Audirvana Plus 2.5.1 with PCM->DSD Conversion

[damien78]

Just released a new beta for 2.5.1: 2.5.0.11, available here :

• Fix DSD to PCM conversion (regression that was in the beta)
  
• TIDAL catalog browsing: show album description (when available), along with the hyperlinks it contains, to navigate to related artists/albums
  
• TIDAL catalog browsing: show album credits (click on the i icon top right of the album tracks list), with hyperlink navigation to artists (for artists referenced in TIDAL)
  

Example of such TIDAL browsing, with artist bio, album description & credits:

[juliocat]

Just released a new beta for 2.5.1: 2.5.0.11, available here :

• Fix DSD to PCM conversion (regression that was in the beta)
  
• TIDAL catalog browsing: show album description (when available), along with the hyperlinks it contains, to navigate to related artists/albums
  
• TIDAL catalog browsing: show album credits (click on the i icon top right of the album tracks list), with hyperlink navigation to artists (for artists referenced in TIDAL)
  

Example of such TIDAL browsing, with artist bio, album description & credits:

[ATTACH=CONFIG]26019[/ATTACH]

 

Thanks, This version is Ranger Safe Now

[RunHomeSlow]

First couple of tracks played now with 2.5.0.11

seems faster now for loading tracks in memory (grey bar), but my CPU remains at 213%-224%

in Activity Monitor, even after track loaded completely...

DSD 128 Filter C

no bug, no skip for me, no fan noise, not hot when i touch the Mini just normally warm i guess.

no more sample rate changes skipping, i'm HAPPY, MERCI Damien!

[YashN]

Just released a new beta for 2.5.1: 2.5.0.11, available here :

• Fix DSD to PCM conversion (regression that was in the beta)
  

Gave it a long try today after changing my settings based on our email exchange.

 

The sound in PCM-DSD256 is absolutely amazing with the iFi iDSD Nano together with my SET Tube Amp: it gets a breadth extension in ranges - I can hear texture for the bass instruments, a very clear mid-range too, dynamics are excellent as well.

 

As I still get the 5s and 5 multiple glitch sometimes (coming back to the start of the track sometimes resolves it) as well as the glitch where one side disappears after a high squeak (much rarer), I tried changing some settings as well as launched Activity Monitor this time.

 

CPU usage hovers around 180-193%. I don't see additional peaks when the 5s glitch occurs.

 

I haven't even changed from the default filter, so as I said before, this sounds incredible, just too bad that my current computer cannot seem to be able to keep up.

 

Is it possible to optimise CPU usage further?

 

Well done so far.

[Le Concombre Masqué]

is there a technical background justifying a difference (my preference goes to the earlier) between iterations 7 and 10 of the beta ?

[rickca]

I go away for two weeks and come back to discover that A+ is now doing PCM to DSD! I had no idea Damien was planning to do this. I sure have a lot of catching up to do.

[MrTHX]

Is there any guidance as to what altering each of these parameters actually does? Like what effect it would have on the sound? I appreciate there might be differences audible of each own DAC etc

 

Steepness: 23

Filter max length: 10,000 (currently in flux)

Cutoff freq: 1.02

Anti-aliasing: 200.0

Pre-ringing: 0.55

 

Make of it what you will.

 

--David

[tipunch]

Bravo Damien!

 

Yes the new beta 2.5.0.11 seems faster now for loading tracks in memory. I can skip tracks or disks with no problems and no more crash with any filters.

 

 

I still hears little the fan noise, but less often and less hardly. It is better that v2.5.0.10.

 

 

I have another last small bug: When I upsample a PCM in DSD128, then after a DSD64 (DSF) play, my DAC Mytek indicates "176.4" instead of "DSD". Audirvana indicates right (entered "DSD64" and in out "DSD64"). But If I press stop, and if later I play a DSD64 (or DSF) my DAC Mytek indicates right "DSD".

I do not think that it is a bug of Mytek because this arrives only with the upsample in DSD. With the upsample 2X or no upsample, the DAC Mytek indicates the right format.

 

MERCI BEAUCOUP Damien

[damien78]

Bravo Damien!

 

Yes the new beta 2.5.0.11 seems faster now for loading tracks in memory. I can skip tracks or disks with no problems and no more crash with any filters.

 

 

I still hears little the fan noise, but less often and less hardly. It is better that v2.5.0.10.

