Audirvana Plus 3 - Released

Discussion in 'Computer Audiophile: Software, Configs, Tools' started by Torq, Mar 8, 2017.

  1. Torq

    Torq Last Remaining Good HF Poster

    Friend
    Joined:
    Jan 15, 2016
    Likes Received:
    4,521
    Dislikes Received:
    1
    Trophy Points:
    113
    Location:
    Seattle, WA
    The new version of Audirvana Plus is finally available.

    $74 for new customers, and $39 as an upgrade (if you purchased Audirvana Plus 2 after 12/25/16 then the upgrade is free).

    I'll post feature additions/changes for the new version once I can find a proper list (or change log), though one of the major features is MQA software decoding (the first unfold). Not that MQA is necessarily something people here will desire, but that's the only new feature I'm sure about at present!

    I'll be sticking this on the new laptop later this afternoon to see how it behaves.
     
    Cryptowolf, GanGreinke and gbeast like this.

  2. gbeast

    gbeast Mighty Moral Power Ranger

    Friend
    Joined:
    Nov 8, 2015
    Likes Received:
    924
    Dislikes Received:
    17
    Trophy Points:
    93
    Location:
    Portlandia
    Home Page:
    Nice. looking forward to your posts. I rock Audirvana Plus since I use mac. It has had tons of issues but the latest updates fixed a lot of them.
     
  3. jowls

    jowls Something related to poop - Friend

    Friend
    Joined:
    Oct 3, 2015
    Likes Received:
    1,515
    Dislikes Received:
    0
    Trophy Points:
    93
    Location:
    Australia
    Audirvana > everything else. At least for OSX.
     
    Abraham Lincoln likes this.
  4. jexby

    jexby Raised by Carthusian Monks - Friend

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    1,471
    Dislikes Received:
    2
    Trophy Points:
    93
    Location:
    CO
    truth. except Apple has still borked the "Direct Mode" feature (for security implications) and A+v3.0 still requires a kext hack to restore "Direct Mode" DAC access.
     
  5. Madra

    Madra Rando

    Joined:
    Oct 18, 2015
    Likes Received:
    15
    Dislikes Received:
    0
    Trophy Points:
    8
    Downloaded A+ 1.3 and have been listening for a couple of hours.
    So far, glitch free operation.
     
  6. Merrick

    Merrick Friend

    Friend
    Joined:
    Jan 6, 2016
    Likes Received:
    3,240
    Dislikes Received:
    4
    Trophy Points:
    113
    Location:
    Portland, OR
    Audirvana is what I use as well, but if the only benefit of the new version is MQA then I will save my money.
     
    TRex likes this.
  7. Orkney

    Orkney Rando

    Joined:
    Jul 15, 2016
    Likes Received:
    9
    Dislikes Received:
    0
    Trophy Points:
    3
    I've been using Audirvana since the get-go but am likely to skip this version unless it's earthshatteringly great -- or, at least functional. I've had tons of problems with v2x and essentially abandoned it for daily use.
     
  8. Torq

    Torq Last Remaining Good HF Poster

    Friend
    Joined:
    Jan 15, 2016
    Likes Received:
    4,521
    Dislikes Received:
    1
    Trophy Points:
    113
    Location:
    Seattle, WA
    So far, other than the MQA stuff, and the claimed improvements to the audio-engine, the only changes I've really found have been some squashed bugs and some basic (if overdue) enhancements to library/playlist management and handling of metadata.

    I never really had problems with 2.x ... other than the "Direct Mode" problem (which is handled without issue in some other products, so I'm not sure what Damien is doing differently there), however I will say that at it's worst Audirvana has been vastly more usable and stable than Amarra has ever managed at it's best. Amarra 4.x does not seem to be doing much to change that trend.
     
  9. drfindley

    drfindley Secretly lives in the Analog Room - Friend

    Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    1,217
    Dislikes Received:
    1
    Trophy Points:
    93
    Location:
    Palo Alto
    I've been using it for a day and I haven't see many improvements myself aside from the ones you mentioned. I need to file a bug as the tidal playlist reordering is terrible.
     
