Life after Yggdrasil: Watering the Ash

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

Tags:
  1. murphythecat

    murphythecat GRU-powered uniformed trumpkin

    Pyrate Banned
    Joined:
    Apr 4, 2016
    Likes Received:
    1,201
    Trophy Points:
    93
    there's so many more DAc that have not yet been tested.
    T+A dac 8, Ec designs, Trinity dac, and many other DAC have not been tested.
    shouldnt jump to conclusion.
     
  2. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Yes, you suggested I consider the Mosaic "T" DAC back when I was actually looking for a new DAC for myself. At the time, despite talking with the manufacturer, there wasn't a practical way to get a unit for audition. Hopefully that will be different this time around - especially as where I'm NOT looking for a new DAC myself at the moment I'm a lot less inclined to jump through hoops to make auditions happen than I was in the original cycle (and I wasn't very inclined to do that even back then).

    So we'll see ... I'll throw it up on the candidate list and we'll see if the manufacturer/distributor comes through with a demo unit.

    --

    I'm going to say up front that being USB-only and using a single crystal time base is very unlikely to do positive things for the sound of the thing though. The Mosaic T looked a LOT more interesting that this new thing.
     
  3. murphythecat

    murphythecat GRU-powered uniformed trumpkin

    Pyrate Banned
    Joined:
    Apr 4, 2016
    Likes Received:
    1,201
    Trophy Points:
    93
    T+A dac 8 should also be on your list!

    as for the usb input on the mosaic UV, I personally find this very interesting. the whole usb input runs on a single XTAL. no more clocking issue possible? all I know is that if john brown changed from Toslink to usb module, its definitely becuse it sounded better, no doubt about it. the way usb is implemented is absolutely unique, nobody else does this. as far as I know, the dac module and most module stayed the same and the difference between the Mosaic T and Mosaic UV is mostly the USB module.
    ''
    That is why we had to make our own USB firmware solution to hand over digital music samples to our discrete DA converters. Thus our goal for the USB firmware in our microcontroller was to keep it as quiet and simple as possible.

    This eventually resulted in using a single core 32 bit microcontroller running on a single 12 Mhz XTAL. The firmware now uses one single interrupt for the sample frequency, uses hardware counters for timing and uses polling to handle the USB stack.

    ''
    https://www.ecdesigns.nl/support_files/mosaic-usb-module.html

    anyhow, we've decided to order the holo spring just to be sure how it stacks to the Mosaic UV.
     
  4. msommers

    msommers High on Epipens

    Pyrate
    Joined:
    Nov 6, 2016
    Likes Received:
    2,750
    Trophy Points:
    113
    Location:
    Calgary, Alberta
    Home Page:
    @Torq you've mentioned a few times now that the Yggdrasil portrays piano the best out of all the dacs you've heard. Within (or maybe externally) your extensive track list you've already outlined, is there a specific song/format/version that is your general go-to for evaluation?

    Fiona Joy is a modern pianist I really enjoy and she records with Blue Coast Records (very high quality recordings) - may be up your alley.
     
  5. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    I need to complete the post that has my audition playlists in it. So far it just has the first one I use (and the one things used to have to get past to have me bother with the rest). I'll try and do that this weekend.

    But the tracks I use for evaluating piano are based on my own recordings of my own instrument (a rather lovely turn of the century Steinway Grand, with ivory keys as that's all you could get at the time, that's been played by some very prominent visiting artists), specifically setup in a way I find revealing. I've been mulling whether or not I would share those as, in general, my music is for me, and haven't come to a decision on that yet.

    Indeed!
     
  6. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Please read the 2nd post. It already is on the list.

    Actually it makes clocking vastly more complicated than it needs to be and is also highly likely to increase jitter at the sample clock. Notwithstanding the fact that there's nothing in USB that has any relation whatsoever to sample clocks.

    It's done this way in hundreds of micro-controller applications. There will be differences in the software implementation, but what amounts to inverted bit-banging isn't remotely new.

    ...

