Audirvana Plus 3 (official thread)

[wgscott]

Somehow I knew you would post a thread like this. (But isn't it a month early?)

 

Audirvana Plus 3.0 is great. I am glad he compiled it linked to libreadmind.3.0.0.dylib. Works better than any remote control.

[wgscott]

I just submitted a bug yesterday. But you guys already knew that.

[wgscott]

Won't be long before upsampling is filed away

 

How do you think a DAC normally works?

[wgscott]

Personally speaking, I could care less. My views are totally based on the music I hear. I find upsampling alters or reconstructs the sound in an artificial manner that has limited appeal.

 

With very very few exceptions, DACs upsample, so unless you are listening with a non-upsampling override setting (which will sound horrible due to aliasing and other artifacts), your expectation bias is what you are hearing. Having software do so instead (or in addition) simply gives you more control. If you don't like those options, no one is suggesting you have to use them, but telling other people that it is damaging to the music is kind of ill-informed, despite the fact that you "could [sic] care less."

[wgscott]

What's the issue with Teac DACs? I can't seem to find it on the website (and I have a Teac NT-503 fwiw).

 

Nevermind; I found it:

 

MARANTZ, TEAC, DENON, LUXMAN, SONY, YAMAHA DACS: SPECIAL BUILD

 

There is currently a compatibility issue between some TEAC, Marantz, Denon, Luxman, Sony DACs with the MQA device detection routine.

This causes the DAC USB interface to fail to initialize.

If you have a DAC from a brand above, you should get the special build that is not including the MQA device detection routine.

[wgscott]

iFi Audio DAC's have a BitPerfect setting where they do not upsample. It is even the best sounding setting!

 

So does my DAC.

 

It might sound the "best" to you, but do you know what happens with the digital to analogue conversion when neither software nor hardware upsampling is used?

[wgscott]

Upsampling is generally caused by an obsessive compulsion to fiddle, in a hopeless quest to extract another 0.1% of "sound quality" from your audio equipment. It's the very definition of expectation bias.

 

That is certainly one point of view.

[wgscott]

It needs a well designed analog filter.

 

Add-on: that's the main reason why up sampling was introduced: easier and cheaper filter

 

If you don't filter or upsample before the conversion to analogue, you will have high-frequency roll-off. (That might be why it sounds better.)

 

 

(White and pink curves are NOS and not filtered). Archimago's Musings: MEASUREMENTS: Tascam UH-7000 USB Interface (Part II: As an ADC)

[wgscott]

I'm with you on that. Software is just another consumer (or B2B) product. You wanna stay in business? Word to the wise, don't piss off your customers - especially over 40 bucks. It's just stupid business. Damien's a fantastically talented programmer and seems like a great guy. But this kind of thing won't serve him well. I've been in consumer products my entire career and I can tell you, at times like this, you suck it up and help your customers. It all comes around.

 

Damien started this as an open-source hobby project, not as a money-maker, and he was hoping that more people would get involved. Since they didn't, or couldn't, he's selling it for a nominal fee in his spare time, but it isn't like this is some large business with six people answering phones at a help desk call center while filing their nails and hanging out at the water cooler. Get a grip.

[wgscott]

OK, you seem to know all about his business model. Or lack of, as he's just a saintly chap doing the world a huge favor. LOL :-)

 

No, but he is a fellow CA member, who decided to take a shot at writing his own audio software. He is one of our own, and he is a great guy who is doing a lot of us a huge favor.

[wgscott]

If it was a 6hr missed cutoff, it would just be better to give the guy a free upgrade and stop all the needless and distracting banter here. No need to be overly rigid!

 

Although I agree with you, it seems apparent that there will be those who feel compelled to whine and bitch about every apparent shortcoming and view it as an atrocity, due to an infinite sense of entitlement.

 

This, by the way, is precisely why I will never write software.

[wgscott]

Some people will find something to complain about, no matter what.

 

Exactly.

[wgscott]

I

And I cant use any version because of that NAS mounting mumbo jumbo.

 

I can use it with my NAS. Try mounting with this: https://osxfuse.github.io

[wgscott]

A hobby is one thing, once he starts charging and making money that's another, and that's when he has a responsibility to his customers....or not. Again he doesn't need anyone answering phones, just answer the few questions out there.

