Massively improve the SQ of computer audio streaming?

[Superdad]

11 minutes ago, Ralf11 said:

YES, it is.  However, I did not make the claims that you did.

 

Well I saw in a recent post (different thread) that your are at last considering moving past Airplay streaming.  The rest of your system looks nice (I like Magnepans), but Airplay with AppleTV3 is limiting the subtleties... 

[Ralf11]

I have a variety of sources; that is one - not that 48 kHz is a SQ problem.

 

Now, please stop trying to divert the discussion and let your friend answer the question

 

- new thread started based on S'Dad's post

 

back to Lack of Controls !!

[asdf1000]

3 minutes ago, Superdad said:

Airplay with AppleTV3 is limiting the subtleties... 

 

Yup AirPlay via Apple TV re-samples everything to 48kHz so bit-perfect playback is not possible (unless streaming a 48kHz source maybe...).

 

 

[Ralf11]

use the other thread please

 

 

Francoise apparently did no testing so the mu metal notion is unconfirmed

[asdf1000]

10 minutes ago, Ralf11 said:

use the other thread please

 

 

Francoise apparently did no testing so the mu metal notion is unconfirmed

 

On topic... one way to potentially (massively?) improve the SQ of computer audio may be to move on from an Apple TV3's optical output (which is non bit-perfect with your 44kHz playback)... ?

 

Nice speakers though - love Maggies.

[sandyk]

1 hour ago, Em2016 said:

 

On topic... one way to potentially (massively?) improve the SQ of computer audio may be to move on from an Apple TV3's optical output (which is non bit-perfect with your 44kHz playback)... ?

 

Nice speakers though - love Maggies.

 

Deleted.

[Ralf11]

3 hours ago, Em2016 said:

 

On topic... one way to potentially (massively?) improve the SQ of computer audio may be to move on from an Apple TV3's optical output (which is non bit-perfect with your 44kHz playback)... ?

 

Nice speakers though - love Maggies.

 

Thx - upsampling is non-bit perfect?

 

There is now another thread on the TV3 with 3 lines of evidence from S'dad, which I am looking at...

[Ralf11]

4 hours ago, mansr said:

You already know the answer.

 

It would have been easy for afrancois to simply say "I did xx with mu metal and while I did no blind testing, there does -seem- to be an improvement."

 

 

[asdf1000]

19 minutes ago, Ralf11 said:

Thx - upsampling is non-bit perfect?

 

Correct - up and down-sampling are forms fo DSP.

 

Keeping in context of the Apple TV 3 specifically, I have no idea if it's up-sampling of 44 to 48kHz is regarded as good or bad or whatever.

 

But if you like the internal DSP your DAC does (if it does DSP - which DAC?) then you MAY not want the bits messed with, before they arrive at your DAC's input.. depending on your personal preferences (what your ear/brain system likes).

 

Maybe it makes no difference to your ears? That's a good problem to have (good for your wallet...)

 

Please note I'm very careful not to claim definitively what sounds good/better/best... AirPlay multi-room audio (from a Mac iTunes Server) got me through years of music enjoyment (well before I got onto Roon), so I'd never put anyone down using AirPlay.

 

 

 

 

 

[afrancois]

51 minutes ago, Ralf11 said:

 

It would have been easy for afrancois to simply say "I did xx with mu metal and while I did no blind testing, there does -seem- to be an improvement."

 

 

Of course, I did not do any blind or ABX test. I’m the only one in the household that can change the cables.  Now with the mu-metal they are very stiff.

In all honesty, I think very few people and not only here at CA do double blind tests. Even at PS Audio Paul explained they don’t do it.

I’ve been optimizing my system for over a year now and while not all changes did at first glance seem to improve my system, I can tell you with absolute certainty that it now sounds a whole lot better than a year ago. This would not have been the case with an attitude “if you can’t prove it, it isn’t true’. The best example is Audiolinux. Nobody can prove it, but everybody hears it. And so did I, while only changing the server to a simple NUC with AL. 

[Jud]

1 hour ago, Ralf11 said:

 

It would have been easy for afrancois to simply say "I did xx with mu metal and while I did no blind testing, there does -seem- to be an improvement."

 

When I asked once upon a time on this site whether people had employed blind testing in the process of purchasing any of their components, not a single person, including folks who have frequently called for blind tests to prove effectiveness here, said they had.  If you employed blind testing in the process of purchasing any of your components, you would be the first I know of on this site. Did you?

 

So rather than demanding a superfluous"I didn't use blind testing" from everyone else, it's probably better just to assume that's the case in the absence of a statement to the contrary.

[pkane2001]

12 minutes ago, Jud said:

 

When I asked once upon a time on this site whether people had employed blind testing in the process of purchasing any of their components, not a single person, including folks who have frequently called for blind tests to prove effectiveness here, said they had.  If you employed blind testing in the process of purchasing any of your components, you would be the first I know of on this site. Did you?

 

I know of at least one such person.

[Ralf11]

besides me?

 

that'd be 2

 

it is one thing to do something; another to advocate for something in public - the latter, if not well described, can mislead others

[esldude]

8 hours ago, Kal Rubinson said:

That was a direct quote.

Below is my original post on the matter:

 

People can 'hear' gigahertz radar signals if strong enough.  But this is not hearing in the normal sense.  Bone conducted ultrasonics can show up as perceived lower frequencies, but listening via the air such things are of no real consequence. 100 khz or so sound high enough in level enters the eye sockets and vibrates some auditory nerves.  Again in normal situations of hearing they are of no consequence.  Infra sound can be felt, and if high enough in level it can vibrate us enough our vision is effected.  Would you say we see infrasound?

