Life after Yggdrasil: Watering the Ash

Discussion in 'Digital: DACs, USB converters, decrapifiers' started by Torq, Mar 1, 2017.

Tags:
  1. winders

    winders boomer

    Banned
    Joined:
    Feb 13, 2017
    Likes Received:
    1,596
    Trophy Points:
    113
    Location:
    San Martin, CA
    Since I own a DirectStream running Red Cloud, used to own an Yggdrasil A1, and have not heard an Yggdrasil A2, I am very interested in reading something from you that discusses how those devices compare to each other.

    When I bought the DirectStream, I had no idea that the Yggdrasil A2 was coming. I felt the DirectStream sounded better than Yggdrasil A1 and liked the idea that it kept getting better with Ted Smith updating the FPGA code on a regular basis. I doubt that I will go back to Yggdrasil now but it would be good to hear what you have to say.
     
  2. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    So, details/specifics will still follow as I can get to them, but as a precursor to that … and given that the comparisons are indirect in places, here’s a summary of my current thinking …

    Musically, at this point, there’s really not a lot in it. It’s as much about signature preferences, sources, features and, to a certain extent … synergy. A bit of a cop-out perhaps, but that’s really how it is. Some blows are traded in terms of which unit wins at which specific technical aspect(s) – and that might swing things one way or another as well for different individuals, but overall, they’re all very close – just different.

    I really think it’s necessary at this level (and recommended at any level) to be comparing, or at least individually auditioning, these units in your own systems.

    But …

    Itend to think that Yggdrasil A2 and DAVE are generally at their best with Red Book content (indeed, DAVEs strength, in particular, for me is with 16/44.1 PCM sources) and are slightly ahead of the DirectStream here. This was definitely the case where the DirectStream was running on “Torreys”; more time with “Red Cloud” in a direct comparison might shift things a bit, it’s hard to say.

    Red Cloud definitely closes the gap, though (and I’d put it ahead of the original Yggdrasil in that form). And the Red Cloud DirectStream is typically ahead when playing native DSD material. However, if you convert PCM to DSD and feed the PCM to either A2 or DAVE, and the converted DSD to the DirectStream, then Yggdrasil and DAVE will generally sound the best.

    Not a lot to go on, I know.

    But consider that we’re talking about DACs that, among all those I’ve heard are performing in the 97-100 out of 100 levels of performance anyway. We just might not agree on which one gets which score, on which aspect of evaluation.

    For me, right now I’ll be keeping an Yggdrasil Analog 2 and DAVE. And I'll spending some more time to properly evaluating “Red Cloud”. And, of course, be paying keen attention to whatever PS Audio do next for the DSD units in terms of firmware.

    ...

    Now … if you let the value elephant in the room, well, things become very focused and rather quickly at that …
     
  3. winders

    winders boomer

    Banned
    Joined:
    Feb 13, 2017
    Likes Received:
    1,596
    Trophy Points:
    113
    Location:
    San Martin, CA
    @Torq

    I don’t convert PCM to DSD. I let the DirectStream do the conversion internally. I did so at first because the Bridge II card supports only DSD64. But I did some testing and I think PCM content sounds better out of he DirectStream if I feed it at its actual sample rate. Have you tried that with PCM content?
     
  4. MyPetSasquatch

    MyPetSasquatch Friend

    Pyrate
    Joined:
    Dec 1, 2015
    Likes Received:
    347
    Trophy Points:
    63
    Location:
    Sasquatch Country
    I actually just picked up a DS a couple weeks ago to try it out, so count me as the third. Ironically I've never heard the Yggdrasil (either version) but I've owned a Gungnir Multibit, Pavane, Onyx and now the DirectStream as the last few DACs I've had in-house.

    Totally agree with you on system synergy. The room, speakers, amp, preamp, etc etc etc can make huge differences, but it's so damn nice to have someone like Torq that has had a ton of experience in his own system with so many different DACs. This thread has gotten such a following because it's rare to have someone so willing to give comparisons to those of us hungry for them.
     
  5. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Yes. When I had the DSD Jr. and in the weeks I've spent with the DSD (on various firmware), the VAST majority of the time it was being fed 16/44.1 PCM.

    I only tried the PCM to DSD conversion while trying to get a feel for the unit's behavior. Some DACs that do their native conversion via D/S fair better when the input is DSD (native or otherwise).

    But yes, I prefer PCM to DSD with the DirectStream. And with NATIVE DSD content (i.e stuff that's never been converted to/through PCM), I felt the DirectStream also did a better job with it than DAVE did.

    DAVE (and Analog 2) gave a better result with PCM vs. the DirectStream fed the same.

    Now, I'd want more time, particularly back to back, with Red Cloud to see if those impressions stick. I'd only consider those observations as "reliable" (from my perspective) comparison based on Torrey's at this point.
     
  6. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    And ... let me simplify that.

    Right now, for native PCM source material I favor Yggdrasil Analog 2 and DAVE.