    Using interrupts, software dividers or hardware counters to convert from 12 MHz (or any other arbitrary frequency) to some multiple of either 44.1 KHz or 48 KHz is an old problem. We don't do it because it tends to suck. Unless there's a strong market for super-precise clocks at 12 MHz (which I can't find), it's likely to be a lot less stable, even accounting for the vastly reduced sensitivity to beat-error/tick-error/drift, than any reasonable audio-grade crystal.

    It also requires a micro-controller, which is one of the noises components one can choose to put on a board - one reason I'm not a fan of XMOS or Amanero implementations. It can be done cheaper, better, and quieter with purpose-built silicon.

    But ... as always I'll give it an unbiased listen (by which I mean I won't trash it if I like it - I'm out of axes to grind) - assuming they can source one.
     
  7. Senorx12562

    Senorx12562 Case of the mondays

    Pyrate
    Joined:
    Oct 26, 2015
    Likes Received:
    3,274
    Trophy Points:
    113
    Location:
    Bird-watcher's paradise
    I'm getting a distinct oder of...is it...shillburger?
     
  8. murphythecat

    murphythecat GRU-powered uniformed trumpkin

    Pyrate Banned
    Joined:
    Apr 4, 2016
    Likes Received:
    1,201
    Trophy Points:
    93
    ive just been a fan of ec designs for years. not a shill, just a fan and as soon as I find a dac I prefer, ill buy it in a heartbeat :).

    the hifiadvice review of the ec designs mosaic T is what made me very interested in them. after being compared and preffered to the metrum pavane, jeff rowland aeris, ect
    http://www.hifi-advice.com/blog/review/digital-reviews/spdif-dac-reviews/metrum-pavane/
    "Like the Mosaic, the Pavane has a highly neutral and unforced sound. And like the Mosaic, it has a tendency to make normal DACs sound etched and hard in comparison. Also like the Mosaic, the Pavane is very subtle and has great differentiation of low level detail."

    ''While the Mosaic T may not be as dynamic and punchy as the Pavane, it does present music with a natural flow and to me sounds emotionally more appealing.''

    http://www.hifi-advice.com/blog/review/digital-reviews/spdif-dac-reviews/ec-designs-mosaic-t-dac/


    the Mosaic UV is a evolution from the Mosaic T. if the Moasic T was already preffered to many reputed dac, im interested in Torq opinion of the Mosaic uv!
     
    Last edited: Mar 26, 2017
  9. murphythecat

    murphythecat GRU-powered uniformed trumpkin

    Pyrate Banned
    Joined:
    Apr 4, 2016
    Likes Received:
    1,201
    Trophy Points:
    93
    interesting, I really dont know enough in DAC. he posted on diyaudio a explanation of his usb module, maybe you can understand more what he explains:
    "
    First picture shows a novel single-chip USB audio receiver & system control prototype module. My brother developed the software for this marvel. It contains many innovations, USB, CPU, and DAC timing signals are all derived from a single local masterclock (single time domain operation). Time domain dithering is applied, it has similar effect on sound like dithering applied after reducing resolution (CD recording).

    So we end up with higher perceived resolution and there is no longer the need for sample rates above 96 KHz. This module talks multi-stream so we could finally dump the very problematic I2S interface and the required I2S to multi-stream decoder. The major problems with I2S are the high interference energy levels (symmetrical square waves with form factor 1) and spectrum (dominating odd harmonics & inter-modulation with every clock or data signal within in a radius of a few meters).

    With the multi-stream interface we have very flexible data transfer and timing and the freedom to shape the digital audio interface spectrum any way we like. We now use very narrow pulses for both data and timing signals and spread spectrum timing.
    "


    ANYHOW, if the manufacturer cannot provide you a demo dac, i will send you mine when it arrives so you can evaluate it. ill cover the shipping! im just too curious about how it stacks up.

    thanks for all your contribution, what a journey your on!
     
    Last edited: Mar 26, 2017
  10. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    So, don't take any of this as being direct AT you @murphythecat ... I'm just responding to the quotes you've provided.

    And those quotes really don't say much that is substantive or necessarily borne out by my own engineering experience - in fact at first blush I thought someone was channeling Rob Watts there. But lets take a few of this things at face value, before getting back on topic (there's a couple of other threads that further discussion of this should probably be directed to), and note, as doing so, that I could be entirely wrong - I'm just raising the questions that come up when claims like this are made with too few details to make sense of.