 

I put the marks around hear about radar because while people will hear it in their head it isn't really heard via the ears.  Nothing about eye sockets there. 

 

Then the comment about bone conducted ultrasonics.  Nothing about eye sockets there.  Though it might have read as if the following sentence went with that. 

 

So then the statement that 100 khz or so sound at high level enters the eye sockets and vibrates auditory nerves.  So the latter was not part of the radar "hearing".  

[Jud]

6 hours ago, Ralf11 said:

besides me?

 

that'd be 2

 

it is one thing to do something; another to advocate for something in public - the latter, if not well described, can mislead others

 

What was it you blind tested in order to purchase (I'm curious, since it would seem to take some doing)?

[mansr]

9 hours ago, Ralf11 said:

upsampling is non-bit perfect?

It alters the bits, so no. The term bit perfect doesn't really make sense where an intentional alteration takes place.

[Kal Rubinson]

7 hours ago, esldude said:

So then the statement that 100 khz or so sound at high level enters the eye sockets and vibrates auditory nerves.  So the latter was not part of the radar "hearing".  

Understood.  Still questionable.  

[wgscott]

11 hours ago, Jud said:

 

When I asked once upon a time on this site whether people had employed blind testing in the process of purchasing any of their components, not a single person, including folks who have frequently called for blind tests to prove effectiveness here, said they had.  If you employed blind testing in the process of purchasing any of your components, you would be the first I know of on this site. Did you?

 

So rather than demanding a superfluous"I didn't use blind testing" from everyone else, it's probably better just to assume that's the case in the absence of a statement to the contrary.

 

(1) I have.

 

(2) The point is not that a potential purchaser "should" conduct testing, any more than it is to have a potential car purchaser do insurance industry safety tests, or Consumer Reports type testing.  Rather, the point is for reviewers or others to do tests to establish the veracity of marketing claims, rather than simply parrot them.  Also, if someone makes a claim that (eg) modifying a mac mini is essential to obtaining high quality sound, there should be some objective evidence to substantiate the claim.

[Jud]

3 hours ago, wgscott said:

 

(1) I have.

 

Same question - what component(s), and can you describe a little about the process?  I'm thinking it would be kind of a PITA, so I'm curious.

 

3 hours ago, wgscott said:

(2) The point is not that a potential purchaser "should" conduct testing, any more than it is to have a potential car purchaser do insurance industry safety tests, or Consumer Reports type testing.  Rather, the point is for reviewers or others to do tests to establish the veracity of marketing claims, rather than simply parrot them.

 

Agreed.

 

3 hours ago, wgscott said:

Also, if someone makes a claim that (eg) modifying a mac mini is essential to obtaining high quality sound, there should be some objective evidence to substantiate the claim.

 

Or you can simply decide, as we all do, what if any credence to give such claims in the event that only subjective experience is forthcoming.

[Ralf11]

7 hours ago, Jud said:

 

What was it you blind tested in order to purchase (I'm curious, since it would seem to take some doing)?

 

Oppo - very simple for anything that feeds into a pre-amp - on here with key search words oppo, cal audio icon

 

I tried to compare big new & small old Maggies as closely as possible but obviously a huge undertaking to do any blind testing with speakers.  I've described Maggies vs. Vandies previously - upshot: can't easily blind but 3-4 months of comparison allowed me to make the difficult decision in the mid-90s.

[Ralf11]

6 hours ago, mansr said:

It alters the bits, so no. The term bit perfect doesn't really make sense where an intentional alteration takes place.

 

Quite true.  The Apple TV3 upsampling degrading SQ issue has reared its head again BTW

[Ralf11]

52 minutes ago, Jud said:

what if any credence to give such claims in the event that only subjective experience is forthcoming.

 

I agree with this.  A blind test is always best, but I pay a lot of attention to Rajiv and John's Eye - both are very good at verbalising their perceptions.

 

One more tidbit: no need to test much at all if ergonomics or aesthetics would lead you to buy X over Y anyway.  And that is a perfectly valid consumer choice.

[Jud]

1 hour ago, Ralf11 said:

One more tidbit: no need to test much at all if ergonomics or aesthetics would lead you to buy X over Y anyway.  And that is a perfectly valid consumer choice.

 

Yes, for example I use a streamer (microRendu) because the desktop running the software I like is too big and acoustically noisy to go in the living room where the main system lives.

 

I also have a NUC (though 12 volt with a relatively powerful processor) and an Audiolinux license, so I can play with that if I'd like. But Jesus' noise measurements seem to indicate I should expect more noise with that setup unless I buy a much better power supply, so at this point I don't know whether I'll bother.  I've tried the NUC with another couple of Linuxen, and subjectively I preferred the microRendu.

[rickca]

23 minutes ago, Jud said:

I've tried the NUC with another couple of Linuxen, and subjectively I preferred the microRendu.

So you haven't tried AL/ramroot on your NUC yet?  Of course, this is what Jesus calls 'forum jitter'.

[Jud]

1 hour ago, rickca said:

So you haven't tried AL/ramroot on your NUC yet?  Of course, this is what Jesus calls 'forum jitter'.

 

Nope. Not sure it would be the magic elixir with a NUC that isn't one of the particular ones people are talking about (has a more powerful CPU) using the inexpensive SMPS power supply that came in the box.

 

I've used the NUC with a reasonably stripped-down Linux (Ubuntu Server with some of the unneeded servers turned off).  I've tried ramdisks and low latency kernels too.  They didn't seem to me to make sufficient difference to what I heard to overcome the power supply disadvantage. And I'm quite happy with the sound I've got.  But maybe one of these days when I have more time than I do now, I may try it out of curiosity.