    For native DSD it was generally the DirectStream that won (of course Yggdrasil won't play that anyway).

    And if we're just talking DirectStream, then I like it better with PCM than with DSD, but it still has the best DSD performance of the three.
     
  7. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Another thought, or at least observation, while I've got a few minutes and am feeling "chatty" ...

    It has taken several months with DAVE ,and a LOT of critical listening, for it to re-elevate itself to the level of Yggdrasil Analog 2. When I first got the Analog 2 pre-production boards it was, for me (as the notes earlier in the thread should show) the "Yggdrasil+" that was the clear and immediate winner and my musical preference.

    More time with both, in the same system, has given me a better appreciation for the details and differences between these units some of which are very fine-grained and minute. Subjective reviews exacerbate these to a huge extent. Preferences play a huge part.

    And, again, one of these units is 1/5th the price while being as good as, or better than, the other.
     
  8. Thenewerguy009

    Thenewerguy009 Friend

    Pyrate
    Joined:
    Oct 11, 2015
    Likes Received:
    385
    Trophy Points:
    63
    Did you ever give up-sampling a try? Energy over at head-fi is raving about the benefits it gives.
    https://www.head-fi.org/threads/metrum-pavane.766568/page-23#post-14056203
    He is using HQPlayer & up-sampling is done via his PC, which you need a 12-16 core CPU to get it running without any hitches & to get the better sound quality.
     
  9. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Yes, I've tried it - with pretty much every DAC I've had in my hands that supports multi-rate DSD.

    With some DACs it results in that DAC performing better than it does when fed PCM (and where this has been the case, it is noted in my review/impressions). Sometimes there's no audible difference. When there is an audible difference, it's by no means always positive.

    The little Pro-Ject Pre Box S2 Digital sounded best when fed DSD512 via upsampling. Things like DAVE and the DirectStream DACs sound worse or, at best, no different to using their own internal upsampling/conversion.

    In general, I find it to be a lot of effort for generally no gain. This has been the case during both sighted and blind testing. And the best results, in absolute terms, I've gotten across all the DACs I've sampled have been using PCM material fed to the DAC in native form.
     
  10. CoLdAsSauLt

    CoLdAsSauLt New

    Joined:
    May 11, 2016
    Likes Received:
    11
    Trophy Points:
    3
    I don't want to swear in this thread... but I own an Auralic Vega. Excellent unit imo, but I've never had the chance to hear Yggdrasil or something R2R.

    Looking for an upgrade path however. Nonetheless you can count me in as both morbidly curious and actually hoping to be guided in my pre-purchase quest.
     
  11. mitochondrium

    mitochondrium Friend

    Pyrate
    Joined:
    Sep 1, 2017
    Likes Received:
    1,115
    Trophy Points:
    93
    Location:
    A Cell
    An observation about system synergy

    Some time ago I tried a Gungnir Multibit Gen 5 against my Armature Cronos (Denafrips Ares clone) in my system. My conclusion was that the difference was so small that a) it took intensive listening to perceive it b) this did not warrant the extra amount of money. The Gungnir Multibit seemed to render the overtones of the piano on this:

    [​IMG]

    slightly better


    How come? I have thought about that.
    My conclusion is, that the outcome is due to the way my system is set up:
    I am listening to active speakers in a room where T60 is below 0.3 s down to 90 Hz and at 60 Hz it is 0.45 s for the left channel and 0.3 s down to 96 Hz and 0.45 s down to 75 Hz for the right channel. Due to the asymmetry of my listening room even with absorbers I am not able to achieve perfect stereo imaging, this is why I use DRC. I measure using REW and make the filters with DRCDesigner. I then upload these filters into Daphile which runs on an UpBoard. In order not to have to switch between filters for different sampling rates I upsample everything to 96 kHz before applying DRC. So Daphile upsamples or downsamples everything to 32/96 with sox and brutefir is then doing the DRC with 64 bits and then downsamples to 32/96 for the output to the DAC. There is one more reason for upsampling everything to 96 kHz besides ease of use:
    I cannot hear differences between different digital filters at sample rates higher than 48 kHz. I hear differences when music plays at 44.1 or 48 kHz.
    I guess this is the reason why the Gungnir Multibit did not shine, I guess under these cirumstances the megaburrito whatever it is called filter Schiit uses can't work its magic. That would tie in with Torq's experience that Yggdrasil (A1 or A2) are hard to beat when listening to 44.1 kHz but things can be different when using higher sample rates.
    Now I always felt that the impact of DRC beats the inconvenience of upsampling but maybe the better comparison of the two DAC would have been with no DRC, maybe the Gungnir Multibit without DRC would have outperformed the Cronos with DRC? In a room which would be highly symmetric and where the two channels are matched close enough to be able to not use DRC this might well be.
    Listening to the well known Toccata by Bach through my ADI-2 PRO sounds nice. One advantage is that it does not truncate the incoming 32 bits nor does it dither. The Cronos truncates to 20 bits, this too, might not be a good idea. The same thing would apply if I used a soekris DAC, even if it would only truncate to 27 bits. I also prefered the Master 7 I once had with DSP set to dither, when feeding it 32/96.
     