    I've seen this done all over the place - including on very basic micro-controllers like the Atmel tinyAVR. They way he is doing it might be novel, but deriving other clocks from your system master clock (CPU clock) is more common than not. And how good it is lives or dies on a number of factors not directly in "your" control. How good is that master clock in optimum conditions? How sensitive to thermal variation? What compensations are in place for this?

    Where is this applied? As my first suspicion is that it's an attempt to compensate for generally poor clock stability/accuracy.

    I strongly suspect that it's not a case of "no longer need" and much more a case of "couldn't get it to function reliably". But, it's not clear if this is specifically a limitation from the USB perspective (since that's being done via polling) or the derived sample-clock, so who knows? That, or it's weasel-words for a decision to only support UAC1 so they didn't have to write drivers for Windows.

    Now I really want to know what micro-controller is being used here as it is quite likely that it's a vastly more problematic offender for that sort of thing. Its one reason why the massively over-powerful XMOS and Amanero SoCs can cause issues if not treated sensibly (and they almost never are).

    --

    Anyway, these aren't things you need to answer, and I don't really care at the end of the day - they just aren't specific enough not to cause my engineering-BS meter to get all antsy. Ultimately, if it sounds great then that's fantastic. Ant the only way to know that is to hear one in a familiar system and see what it can do ...

    Very generous, but not necessary.

    If they, or their direct agents, cannot arrange for a demo unit then it's unlikely that most of the folks here are going to fare any better in getting to hear one and I would not advise anyone to take my inane drivel as any kind of rationale to "buy without hearing for yourself".

    There's a difference between "someone can loan me one because it's convenient" and "someone has to loan me one because they're impossible to hear without buying".
     
  11. murphythecat

    murphythecat GRU-powered uniformed trumpkin

    Pyrate Banned
    Joined:
    Apr 4, 2016
    Likes Received:
    1,201
    Trophy Points:
    93
    hi torq, let me quote what you just said on ec designs diyaudio thread and see what he answers.

    still, ill send you mine, I really value your opinion!
     
  12. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    No need.

    If I cared about the answers, I'd go over there and ask myself.

    And to be clear, I don't want to spend time auditioning stuff that anyone else is going to have to buy blind. While it might be academically interesting, that's very low-priority stuff. It would be different if this was the first cycle where I was actually looking for a new DAC for myself - where it didn't really matter how I got the audition. Here, it matters a lot more.
     
  13. Kattefjaes

    Kattefjaes Mostly Harmless

    Pyrate
    Joined:
    Sep 5, 2016
    Likes Received:
    4,521
    Trophy Points:
    113
    Location:
    London, UK
    Sadly, some of us have to take the inane wibbling of ne'er-do-wells like you at face value, and use it to inform purchases, especially in the case of American brands. Either they don't have dealers here, or the dealers are so monumentally poorly-stocked and run that getting them to actually ship the totality of your order at all can be a challenge. I'm seriously considering a Gungnir Multibit (or similar, depending on what happens to the market in the near future), and when I do pull the trigger, it will probably have to involve getting one Schiipped direct.

    I do make an effort to super-sample opinion, and understand the context and preferences of each reviewer, though. Given that the vast majority of hifi reviewers write "everything is awesome" drivel to keep the free stuff flowing, very few reviewers are actually worth the oxygen that they use. It's a jungle out there, which is why I value certain grumpy bastards who aren't afraid to call a spade a spade.
     
  14. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Since this came up, as a point of contention, in the original thread, I wanted to be absolutely clear on one point here (quoting myself for simplicity):

    I do not have a problem auditioning DACs that don't have a traditional dealer network (e.g. direct-sales models like Schiit) and that might be best handled for my audition by a loan from an owner. However, I do not want to invest a lot of time dealing with units that are going to be a problem for everyone to get hold of to hear.

    Keeping with Schiit as the example, most people won't (or shouldn't) have trouble finding someone local with a suitable Schiit DAC to audition, and can easily bypass the buy/return/restock model. It won't always work, but it's a lot more likely than a lot of brands - especially smaller, ultra-niche, ones.

