
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Less congested and less dark over time. The changes are front loaded. More happens in the first few hours than last hour. You will get continued changes in the next 24 hours, but not much. I wanted to repeat this because the LIM is rather dark sounding when cold.

    Practically every R2R DACs sounds congested when cold. Who knows if this is because the resistor ladder needs to heat up and stabilize (we see from the datasheets that the DNL/INL can be all over the place depending upon temperature) or that maybe that the same is needed for the oscillators. R2R is supposedly more prone to jitter than delta sigma.
     
  2. Scott Kramer

    Scott Kramer Friend

    Pyrate
    Joined:
    May 3, 2016
    Likes Received:
    1,459
    Trophy Points:
    93
    IMG_0042.jpg
    Screen Shot 2021-09-17 at 4.02.00 PM.jpg
    @purr1n did you notice those TI DAC8812's have 2 DAC channels each? 4 total per card (8 per Yggdrasil)? I wonder how they use those...
     
    Last edited: Sep 19, 2021
  3. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    90,849
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Yeah. I just noticed it. I thought it was x4 chips, but it's x2 because each chip has x2 channels. I want to go back and redo the THD vs level measurement to verify. I have a hard time believing that Mike and Dave would do the gluing of x2 DAC ladders vertically (they've never done that in the past - it had always been a lateral approach). It could be that the DAC8812 does the glue internally, and then they are used x2 for averaging. Who knows. It sounds good. The 8812 is weird though with it being a multiplying DAC.
     
  4. sp33ls

    sp33ls Friend

    Pyrate Contributor
    Joined:
    Oct 24, 2018
    Likes Received:
    171
    Trophy Points:
    33
    Location:
    PNW
    I'd love to hear more from Mike on the nitty gritty during the next Schitt Livestream.
     
  5. Scott Kramer

    Scott Kramer Friend

    Pyrate
    Joined:
    May 3, 2016
    Likes Received:
    1,459
    Trophy Points:
    93
    Haha, agree... left a very technical post on Mike's head-fi thread (something about spine-tingling).

    Need to read through the data-sheet, maybe there's some funky mode utilizing both channels.

    @rlow: Remember when Mike talked about a cheap DAC chip that sounded way too phucking[sic] good? Think LiM is that one?
     
    Last edited: Sep 17, 2021
  6. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    90,849
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    The best way is to educate yourself as much as possible and post good questions. Mike ain't gonna volunteer his secrets. The guys at Schiit do read SBAF. The trick it so make it irresistible so that they will answer the questions. More often they will drop hints. The more you do to connect the dots, the more they will reward you.

    It wasn't that long ago that I wanted to figure out what Mike was cooking up. This was pre Yggdrasil. MIke was talking about military / medical spec AD and DA chips. You know, the kind of chips today that enable a Hellfire R9X to drop directly in the driver's seat from a Reaper drone up a few miles high, shredding a well defined star pattern into the roof of the car (the R9X doesn't carry an explosive, but instead extends six blades before impact - you know, so the terrorist's wife sitting next to him and the kids in the back don't get killed).

    I found out Mike made a temperature controlled AD converter for Mobile Fidelity in its heyday that replaced a prior ladder type AD converter. I got whatever Mobile Fidelity CDs I could before and after Mike's converter was used at MF. Mike's AD converter was more resolving than the prior stuff. The AD converter that eventually replaced Mike's was a one-bit, because DSD was all the rage. Guess what, the highs sounded like ass on that one-bit converter. I'd like to say that Mobile Fidelity made the wrong decision by going to an AD which sounded not as as good; but heck, SACD was all the rage then (not too different from MQA - history simply repeats itself).

    I also knew that Theta DACs were all the rage in the early 90s. Back then I was out of school and couldn't afford much, so I bought Theta's entry level DAC, the Cobalt. At meets ten years ago, people old enough to remember Theta would remark that Theta was the bomb, that they made the best DACs, among others such as Wadia, etc. As an aside, I find out later that the Cobalt was designed by Jason under MIke. Funny how these things happen. I auditioned quite a lot of stuff too before going with the Cobalt. (There were these things called physical stores back then).

    I put two and two together and realized that Mike was behind the Theta Gen V. I picked up a very well used Gen V to find out more what Mike's DACs were all about - since Mike kept hinting at what was to come, the Yggdrasil. The Gen V DAC was nuts. Three different sections. Power supply, balls to the wall output stage that melted the PCB after five years. And a DSP section composed of three, not one, but three Motorola 68000 CPUs (used in Apple computers at that time). The most spooky aspect of the Gen V which I remarked to Mike was its imaging, localization, depth discernment and ability for a listener to get a sense of the venue (this is with speakers, headphones don't do soundstage or imaging.) That's the megacomboburrito filter. That's what all those CPUs were for. Fortunately we don't need three Macintosh CPUs to do the same processing today.

    Anyway, some investigative reporting always gets the conversation going. The Schiit guys did not feed me info. They gave me hints for sure, but I had to do the work.