  12. winders

    winders boomer

    Banned
    Joined:
    Feb 13, 2017
    Likes Received:
    1,596
    Trophy Points:
    113
    Location:
    San Martin, CA
    After reading that all I want to do is take a shower...nor can I stop shaking my head.
     
  13. Azteca

    Azteca Friend

    Pyrate
    Joined:
    Sep 26, 2015
    Likes Received:
    1,581
    Trophy Points:
    93
    Location:
    Cincinnati, Ohio
    DRC is a valid application. I appreciate the contribution even if it doesn’t line up with most posters use cases.
     
  14. winders

    winders boomer

    Banned
    Joined:
    Feb 13, 2017
    Likes Received:
    1,596
    Trophy Points:
    113
    Location:
    San Martin, CA
    DRC was not my issue.....
     
  15. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Nothing wrong with the Vega, except the price.

    I owned one for a while, along side my first Yggdrasil. My biggest issue with it remains the price (or at least the value of the unit) more than it does the sound. It's one of the better examples of it's breed. Not my long-term preference overall, but it's a very good unit ... just priced at about what double what I think it should be.

    The more recent Altair is a much better value proposition in my opinion. It's on the money in terms of price/performance, sounds about 99% as good as the Vega, has a ton of extra functionality, and costs $1,000 less.
     
  16. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    I think they're hard to beat regardless (Yggdrasil and DAVE). But that comment was more about them doing a better job when you let THEM do the up-sampling (rather than doing it in software, ahead of them) than any deficiency in their performance with nominally high-resolution source material. Remember, the context was in the use of extreme up-sampling in player software, not in feeding them other, nominally hi-resolution, content.
     
  17. mitochondrium

    mitochondrium Friend

    Pyrate
    Joined:
    Sep 1, 2017
    Likes Received:
    1,115
    Trophy Points:
    93
    Location:
    A Cell
    Yes I should have phrased that differently. The prerequisite for Schiit MB DAC to work their magic is not low sample frequency but the fact that they are fed unadulterated pcm data regardless of sample frequency.
     
    Last edited: Mar 12, 2018
  18. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    That's a long-winded way of saying that the higher your sample rate in relation to the highest frequency that's actually present in the content being sampled, then the less critical the filters are. Similarly, oversampling post-capture helps in this regard also.

    This isn't really news.

    And is definitely way off topic.
     
  19. StefanAC

    StefanAC Acquaintance

    Joined:
    Oct 24, 2017
    Likes Received:
    32
    Trophy Points:
    18
    Location:
    Germany
    I‘m in the same situation. And while my limit is a bit below the mentioned $5K, I prefer to buy used - which makes some of the higher priced units @Torq reviewed very interesting.

    BTW: This is the thread that made me join SBAF after lurking a long time in the background.
     
    Last edited: Mar 20, 2018
  20. drgumbybrain

    drgumbybrain Science Nut

    Pyrate
    Joined:
    Jan 9, 2017
    Likes Received:
    2,219
    Trophy Points:
    93
    Location:
    Lives in Fortaleza, Heart in Girona
    Just Had the Psaudio DSD and Yggdrasil A2 in the house for the weekend with some nerd and weirdo friends from university. I dont have the meanings to express myself about audio diferences like Torq, but for me both DACs are top of the line technology. PSAUDIO DSD it's more relaxed and "soft" ** and much ***(IMHO) more emotionally connected to music when compared with Yggdrasil A2. The Schiit Dac was more energetic, had a little better soundstage but could get a little fatiguing with focal utopia - Zana Deux Combo. Timbre was better in Schiit when playing 44k and better with Psaudio when playing 192k or DSD. PSAUDIO DSD was with red cloud firmware, and overall impressed me A LOT. The sound was so addicting, comfortable ... maybe this is the kind of sound that many people talk about with CHORD DAVE... dunno. Before that experience I thought that delta sigma dacs where all poop. I will try to demo a chord Dave in England ASAP this year.
    Both dacs where connected to Auralic Aries AES. (When playing DSD 64 - Beethoven 5th, the psaudio blow my mind away) In the end, if I could buy it (PSAUDIO DSD) for 3.000 dollars, I definitely would. But for 7.000 dollars in my country, I really cannot afford it. With 7000 dollars in my pocket I would buy a Yggdrasil A2, an used Focal utopia and a nice Amp lIkeIFI pro ican and would have money for beer and barbecue. Please remember that was just my opinion and no psychedelic drugs were used in the process!
    :drunk:Cheers


    ** Edited after @bengo quote: SOFT was the best word in English yo express my self. in my native language I would use the word "redondo" or lush? In english

    *** those impressions are always only words, because dac differences sometimes are minimal, but do change my music reactions
     
    Last edited: Mar 26, 2018

Share This Page