Audirvana Plus 3 (official thread)

[Fridolin]

1 hour ago, rodrigaj said:

Onkyo/Pioneer , Sonore, Bricasti

 

Thanks @rodrigaj.

I saw these few examples on their site but couldn't believe that's all.

 

My KEF LS50w work pretty well with Audirvana in UPnP mode, however. 

[rodrigaj]

3 hours ago, Fridolin said:

 

Thanks @rodrigaj.

I saw these few examples on their site but couldn't believe that's all.

 

My KEF LS50w work pretty well with Audirvana in UPnP mode, however. 

I couldn't believe it either.

 

I'm sure there are more manufacturers that work, but it is useful to know that those on their site are the only ones that are guaranteed to work.

 

Your KEF's were used the Darko's review of A+, BTW:

 https://darko.audio/2017/08/network-streaming-comes-to-audirvana-plus/

 

It was in that review that I discovered the work that needs to be done between renderer developer and Plisson. Doesn't sound like it is all that difficult, but it does require a certain degree of openness on the part of the manufacturer. I would contact Auralic and ask them if they wouldn't mind bringing Audirvana+ on board.

 

I tried with PS Audio, but they are looking at their own solution in the form of Octave which is due to be finished by the end of the year. 

[pl_svn]

Mr. Wang (Auralic), a few days ago on their community forum, answered (question was about Qobuz own app) that UpNP Renderer mode will no longer be developed and is not even available on their new products

[Browniesbane]

Sorry if this has  been asked before. I have A+ on an iMac and a MacBook, but I am thinking about setting up a Mac mini  i n a second system that can connects with ethernet to the network  and an mr as suggested  in the chain on massively improving sq. I know A+ only allows two installs, but are

there any discounts for the right to use A+ on a third mac?

 

if not I’ll take A+ off the MacBook, but its nice to keep that option  open as a backup.

 

[Aboulfad]

A bit OT, but I am recently trialing an Anthem MRX-720 AVR, and it supports only that lousy Play-Fi app which doesn’t seem compatible with AV/UPnP. Which led me to trying all different media streamer, Roon, twonky, plex,... and honestly even with the ordinary UI of AV, it just rocks, overall best value from streaming services support, SQ and much more. Now if we can only get both Play-Fi and AV to become BFF 

[MacAudio]

Does it make a difference which SRC I choose if I do not upsample ?

I believe that the SRC is always running , so that it recognizes the native sampling rate of the current track and changes it if necessary.

 

[Jud]

It makes no difference because the sample rate conversion doesn't run when you don't convert the sample rate in the computer. (If you don't convert in the computer, sample rate conversion will be done by your DAC.)

[MacAudio]

Yes, some dacs have this option, some just do it in the background.Rob Robinson (channel D ) once explained that it would be better to do the upsampling (if necessary) in the computer, because the computer calculates in 64bits. Hardware however mostly "only" in 32bit or maximum in 48bit.

[Jud]

13 minutes ago, MacAudio said:

Yes, some dacs have this option, some just do it in the background.Rob Robinson (channel D ) once explained that it would be better to do the upsampling (if necessary) in the computer, because the computer calculates in 64bits. Hardware however mostly "only" in 32bit or maximum in 48bit.

 

Nope, virtually every DAC does this. You are somewhat confusing word length (the "left side" of the resolution numbers, for example 16 in 16/44.1 CD resolution) and sample rate (the "right side," for example 44.1).

 

Macs use a 32-bit word length for internal calculations, and so do many (most?) DACs.  I think iZotope may use 64 bits, but I'm not sure.  Since the thermal noise of the electronics kicks in at about 21 or 22 bits, I don't think this matters a great deal, unless you're in a studio wanting to do inaudible sonic manipulation.

 

The SRC in A+ and DACs is much more about the number on the right side, the sample rate, than it is about word length and internal computations. That's why it's called sample rate conversion rather than word length conversion.  The part of A+ that has to do with word length is integer mode.