    In other cases, even if it might be fiddly to secure an audition, if it's the manufacturer/distributor providing the unit (e.g. that's how the Denafrips stuff will get into my hands, courtesy of @alvin1118), then that's okay too. But if a manufacturer/distributor can't make it happen for something like this, then I don't imagine it being something that'll happen for myriad other people that want an audition. Which means buying on faith, often from other countries, and hoping that a) nothing happens to the unit in transit, b) incurring potentially significant (relatively) expensive and c) trusting to the return/restock policy.
     
  15. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Yes, geography can make some of this a lot harder to get hands-on (or ears-on) time with, for sure.

    One thing we can do here, for super-niche units that are a pain to get hold of, but for which kind souls like @murphythecat are willing to provide access to their own units, absent a manufacturer wanting to do the same (and EC Designs might, they were communicative in the first round, just couldn't make it happen), is to put the unit to the vote. At least that way we can see if there is enough interest in it from the SBAF membership that it's worth it for them - in which case I'm happy to do it.

    If that's interesting, I'll have a think about the best way to facilitate it.
     
  16. murphythecat

    murphythecat GRU-powered uniformed trumpkin

    Pyrate Banned
    Joined:
    Apr 4, 2016
    Likes Received:
    1,201
    Trophy Points:
    93
    amazing and yes, some european company may not have the means to adequately distribute demo units. I dont think it will be a problem for ecdesigns if you were to contact them this time around...

    in any case, lending you my Mosaic UV is something Im happy to do. the more DAC tested, the more option is there for all of us! Mosaic UV is very cheap (relatively) and the Mosaic T compared favorably to the Pavane so I think its even more interesting due to its pricing.
     
  17. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,777
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    @Torq: When are you going to review the latest Lampizator 11.B13 DAC and OJnegDAC? I demand that you review them. Why are you taking so long?
     
  18. FredM

    FredM New

    Joined:
    May 24, 2016
    Likes Received:
    20
    Trophy Points:
    8
    1+
    And now I understand your previous comment, not to jump to conclusions. The Mosiac price point is interesting indeed.
     
  19. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    [​IMG]

    It's all down to me and my terminal case of procrastination ...

    That, and a strong sense of auditory-preservation.
     
  20. Torq

    Torq MOT: Headphone.com

    Pyrate
    Joined:
    Jan 15, 2016
    Likes Received:
    8,193
    Trophy Points:
    113
    Location:
    Seattle, WA
    Linn

    Before reading about the latest Linn “Network Music Players”, you might want to go back and read the intro to Linn’s current systems in this post … specifically the “Linn” and “What is Exakt?” sections. They will make a bit more sense of what I’m going to talk to here.

    Klimax DSM/2 & Klimax DS/3 (w/ Katalyst Architecture) (+) (!)

    I’m treating these units together as they’re variations on a theme. So … let’s begin by looking at the difference between these two units (the different version numbers are a product of the DS being the first streamer, and it was updated originally when the first DSM unit was released, so a KDSM/2 is just the “M” version of a KDS/3) – which essentially comes down to what inputs they’ll accept:
    • The DS/3, which is the (much) cheaper option, only accepts music over Ethernet using one of a number of protocols (typically OpenHome which is an extension to UPnP/DLNA, but others are supported, including native capability for both TIDAL and Qobuz).
    • The DSM/2, in addition to its operation as an Ethernet-fed streamer, offers HDMI, TOSLINK and COAX digital inputs and balanced XLR analog inputs. Unlike the Akurate-level units, the Klimax DSM/2 does not include a phono-stage … if you want that, you buy something with a balanced output … and if Linn had their way it’d be their excellent Urika or Uphorik stages.
    The considerable difference in price comes from not just the lack of inputs on the DS/3, but the need for the DSM/3 to implement a VERY high-spec ADC stage used to run its analog input. Why have an ADC here at all? This is due to the need to get everything into the digital domain so that you can use the Exakt and Space Optimization features of the units.

    The ADC on the Akurate level gear, which supports more analog inputs than the Klimax stuff, is good enough as to clearly allow upstream changes to come through (be it a change in tracking force, VTA, a different digital input etc.). I’ve done extensive back-to-back comparisons here and the results are so good that using the ADC-input into Linn’s internal, software, RIAA EQ is as good as any of the external phono-stages I’ve had my ears on (by no means all, but quite a few that would be appropriate for a >$50K system).