 

So would you rather he charged several hundred bucks like some of the other options, and answered every inane email within 6 hours of its receipt from people who can't be bothered to RTFM? I know I wouldn't.

 

How often does Apple answer your emails about OS X or iTunes?

[wgscott]

No, I apologize. My brain saw "Dislike" when the 1 was next to "Like". My fault...and thank you!

 

Once you like, something, if you change your mind, you can only dislike it after you unlike it (which also gives the option of re-liking it).

 

Integer modes work exactly the same way. Except different.

[wgscott]

now I have to pay another 39 dollars to upgrade

 

Why do you have to upgrade?

 

I'm a huge fan, but haven't upgraded yet. Do it when you feel you have got your money's worth out of your current investment and/or need the new functionality. Version 2 hasn't stopped working the same as it did the day before version 3 was released.

[wgscott]

19 minutes ago, OldBigEars said:

TommyCat - agreed on all points.  I went through an extended loop - testing all kinds of upsampling to high res PCM and DSD, as well as fiddling around with  filtration tweakery.  Eventually I came to the realization that non-upsampling sounds best on my system.  It's like drinking coffee without lumps of sugar and dollops of milk distorting the flavor.  I just like it better. And yes, good cable selection makes a far more satisfying contribution to enjoying the music. YMMV :-)

 

I forget.  Which non-upsampling DAC are you using?

[wgscott]

15 minutes ago, OldBigEars said:

I use a Hegel HD25.  According to the LED display, it processes redbook files at 44 kbps.  I dunno, does that mean it upsamples them?     

This is all I found:  http://www.avmentor.net/reviews/2014/hegel_hd25_1.shtml

Quote

Main digital processing in the HD25 is performed via three Asahi Kasei chips. The trio includes the AK4115 digital interface, the AK4127 sample rate converter and -finally- the AK4399 d/a converter, which operates at 32bit. Of these three components, the role of the SRC seems to need some clarification. Hegel does not mention that the converter performs any kind of upsampling (all AK DAC chips can handle all standard sample rates, anyway) so the usefulness of a sample rate converter at this point could be associated with the re-clocking the company seems to carry out, but without giving more details, one cannot say for sure. 

Someone who knows a lot more about this stuff than I do would have to help us out here. Sorry.

[wgscott]

10 hours ago, dahlberg said:

Is there a convolver plugin that will work with Audirvana ?

Anything that has an AU (eg: Dirac [beta]).

[wgscott]

29 minutes ago, dahlberg said:

 

I already have the eq functions in the dsp-xo, what I'm after is impulse and phase response compensation. The frequency response in this room is actually quite good without any eq directed at the room response. I'm only using these functions for dipole roll of compensation. If I'm reading it right Dirac would work right away or ? Only thing missing would be the measuring microphone. I'm running a Mac mini and [Audirvana?] 3.0

 

The nice thing about Dirac is it is completely self-contained and relatively user-friendly.  It's main limitation is that it doesn't explicitly do dsp crossovers, so either (a) it isn't for you, or (b) you have to do your dsp-xo first, and then apply Dirac to the output (which should work fine, but you might not want to do things that way).  I use a $100 calibrated USB mic.  You could probably borrow one from someone here or locally, to check out the free trial of Dirac.  We have several threads in the DSP/room correction sub-forum on this, including discussion of the advantages and disadvantages by people who understand this better than I do (including Flavio (flak) from Dirac).

 

[wgscott]

20 minutes ago, steveoat said:

When will the version for Windows be available?

 

A few weeks subsequent to the Vax-VMS and SGI-Irix versions.

[wgscott]

Does version 3.X still suffer from the "Strong issue"?

 

 eg:

 

May 23 12:25:51 tv-stereo-mini Audirvana Plus[64009]: Strong issue, need to restart playback
May 23 12:25:51 tv-stereo-mini Audirvana Plus[64009]: Restarted: 0

 

[wgscott]

Does the problem persist if you reboot the computer?

 

Does the same problem occur if you create a new user account (for diagnostic purposes) and run Audirvana from that?  If not, it implies it is a corrupted preference file so something similar in your home directory.