[jas]

Jud.  I read many (most?) of the CA threads regarding IZotope settings and was guided/helped by your progressive settings.  I am not a DSD user, though, I use a Yggy, so am interested in the interaction between Steepness and Pre-ringing in non-DSD situations.  I am taking the settings one a a time and am currently testing Pre-ringing at the right end of the scale and landed in the 70's, though the full linear setting wasn't bad, except for some echo-sounds on a few tracks.  Can you, in basic terms, explain how and why these two settings interact?  (I've read that they do) As you can tell, the science is new to me and though I do hear differences in settings I don't know what iZotope would call what I am hearing. 

Thanks for any response you have time for.

[buonassi]

5 hours ago, jas said:

Jud.  I read many (most?) of the CA threads regarding IZotope settings and was guided/helped by your progressive settings.  I am not a DSD user, though, I use a Yggy, so am interested in the interaction between Steepness and Pre-ringing in non-DSD situations.  I am taking the settings one a a time and am currently testing Pre-ringing at the right end of the scale and landed in the 70's, though the full linear setting wasn't bad, except for some echo-sounds on a few tracks.  Can you, in basic terms, explain how and why these two settings interact?  (I've read that they do) As you can tell, the science is new to me and though I do hear differences in settings I don't know what iZotope would call what I am hearing. 

Thanks for any response you have time for.

I too am a big reader of @Jud.  I'll take a stab at answering this in the meantime, though I'm sure he'll elaborate soon.  

 

The higher the steepness of your filter setting, the more the overall ringing will be present.  Regarding phase setting, linear means that there will be a symmetrical impulse response with equal parts of ringing before and after the transient.  Minimum will have take all the ripples of the pre ring and add them to the post ring, so it increases the post ring at the expense of the pre ring.  Anything in between min and lin phase is called intermediate phase and can be helpful if you're trying to reduce phase distortion but enjoy the sound of a min phase setting (as one example). 

 

these two settings work together to shape the transient response.  one defines how much ringing in general, the other determines where that ringing ends up.

[Jud]

19 hours ago, jas said:

Jud.  I read many (most?) of the CA threads regarding IZotope settings and was guided/helped by your progressive settings.  I am not a DSD user, though, I use a Yggy, so am interested in the interaction between Steepness and Pre-ringing in non-DSD situations.  I am taking the settings one a a time and am currently testing Pre-ringing at the right end of the scale and landed in the 70's, though the full linear setting wasn't bad, except for some echo-sounds on a few tracks.  Can you, in basic terms, explain how and why these two settings interact?  (I've read that they do) As you can tell, the science is new to me and though I do hear differences in settings I don't know what iZotope would call what I am hearing. 

Thanks for any response you have time for.

 

Pardon the brief answer, but I think it should be helpful:

 

- Ringing is the product of something called the Gibbs Effect, and occurs at a "discontinuity."  (There is a Wikipedia article if you want to read more.)  The steeper the filter, the sharper the discontinuity, and thus the more ringing.

 

- It is reasonably certain that any echo you heard was not the result of ringing.  Ringing occurs at the frequency of the discontinuity, which for DACs and A+ should be in the ultrasonic range (or at least we'll above inaudibility for adult males).  If your filtering is set to start somewhere in the audible range (really not necessary), it can be quite gentle (low steepness) and therefore shouldn't ring much anyway.

 

There is some dispute (to put it mildly) about whether ringing has audible effects. Those who say it does feel it results in a "smearing" of transients, making them less sharp. But I have never read of anyone claiming ringing from the filtering commonly used in digital audio results in audible echoing.

[jas]

Jud and Buonassi:  Thank you both for responding and for eliminating a possible source of the echoing.  

 

I did read but struggled with the Wikipedia article on the Gibbs Effect, so at the moment, I found both of your explanations more useful. . Since it sounds as if the echoing I am hearing can not be put at the feet of ringing, it occurred to me that the 2-3 tracks I heard it on were new to my system since adding the Yggy and Ayre amp.  What I hear could be, then, baked in to the recording but not having heard these tracks before on my current or any revealing system, I have no real comparative basis to assume that, either.

 

I will eliminate what I can control which is to test steepness at different levels over extended listening periods.   I feel like I am dancing around the target here, but a target that is, (pun intended), within range. 

 

Thanks again.