    Anyway, Linn’s current “everything winds up digital” approach is quite controversial amongst audiophiles. Indeed I was very skeptical about it myself to begin with. And, to be honest, I wouldn’t deal with it at all if it were not for how good both Exakt and Space Optimization* actually are (and how much I'm now needing to rely on them).

    What is “Katalyst Architecture”?

    In short, it’s Linn’s fancy name for their latest DAC implementation, and their “Katalyst” page gives the details. The information disclosed about it won’t rattle anyone’s cages. Super-stable/accurate reference voltages, discrete, tasks-specific, power-supplies, ultra-high-quality clocks, closely-toleranced discrete analog stages etc. You’ve heard all that before. But regardless, in this case, it is responsible for perhaps the biggest single step forward in performance for these units in their 10-years (or so) history.

    And then that is coupled with updated**, FPGA-implemented, oversampling, filtering and processing, as has been common to all Linn “DS/DSM” series components going all the way back to 2007.

    In essence, Linn’s approach is to use CoTS DS-DAC chips for the pure digital-to-analog conversion, in this case an AKM4497, and then do everything pre-and-post that raw conversion in their own custom code. That means that while the 4497 is still considered part of AKM’s “Velvet Sound” family, you’ll hear NONE of that signature in the Linn’s output.

    Auditioning Notes:

    My listening to the new Klimax units was not done in an Exakt system. Doing so would mean I was listening to the DACs in the Exakt modules not those in the source***. Also, you’ll note that, for the sake of simplicity, I’m just going to refer to the KDSM/2 and KDS/3 as “KDS/3”.

    For headphone listening it was my usual setup, and for speakers it was run via conventional analog components from the analog outputs on the Klimax units (KDS/3 -> BAL -> AK Kontrol/1 -> AK4200 -> Akudorik [Passive]). And that speaker listening was done in my original, dedicated, listening room.

    Aside:

    This marks the last time I’ll be able to audition a DAC “properly” in my speaker system. Changes resulting from my recent move mean I no longer have the Kontrol/1 & AK4200 combination necessary to run an analog output/pre/power setup to my speakers. I have switched to a solely-Exakt configuration there (both for quality and sanity purposes).

    How Does it Sound?

    Linn’s presentation tends, across all of their units, to be on the analytical, highly-detailed and dry side of “musical”. That’s apparent here, too, but musical the KDS/3 most definitely is. And more so than ever.

    I had a brief opportunity to compare a KDS/3 back-to-back with its predecessor. Even with more modest hardware, normally such comparisons yield only very subtle differences that take a while to acclimate to. For example, when I compared Bifrost MB to Modi MB, it took a good deal of listening and repetition to be able to tell them apart. With the KDS/2 vs. KDS/3 this was not the case. And, while I certainly wouldn’t call the differences “night and day”, they’re definitely and reliably discernible and there’s not a lot of wiggle room as to which is better unit. It’s also possible to say that the difference is not just down to the latest filter code … as having them running on the same version the differences were unchanged and still quite apparent.

    But … we’re not really here to compare things to earlier iterations of themselves … so while this is definitely the best-sounding, most musical, digital component Linn has ever produced (and I’ve heard all of them, including the short-lived and not-very well known “Kivor” system), there's more to dig into.

    If had to summarize how this thing sounds in a couple of words I’d say:

    Bloody marvelous!

    Whether in my headphone rig or speakers, and especially with electronic music (which has never been subject to tape or other analog stages) … it renders the blackest background of any component I’ve ever heard … beating DAVE, Yggdrasil and the Spring DAC comfortably here. The difference between KDS/3 and Yggdrasil is a bit like comparing the black-levels of a regular single-backlight LCD-TV to that of the very best OLED panels. Music emerges from complete and utter silence. Individually picked notes are startling in their immediacy, emerging from a total void and fading back to absolute nothingness. This contrast between sound-and-none sets a compelling stage upon which everything else is played out.