  10. Torq

    Torq Last Remaining Good HF Poster

    Friend
    Joined:
    Jan 15, 2016
    Likes Received:
    4,521
    Dislikes Received:
    1
    Trophy Points:
    113
    Location:
    Seattle, WA
    After listening with Audirvana 3.x extensively over the weekend, I cannot say I hear any difference between it and the 2.6.x releases.

    General playback of TIDAL raw PCM content sounds better via Audirvana than with the native macOS Tidal client, but it's entirely possible (even likely) that that's down to expectation bias. This is not a change from 2.6.x.

    MQA playback from either client sounds identical ... and is subject to the same, fucking annoying, several-second pause before MQA files will actually start to play.

    I like Audirvana as a simple, self-contained, fast, and flexible (AU support, OTF resampling and DSD conversion if I want it), local player - which is useful on extended trips where I have the space for a laptop. In home-use, I don't find any SQ benefit to it compared to Roon.
     
    Cryptowolf and drfindley like this.
  11. iDesign

    iDesign Acquaintance

    Joined:
    Oct 26, 2015
    Likes Received:
    36
    Dislikes Received:
    1
    Trophy Points:
    18
    I have been comparing 2.6.5 and 3.0.1 there is a small discernible difference in sound quality with lossless files in my library. I also believe there is a distinct difference between Audirvana and Roon-- in comparison to Audirvana, I have found Roon has more bass which has the effect of slightly masking the treble and clarity. I also find that Audirvana is more dynamic, enveloping, and clearer in its presentation which is especially noticeable in old symphony recordings. Audirvana is my preferred music player because it supports Audio Unit plugins and it has better customer support.

    I also should mention that I took the time to fix the kernel extension in Sierra to enable Direct Mode but I still hear the benefits of Audirvana with it switched off.
     
    Last edited: Mar 13, 2017
  12. jexby

    jexby Raised by Carthusian Monks - Friend

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    1,471
    Dislikes Received:
    2
    Trophy Points:
    93
    Location:
    CO
    FWIW:
    found this "Direct Mode tip/hack" on the Audirvana Board:

    "There's a hack which will help circumvent the Direct Mode problem (which is an Apple issue, so out of Damien's control).
    What you need to do is: Select some music (any music) to play and when it starts check the direct mode box (this will also enable you to alter the integer setting if you wish), then select the music you really want to listen to, activate play and the Direct Mode remains checked. You have to repeat this each time and I'm not sure if it actually impacts on the music, but it might."

    haven't heard a clear improvement yet tho and need to do more comparisons.
    it's possible the Direct Mode checkbox stays on, yet Direct Mode is not truly enabled to the DAC.
     
  13. iDesign

    iDesign Acquaintance

    Joined:
    Oct 26, 2015
    Likes Received:
    36
    Dislikes Received:
    1
    Trophy Points:
    18
    This does not work because the setting is not saved while a track is playing. You will notice that tracks will fail to play after the the song is stoped or If you quit and reopen Audirvana. The only way to enable Direct Mode is to fix the kernel extension in Sierra which is not a simple process unless you're very experienced with macOS Terminal.
     
    Last edited: Mar 15, 2017
  14. jexby

    jexby Raised by Carthusian Monks - Friend

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    1,471
    Dislikes Received:
    2
    Trophy Points:
    93
    Location:
    CO
    not in my case. once i get song #2 playing (with Direct check box on) all songs continue to play fine.
    of course quitting the app entirely disables the Direct work around.

    I'm not sure Direct mode is really enabled for song #2, can't hear a bit of difference.
     
  15. iDesign

    iDesign Acquaintance

    Joined:
    Oct 26, 2015
    Likes Received:
    36
    Dislikes Received:
    1
    Trophy Points:
    18
    Im afraid it does nothing because the setting is not saved while a track is playing.
     
  16. jexby

    jexby Raised by Carthusian Monks - Friend

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    1,471
    Dislikes Received:
    2
    Trophy Points:
    93
    Location:
    CO
    suspect you are right, and the checkbox staying checked is an artifact.
    I will do the real kext replacement at OS level this week, enable real Direct mode and re-listen closely.
     