    Of course one can always do the easy thing: play a 1000 cycle per second sine wave and see what the SINAD panel reads. My guess Mike will probably look at you funny for a few seconds, and then turn around and slowly walk away.
     
    • Like Like x 17
    • Epic Epic x 1
    • List
    Last edited: Sep 17, 2021
  7. Kolozub

    Kolozub New

    Contributor
    Joined:
    Jan 30, 2021
    Likes Received:
    50
    Trophy Points:
    13
    Location:
    Slovakia - Krásno nad Kysucou
    I think it was delta sigma proto that sounded phucking good to him, but i am not sure :)
     
  8. futbutts

    futbutts Friend

    Pyrate
    Joined:
    May 15, 2018
    Likes Received:
    471
    Trophy Points:
    63
    Location:
    Shermie Oaks, CA
    Yeah I'd add that today's improvements seem to be an increased sense of airiness without loss of body or impact. Microplankton is still well-defined but a tad more natural sounding too. And yeah, Yggdrasil LIM over 10 hours was definitely a bigger change than the Gungnir displayed in even the first few days.
     
  9. Merrick

    Merrick A lidless ear

    Pyrate
    Joined:
    Jan 6, 2016
    Likes Received:
    12,632
    Trophy Points:
    113
    Location:
    Portland, OR
    The GAIN I system that Mike designed for MoFi was f'ing awesome. Almost any GAIN I mastered CD is worth owning, and are among the best sounding MoFi releases. GAIN II uses DSD and many MoFi SACDs do sound very good but they lost the magic that GAIN I had. GAIN I discs on a Moffat designed DAC sound extremely holographic and realistic.
     
  10. Josh83

    Josh83 Friend

    Pyrate
    Joined:
    Mar 1, 2017
    Likes Received:
    1,388
    Trophy Points:
    93
    Location:
    Charlottesville
    Thanks for the fascinating information, @Marvey and @Merrick. I had no idea GAIN I was a Moffat-designed ADC.
     
  11. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,802
    Trophy Points:
    113
    Location:
    Winnipeg
    Is there perhaps a master list of the GAIN I and II recordings somewhere?
     
  12. Merrick

    Merrick A lidless ear

    Pyrate
    Joined:
    Jan 6, 2016
    Likes Received:
    12,632
    Trophy Points:
    113
    Location:
    Portland, OR
  13. MellowVelo

    MellowVelo Friend

    Pyrate Contributor
    Joined:
    Sep 5, 2016
    Likes Received:
    937
    Trophy Points:
    93
    Location:
    Chicago
    This is really helpful. I always associated Ultradisc and Ultradisc II with Gain and Gain 2, respectively. Based on your findings, it appears that several of the Ultradisc II entries still used the original Gain system. This actually explains something I’ve wondered about. A few years ago, I purchased “Disraeli Gears” by Cream on both Ultradisc and Ultradisc II. When I compared them with AccurateRip, they were identical. So, it makes sense now that the discs had the same original Gain mastering; they were simply pressed in different locations (Ultradisc in Japan, and Ultradisc II in the United States).
     
  14. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    90,849
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    The MFSL GAIN I Era:

    START: Dec 1994 UDCD 593 Muddy Waters Folk Singer
    END: Nov 1998 UDCD 734 Jehtro Tull songs From The Wood​

    It's a chance to hear a full A to D and D to A chain from Mike.

    Now the next thing to do is for us to convince Mike to make a multibit A to D. I think Jil didn't sell because it lacked basic pro feature sets of x2 microphone inputs and x2 BAL/TRS inputs.

    With more people working from home and a pro feature set, I bet a multibit AD version would sell like hotcakes. It wouldn't be as cheap as a MOTU M4, but I bet it would be a better microphone pre and A to D.
     
    Last edited: Sep 18, 2021
  15. GoodEnoughGear

    GoodEnoughGear Evil Dr. Shultz‎

    Pyrate
    Joined:
    Oct 25, 2015
    Likes Received:
    3,074
    Trophy Points:
    113
    Location:
    Cape Town, South Africa
    This could carve out a niche for influencers and Eilish-styled home producers too.
     
  16. Josh83

    Josh83 Friend

    Pyrate
    Joined:
    Mar 1, 2017
    Likes Received:
    1,388
    Trophy Points:
    93
    Location:
    Charlottesville
    Yeah, a bunch of MoFi CDs, like Gaucho and Dark Side, were released as Ultradisc I and II with identical transfers and mastering. The Ultradisc I/II designations don’t map onto the GAIN I/II designations.
     
  17. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,802
    Trophy Points:
    113
    Location:
    Winnipeg
    Copying Marv's info here so we have the start and end dates...

    Then I went through the link above from the stevehoffman forum and I tried cleaning up that list because all those spaces was bugging me... I'm sure someone who has basic programming skills can extract that into a spreadsheet or something. Note that the below list starts at 564, and Marv's dates start at 593.