    Whether it is down to this utter blackness, or some other factor, music seems somehow more vivid played via the KDS/3 than anything I’ve heard before. And the level of detail realized is also absolutely first class. The difference here is smaller than when comparing “blackness”, but it is apparent nonetheless. I had no trouble at all discerning splits in wire-groups with brushes over cymbals, the post-hit reverberation of timpani – down to being able to easily tell where the thing had been hit (from a tonal perspective), or the overzealous hammering of keys in one of my piano pieces … down to the rebound of the key.

    Tonality was as good as I’ve ever heard. Instruments were portrayed realistically with nothing at all calling attention to itself compared to how I expect things to sound. Familial variances in an instrument family are readily delineated. It was also possible to tell the difference between different instruments … for example a pair of flutes … though I am not sure if it was really the natural tone of the instruments differing or if that effect was down the technique of the artists playing them.

    I will say that the lowest registers are more compelling with Yggdrasil, and the impact/slam there is perhaps a little harder. I don’t, honestly, know which is better here. Over time, and more listening, I’ve come back to liking Yggdrasil’s bass rendering more than DAVEs – but calling it between Yggdrasil and the KDS/3 is harder. Really low notes on big church/cathedral organs are, I think more believable on the Linn unit … but I can’t reliably say if that’s down to the (relative) way the overall ambient acoustic is represented (perhaps down to reference to the background) there or if it’s actually the way the raw notes sound (and feel). But this flips the other way with a driving bassline in more contemporary and/or electronics music (e.g. “Partition” and “Chameleon” which I reference for this sort of thing quite a bit).

    Imaging and staging, which is traditionally a strong-point for all-Linn systems, is superlative here. DAVE and Yggdrasil already were top-of-the-tree for this sort of thing in my experience, with only high-end vinyl replay through proper tube-amplification really besting it. The KDS/3 draws even here … at last … throwing an entirely stable, appropriately sized (in both width and depth) and coherent image. Finding instruments in space is easy. The natural “height” of the presentation (a factor of visual cues and, mostly, speakers operating as point-sources with proper driver-time alignment) is preserved. This is not to say there’s “vertical delineation” at work here (I’m not sure that’s possible with two channels), but there is a definite sense of vertical scale to the presentation.

    Instrumental separation is phenomenal, no doubt helped by, once again, the uncannily black background. No question about what’s playing, where it’s playing, who’s not perfectly in time, and so on.

    The sense of air and space in recordings is nothing short of breathtaking. You can almost feel the closeness of the small Jazz clubs, smell the smoke and spirituous vapors, and see the glistening of the lipstick on the singer’s lips (yes, I listen to quite a lot of female vocals … especially so during my time with the KDS/3). Even the sense of the darkness of the club vs. the illuminated performer(s) seems enhanced by that void-black background (yes, I know, I keep going on about it).

    Dynamics, both macro and micro, are finally on par with Yggdrasil and micro-dynamics best even DAVE, although Yggdrasil exhibits slightly more slam in some cases. This doesn’t seem to be a uniform thing, but was apparent often enough that I’d give it the edge here. And, again, due to more listening time, I’d put Yggdrasil ahead of DAVE in terms of slam here too****.

    Listening with the KDS/3 wound up consuming almost every available moment I had with the unit. Yggdrasil was the last component I had that strong a reaction to. DAVE was very impressive, and definitely induced the “just one more track/album” reaction, but not to the same degree that either Yggdrasil or the Linn unit managed.

    Music was rendered as a fully immersive, communicative, expressive, emotional, soul-stirring experience. Yet that was achieved without giving up any of the technicalities … a trade-off that happens too frequently with a lot of gear. The “analytical” and “dry” nature of the Linn’s sound did not intrude on this at all … but it did make it an entirely fluid thing, yielding the ability to easily focus on the details and technicalities if you wanted to, and then just drift back into the full rapture of the performance being laid before you.

    It is fair to say that I did not want to let it go … a point I’ll come back to momentarily.

    This unit, in my opinion, finally elevates Linn’s DS performance to be genuinely competitive with some of the best DACs in the world.

    So where does this thing fall overall?

    In short, as of my last audition, the KDS/3 takes the top spot as the best digital music reproduction I’ve heard.