 

 

I have another last small bug: When I upsample a PCM in DSD128, then after a DSD64 (DSF) play, my DAC Mytek indicates "176.4" instead of "DSD". Audirvana indicates right (entered "DSD64" and in out "DSD64"). But If I press stop, and if later I play a DSD64 (or DSF) my DAC Mytek indicates right "DSD".

I do not think that it is a bug of Mytek because this arrives only with the upsample in DSD. With the upsample 2X or no upsample, the DAC Mytek indicates the right format.

 

MERCI BEAUCOUP Damien

 

Other than the Mytek LCD display that is wrong, is the music playing correctly at DSD64?

[damien78]

Gave it a long try today after changing my settings based on our email exchange.

 

The sound in PCM-DSD256 is absolutely amazing with the iFi iDSD Nano together with my SET Tube Amp: it gets a breadth extension in ranges - I can hear texture for the bass instruments, a very clear mid-range too, dynamics are excellent as well.

 

As I still get the 5s and 5 multiple glitch sometimes (coming back to the start of the track sometimes resolves it) as well as the glitch where one side disappears after a high squeak (much rarer), I tried changing some settings as well as launched Activity Monitor this time.

 

CPU usage hovers around 180-193%. I don't see additional peaks when the 5s glitch occurs.

 

I haven't even changed from the default filter, so as I said before, this sounds incredible, just too bad that my current computer cannot seem to be able to keep up.

 

Is it possible to optimise CPU usage further?

 

Well done so far.

 

Is this 5s and 5 glitch due to too slow track loading (loading progress bar not in advance)? If no, do you have a track (from TIDAL if possible) where you always have this issue?

For the one side that disappears, what if you make the volume reduction to -1dB or -3dB? Is it still happening?

 

Damien

[tipunch]

Other than the Mytek LCD display that is wrong, is the music playing correctly at DSD64?

 

Yes, the music is playing correctly...

But it is really a bug of the LCD Mytek or Audirvana transforms me the DSD64 into PCM 176.4?

 

Best regards

[YashN]

Is this 5s and 5 glitch due to too slow track loading (loading progress bar not in advance)? If no, do you have a track (from TIDAL if possible) where you always have this issue?

 

Salut Damien,

 

I don't think I noticed that about the progress bar, I may need to re-check.

 

When it happens on a track, if I go back to the start of the track by using the location slider, the glitch can disappear, so this is why I do not think it is the track itself (I don't have TIDAL, I seem to have an aversion to music subscription services although I am glad about our Netflix sub: I nearly never watch a movie a second time, but I want to own my music for replay).

 

For the one side that disappears, what if you make the volume reduction to -1dB or -3dB? Is it still happening?

 

I will need to try that.

 

Bravo pour le son !

[YashN]

I go away for two weeks and come back to discover that A+ is now doing PCM to DSD! I had no idea Damien was planning to do this. I sure have a lot of catching up to do.

 

rickca, give it a try - amazing SQ and micro-detail resolution and soundstage!

[damien78]

Yes, the music is playing correctly...

But it is really a bug of the LCD Mytek or Audirvana transforms me the DSD64 into PCM 176.4?

 

Best regards

 

Looks like it is, otherwise Audirvana Plus will not display DSD64

[YashN]

Gave it another try:

 

High-pitch 'tick' before one channel goes: happens on Chamfort's 'Souris puisque c'est grave' just as the drum kicks in at around 13s in, repeatable and adjusting vol to -1.2 or -3.2 doesn't change it for me. Checked the track in Audiogate and the Clip indicator doesn't light up.

 

Other 5s and 5 multiple glitch: this seems to happen if one track is playing, then I switch to another track while browsing - observing the progress bar, there are three apparent levels: the normal empty one which is the palest, a slightly darker one for pre-loading(?) during playback, and the initial track start section which is darker.

 

When the glitch occurs, the darker section is to the right of the location cursor, so I believe the player already has data to work on, but what I noticed is that on changing to that track (with no Stop in between), the darkest progress bar will be frozen as well as the location cursor (plus Mac beach-ball for a little while), so you see 0s while the track has already started playing, for a little while.

 

Hope that helps.

[mansr]

High-pitch 'tick' before one channel goes: happens on Chamfort's 'Souris puisque c'est grave' just as the drum kicks in at around 13s in, repeatable and adjusting vol to -1.2 or -3.2 doesn't change it for me.