     
    • Epic Epic x 10
    • Like Like x 3
    • List
    Last edited: Sep 19, 2021
  18. Gazny

    Gazny MOT: ETA Audio

    Pyrate Contributor
    Joined:
    May 11, 2020
    Likes Received:
    2,274
    Trophy Points:
    93
    Location:
    open sky
    I would dump what I got for to have a MM designed filtered AD/DA interface!
     
  19. sphinxvc

    sphinxvc Gear Master (retired)

    Staff Member Pyrate BWC
    Joined:
    Sep 28, 2015
    Likes Received:
    3,320
    Trophy Points:
    113
    I've had the LiM warming up here since this past Wednesday and the loaner 2541 since Thursday. Figured I would post some thoughts after I saw @GoodEnoughGear asking about a comparison.

    The chain, at the moment, is a Pi2AES going AES (for slow switching) or coax simultaneously (via splitter, for db-matched fast switching) to both DACs. Unfortunately, I don't have an XLR splitter handy for simultaneous output via AES. The DACs output to a Goldpoint SA2X (big shout out to @AdvanTech for getting this to me so quickly for the fast-switching comparison) and then to a pair of Dynaudio X14A monitors. This was the first time I heard 2541, so I spent some time with the head-out, my JAR650s and separately with the digital pre. I will post thoughts on these in the 2541 thread. This will just be about LiM vs. 2541 in an apples-to-apples comparison.

    Before that though, I want to say Marv does a really fantastic job comparing all these DACs for us. Even though I hadn't heard either of them, I would say his bar graphs and scores line up really well with the references I'm familiar with, and when the DACs came in, they too showed more or less the same relative differences that his charts showed. Also, describing DACs is by definition an exercise in magnifying subtleties, so it takes a careful touch not to overstate, and also to give people a sense of what they'll hear - whether they'll be listening closely for one particular sonic characteristic, or chilling and taking the whole thing in holistically. I'm just getting that out of the way so you know I'm not going to be trying nearly as hard. =P

    LiM versus previous Yggys
    The LiM is the best iteration of Yggdrasil I've heard yet (haven't heard MiL). I think the blackground takes this one over the top for me as a package. If A2 has more details, great, but those details are sitting in a murkier water, visible but submerged. The LiM may have fewer details but most of it is above the water line, dry; easy for the mind eye's to take in. I wouldn't dig into this analogy much further though, I've thought about it a little too much and it stops making sense beyond what I just wrote. If the MiL is a more tonally balanced/neutral version of LiM with just as good a blackground, that's probably the one I would gravitate to.

    LiM vs. 2541
    So regarding the 2541, the LiM seems to "dig" deeper in the bass and lower midrange. The Soekris is no slouch though, it in turn "digs" deeper in the treble and air/presence regions. The treble on the 2541 is fantastic, this DAC is how warm should be done. The overall profile seems to be a straight line going from low bass to treble with a linear downslope (subjectively speaking, of course). The LiM comparatively has more of a "knee" or Q to its slope starting in maybe the midrange. Sure, the LiM digs slightly deeper overall where music matters: anything rhythmic or groovy is naturally more present on the LiM because of its presentation, but every few tracks, it would get showed up by details on the high end that the Soekris easily brought forward. This makes the 2541 more genre-versatile for me, and that was pretty evident in my music choices over the past few days while listening to it. What's ironic about the LiM's tonality is that the burrito filter is known for its excellent spatial rendering, but that sense of space in a recording often comes from high frequency information as small wavelengths bounce around objects in a room. So like I said, I think maybe MiL is more in line with my sensibilities. That said, the 2541 works just fine. It's got such a deep blackground that the smoothness counteracts the grain party that is my JAR650 and the hiss party if connected to lower end studio monitors.

    Caveat on low bass
    There's one area where I'm going to decline to comment: low bass. The Dynaudio monitors have some 5.5" woofers and from previous experience integrating them with big woofers, there's certainly a difference. From a 2020 BF2 vs. Yggdrasil A2 vs. Wavelight comparison, I also know that the marked difference between the big boy DACs vs. the BF2 is that ability to produce low bass. I don't know if this is down to the no-compromise power supplies or what, so I really wouldn't be surprised if I found the LiM to extend further or slam harder than the 2541 on a fully capable speaker system. I personally will have no way to exploit this low bass advantage--if it even exists--for a few years, and so I don't plan to stress evaluating that much.

    Also, some of you may have seen my LiM listing up, this is not only because of the sonic preferences I talk about above, but because I need something lightweight and internationally voltage compatible to take with me as I rove for the foreseeable future. The digital pre, engagement and head-out don't hurt either.

    I'll have both of these with me for a bit, so if there are any specific questions from anyone or things you want me to listen for, let me know. Also, ugh, please don't call this a review.
     
    • Like Like x 26
    • Epic Epic x 7
    • List
  20. rhythmdevils

    rhythmdevils MOT: rhythmdevils audio

    Pyrate
    Joined:
    Apr 15, 2020
    Likes Received:
    12,584
    Trophy Points:
    113
    Location:
    Bay Area, CA
    Home Page:
    Great review @sphinxvc ! :p

    couldn’t help myself sorry
     

Share This Page