    It’s not that far ahead of Yggdrasil in any area, in most cases just edging it out, except in terms of blackness of background where the Linn currently reigns over anything else I’ve heard. Schiit’s DAC is going to be more persuasive for slam-junkies and bass-heads, as the Linn, while not reticent there - neither is it as powerful/compelling with some genres.

    For those following along, and noting the changes in some of my impressions over time, this means additional changes in my overall, current, DAC rankings. And the top three units are, now, clearly, between the DAVE, Yggdrasil and KDS/3 units. For me, my tastes, and how I listen to music outside of auditions, as well as the “technicalities” that often become the focus of write-ups like these, my current ranking is now as follows:

    Linn KDS/3 => Yggdrasil > DAVE​

    Mostly, for me, that should be interpreted as saying:

    Linn KDS/3 > Yggdrasil > DAVE​

    If I was using a simple scoring model here, this would put the Linn at 100 (which is nominally how I score whatever I currently feel is “best”), Yggdrasil at 97 and DAVE at 94 or 95. Not big differences overall … a point I generally try and maintain. Given the massive price disparity between these units, however, this changes markedly if one even pauses to consider “value” or “bang for the buck”. And in that case, or even where you’re not already heavily down the Linn path, I’d have to rank it as:

    Yggdrasil > Linn KDS/3 > DAVE​

    How does the Linn at anywhere between $6,000 and $10,000 more than the DAVE best the Chord unit in terms of value? Easy. DAVE is just a DAC/headphone “amp”. To get DAVE to do the same things as the Linn unit (streaming, room correction, speaker correction, analog inputs etc.), even forgiving the slightly lower level of musical enjoyment it delivers, puts it well above the cost of the respective Linn unit.

    The $23,500 Question:

    Am I going to buy a Linn KDSM/2 or KDS/3?

    No.

    Or, at least, “not yet”.

    Why?

    The reason here is simple – running an Exakt system means I won’t get the benefit of their new DAC implementation and analog outputs. The “source”, in an Exakt system, is just a streamer and all the digital to analog conversion, oversampling, filtering, room compensation, speaker compensation, and amplification occurs in the Exakt module – which in my case are built-into my speaker stands.

    So, for me to get the benefits of the new “Katalyst” implementation, I have to wait for it to filter down to the Akudorik Exakt stands.

    What you can be sure of, is that when Linn offer the updated Exakt modules with the Katalyst architecture as an upgrade, I’ll be first in line to switch to them. So much so that I have a “pre-order” logged with my dealer even in advance of any official announcement of availability. For those that know me at all … well, you know I don’t do pre-orders (I can barely be convinced to order things that aren't in stock and available for next-day delivery).

    This will be a particularly big deal for me as it will give me Yggdrasil-levels of listening pleasure in my speaker rig again – something I can’t have actually using an Yggdrasil there now both due to switching to Exakt fully and as a result of now requiring Space Optimization to deal with the peculiarities of my new listening space.

    Actually buying a KDSM/2 is still on the cards … but that won’t make any sense at all until after I have settled on a new vinyl rig … as all I’ll get out of that is the improved ADC.

    --

    *I moved to a new place recently and this presented several challenges for my speaker system, which is where my Linn gear resides. The dedicated listening room in my previous house was a delight, but in the new place the speakers have to live in the main living area, are closer to the walls than I’d like, and have big picture-windows right behind them. Here, without utilizing Space Optimization (which is a built-in room and speaker correction system) the system sounded absolutely awful – as in “Why bother?”; with it, it’s almost indistinguishable from how it sounded before. That, alone, is worth Linn’s high price of admission - for me.

    **These updates to the filters etc. are relatively frequent. While not always universally regarded as positive improvements (you’re not forced to upgrade etc.), more are than aren’t and the iteration that began around this last Christmas have been definite improvements for me.

    ***This means that in any Linn Exakt system, the source is largely irrelevant from a quality perspective, unless you’re using analog inputs, as all the actual conversion and filtering work is done at the Exakt module which is not in the “source”.

    ****I may have to further revise my DAVE commentary as the more I hear it the less enamored with it I become. In fact even a recent opportunity to acquire a used unit from the UK for, effectively, $6,500 was NOT enough to push me over the edge … though it did lead to another audition to make sure.
     
    Last edited: Apr 3, 2017

Share This Page