  17. Tobes

    Tobes Rando

    Joined:
    Jan 29, 2016
    Likes Received:
    3
    Dislikes Received:
    0
    Trophy Points:
    3
    Very surprised you don't find the SQ change substantial compared to 2.66. I actually wonder what Damien tweaked to make it so different.
    To me 3.02 has substantially more transparency and transient clarity. This can also mean it's less forgiving of poor quality recordings - I note some users are complaining 3.XX sounds too bright.
    FWIW I find 3.02 superior in SQ to Roons player or Roon/HQPlayer or HQPlayer alone (when comparing players without up sampling or filter options engaged).
     
  18. Torq

    Torq Last Remaining Good HF Poster

    Friend
    Joined:
    Jan 15, 2016
    Likes Received:
    4,521
    Dislikes Received:
    1
    Trophy Points:
    113
    Location:
    Seattle, WA
    "Substantial" is a rather ambiguous term. In most cases the differences between major hardware components are rarely what would fit my expectations for a "substantial" difference. Excepting an allowance for the effective exaggeration of differences in prose when reviewing something in order to illustrate any difference at all.

    Assuming he's not fiddling with the bit level data (which I do not believe he would do), it's unlikely to be something either a) deterministic (any change is more likely down to chance than intent and b) the same between any two systems.

    ...

    While I will, at some point, write up something on the high-degree of improbability that any code changes running on any reasonably modern general purpose computer (that's running more than one task/thread) can yield any specific or intended effect on sound quality within a bit-perfect digital system, in the interim I do my best when comparing things to eliminate as much potential for expectation bias as I can. And when I do that, in this case, Audirvana, Roon and various other players all show up as being indiscernibly different.

    Now, I will not claim that my methodology is a pure and true DBT, but it's a lot closer than any sighted test is going to be. Nor am I saying that software players cannot sound different. I am saying that you cannot write code in anything above assembly language, for modern CPU architectures, that has enough control over what actually executes when that, even on a dedicated machine (i.e. no OS running random code at random times), you can design for or achieve any specific or consistent effect on sound quality.

    Even coding in assembly language, what instructions you provide and what the CPU actually executes are, except in very specific cases, not predictable. Between the CPU actually re-compiling/interpreting your instructions in it's own internal microcode (hint: X86 CPUs haven't actually executed X8 op-codes directly in at least 20 years) and things like OOE, the actual activity of your CPU is only marginally predictable. 99.999% of programmers are not even aware of these issues, let alone have to deal with the consequences of them.

    And what this means is that while different code can result in different sound (through a very limited set of rather theoretical circumstances), it's almost impossible to code in a way that will deterministically improve things (vs. just having a random effect) and even less likely that such effects would be consistent between any two computers that were not absolutely identical in spec/build (which effectively never happens).

    All that said ...

    If you prefer one over the other sonically, there's no reason not to use it. After all, the point is, well for me at least, all about enjoying music - and anything that helps there (real, imagined or of a "mood changing" nature) works for me.

    Personally, I shall remain happy in either their being no discernible differences and/or that I just can't detect them in a the closest to a DBT I can muster, so I don't have to worry about such things and can spend more time simply enjoying music.
     
    Wilson, Grahad2 and philipmorgan like this.
  19. Tobes

    Tobes Rando

    Joined:
    Jan 29, 2016
    Likes Received:
    3
    Dislikes Received:
    0
    Trophy Points:
    3
    @Torq, thanks for the considered reply.
    Let me say upfront I'm not happy that I hear distinctive differences between supposedly bit perfect players and I too wonder how this could be the case. Nonetheless I am quite satisfied that these SQ differences exist, consistently across different recorded material and listening sessions.
    I would certainly hope that Damien is not dabbling in some bit level manipulation in order to differentiate his product - but something is definitely different about A+3.

    I have purchased both Roon and Audirvana products. I previously found Roon to be superior in SQ but only when I used it in conjunction HQPlayer - a combo many times the cost of Audirvana. I'd spent the money on Roon/HQP so I suppose you could say I had a vested interest in it sounding better. To me that wasn't the case.
    I found Audirvana 3 to sound more transparent (in the sense of hearing into the soundstage and hearing room/ambient sounds and presence of the musicians), it also has more focus and transient clarity. At first I thought it sounded leaner and perhaps brighter - and wasn't sure I liked this - but on good quality material this wasn't the case. Whether more listening will reveal some flaw, time will tell.

    Perhaps you are fortunate that you hear no difference between the players, I think my listening would be less complicated if that were the case.
     

Share This Page