[mansr]

I feel a few words about transients are in order. In signal theory, a transient is a step or impulse event with unlimited bandwidth. While such signals are useful for mathematical analysis, they do not exist in nature. What is referred to as a transient in music is quite different. Here a transient means an oscillation with a sharp attack. The "edge" does have a fair amount of spectral spread, but it is still very much band limited. The high-frequency content is sufficient to cause a bit of ringing in a filter with a cutoff frequency below 25 kHz. The ringing is, however, nowhere near the levels you see with a synthetic step or impulse. Raising the sample rate to 96 kHz reduces the ringing to negligible levels, and a further doubling to 192 kHz eliminates it entirely for any real music source.

[jas]

mansr:  Thank you; your description helps, too.  It seems, then, that even though the finer points of this topic are escaping me, even at the 44.1 sample rate for the tracks in question ringing can not be the cause of what I am hearing.  I believe the Yggy sample rate to 96kHz, and I have, currently,  A+ set to "power of two" oversampling, so between them the chances of hearing ringing are small, if not non-existent. 

 

Appreciate your input. 

 

[57gold]

Just updated A+ and have a couple of questions that I haven't found answers to searching this thread or the A+ manual:

• In Preferences, Mytek Brooklyn + is Active Audio Device, but DSD is not green at any rate? And can not select to force upsample?
• Anyone know about SoX versus iZotope?  Sounds good, very direct, but so does iZotope.  
• Any other B+ owners with suggestions on how they set up A+ to get best out of B+?

Thanks and best wishes to all CA folks.

[mansr]

28 minutes ago, 57gold said:

Any other B+ owners with suggestions on how they set up A+ to get best out of B+?

This calls for a C++ joke.

[RunHomeSlow]

36 minutes ago, 57gold said:

In Preferences, Mytek Brooklyn + is Active Audio Device, but DSD is not green at any rate? And can not select to force upsample?

 

Try this...

 

[miguelito]

9 minutes ago, mansr said:

This calls for a C++ joke.

Geek.

[57gold]

2 hours ago, RunHomeSlow said:

 

Try this...

 

 

Thanks RHS.  Worked.

 

When playing up-converted PCM to DSD files, does A+ play from memory like PCM?  Or are files to big too big?

 

[RunHomeSlow]

I think it loads in memory as fast as your computer makes that loading bar full ?

[57gold]

Forgot about the grey bar...been using HQPlayer recently.  

 

Upgraded A+ several releases from where I was.  Always sounded good. 

 

Really impressed with newest version and the new Mytek Brooklyn +, PCM 384 and DSD 128.  New chips, clock, analog circuit of B+ versus the DSD192 is pretty incredible.  

[Jud]

14 hours ago, mansr said:

Raising the sample rate to 96 kHz reduces the ringing to negligible levels, and a further doubling to 192 kHz eliminates it entirely for any real music source.

 

192kHz original sample rate, conversion from 44.1 to 176.4 or 192, or both?

[MacAudio]

Are there any advantages when using upsampling? I always read here of "bought" disadvantages like ringing  or other compromises ! ???

I can even extrapolate to DSD, but the CPU of my Mac is of course more stressed ... it does not sound any better. Same  with x2 or max up sampling. Default settings in soX or izotopia. Is there a guide to change the parameter / SRC sliders ?
Also the Dac's  switchable upsampling of e.g. 44.1kHz to 192kHz is not really better. It sounds a little different, but not really better. The same is by switching the filter from sharp > slow. 

setting to - slow -  i can recognize a subtle more stable mono-mid ,or a slightly better , perhaps more natural deepness , not imaging !
The increase of the so called Imaging is often a result of a worse designed dac.

Lucas van der Mee, Sr. Design Engineer, Apogee Electronics:

A good converter maintains the energy and does not widen the overall image. Especially the latter is very often seen as an advantage; wide seems better at first. But in the end it is not, as a matter of fact the worse the converter the wider the image. A good converter should give you depth, the width as intended and a solid mono response for parts mixed to the center.

 

[MacAudio]

That's why I used the Apoggee Mini Dac for many years.
This was designed by him.
Unfortunately, the FW interface did not work 100% perfectly, and that's the reason why I switched to another dac.

This dac now is able to upsample via chip and has DSD function. But sound wise  there is no advantage.

But since I use Audirvana I mean to hear a minimal sonic advantage. Even without upsampling.