 

Does this depend at all on the chosen DSD filter?

[Jud]

Gave it another try:

 

High-pitch 'tick' before one channel goes: happens on Chamfort's 'Souris puisque c'est grave' just as the drum kicks in at around 13s in, repeatable and adjusting vol to -1.2 or -3.2 doesn't change it for me. Checked the track in Audiogate and the Clip indicator doesn't light up.

 

Other 5s and 5 multiple glitch: this seems to happen if one track is playing, then I switch to another track while browsing - observing the progress bar, there are three apparent levels: the normal empty one which is the palest, a slightly darker one for pre-loading(?) during playback, and the initial track start section which is darker.

 

When the glitch occurs, the darker section is to the right of the location cursor, so I believe the player already has data to work on, but what I noticed is that on changing to that track (with no Stop in between), the darkest progress bar will be frozen as well as the location cursor (plus Mac beach-ball for a little while), so you see 0s while the track has already started playing, for a little while.

 

Hope that helps.

 

Does this depend at all on the chosen DSD filter?

 

Little bit lengthier discussion of what Damien mentioned: http://www.computeraudiophile.com/f11-software/audirvana-plus-2-multi-channel-sound-5-surround-channels-and-1-lfe-low-frequency-effects-channel-pulse-code-modulation-direct-stream-digital-conversion-28428/index7.html#post540244

 

I believe mansr has recommended the "A" series filters for the iFi Nano (is this correct?).

[Esprit]

Looks like it is, otherwise Audirvana Plus will not display DSD64

Are you sure this is only a display error or the Mytek is reproducing PCM? I'm downloading now the new beta.

[mansr]

I believe mansr has recommended the "A" series filters for the iFi Nano (is this correct?).

 

Yes, to me the A filters sound best with the iFi DACs.

[Esprit]

I know that's probably psychoacoustics, because Damien will tell to me that the audio engine has not changed, between the .10 version and version .11 and it's the same, but (psychoacoustic or slow digestion...) the .11 does not like to me.

 

In live albums the applause are mock, the music seems less natural and less pleasant, less soundstage, not analog.

 

I've put in ramdisk both versions so I can change and choose quite fast which sw listen (but I'm always coming back to the .10 version)

 

So you shouldn't rely on my ears or how my brain processes the audio signal... I'm unreliable (damn psychoacoustics).

 

We'll see the next days

[Le Concombre Masqué]

how about .7?

I know that's probably psychoacoustics, because Damien will tell to me that the audio engine has not changed, between the .10 version and version .11 and it's the same, but (psychoacoustic or slow digestion...) the .11 does not like to me.

 

In live albums the applause are mock, the music seems less natural and less pleasant, less soundstage, not analog.

 

I've put in ramdisk both versions so I can change and choose quite fast which sw listen (but I'm always coming back to the .10 version)

 

So you shouldn't rely on my ears or how my brain processes the audio signal... I'm unreliable (damn psychoacoustics).

 

We'll see the next days

[Lee Henley]

Nice one Damien, just downloaded the latest beta version and the DSD tracks are know back in stereo, many thanks

 

Lee

[YashN]

Does this depend at all on the chosen DSD filter?

 

OK, tried again to determine that: until now I had it left at filter A4 for all my tests (which does sound wonderful when there's no glitching on the iFi Nano (newest upgrade 'Limoncello' today).

 

Changed filter family to B: B0, no glitch (high-pitched 'Tick', one channel down).

 

Changed filter back to A4: glitch.

 

Changed filter family back to A: A0, no glitch.

 

Changed it one last time back to A4: glitch.

 

So A4 appears to have an issue in my system. I have not checked all filter types, but so far, among those above, I still prefer A4's SQ - that will be great if the glitch doesn't occur anymore.

[YashN]

I believe mansr has recommended the "A" series filters for the iFi Nano (is this correct?).

 

It may well be, but in this particular case, one channel goes down, so it's more a glitch than an SQ difference.

[Jud]

It may well be, but in this particular case, one channel goes down, so it's more a glitch than an SQ difference.

 

I was responding with regard to both stability and SQ, though that wasn't clear (my apologies). I guess you'll have to see where on the letter/number scale you find SQ you like without glitching, or Damien puts out a version that doesn't glitch with the filter(s) you currently like, whichever comes first.