Massively improve the SQ of computer audio streaming?

[asdf1000]

12 hours ago, diecaster said:

Why would a NUC with crappy clocks that makes a lot of noise on its USB ports running AudioLinux in RAM be a better endpoint than the SOtM Ultra with a super duper external clock and low noise purpose built motherboard with ultra quiet power and a fantastic USB implementation?

 

From Rob Watts:

"As you know, RF noise creates noise floor modulation, as the intermodulation distortion from random RF noise is a white noise modulated by the wanted signal. This then results in noise floor modulation, and is very very audible. It accounts for the things sounding brighter and less smooth; additionally, when you reduce RF noise, things sound considerably warmer and darker, and one consequence of this is perception of tempo - more midrange gives the impression of a slower tempo, as individual instruments have much more body.

Now if somebody prefers the brighter sound from more noise floor modulation, then fine - that's their taste and preference. But it's not accurate."

 

 

In terms of tweaking with the digital chain, he says that it's technically better to go with a source/chain that results in a smoother, warmer, darker sound...

 

So it MAY be possible that those reporting more detail, more dynamics, are simply enjoying the results of increased RF, causing increased IM distortion and noise floor modulation...

 

Possible. Hard to know for sure.

 

The Sonore ultraRendu + Uptone LPS-1.2 is the smoothest, warmest sounding USB source/chain I've had in my system, to my ears.. I've tried a powerful i7 NUC as an endpoint too but not the lower powered NUC being discussed recently in the other thread.

 

 

[asdf1000]

1 hour ago, lmitche said:

we are starting to see increasing returns, not little subjective impact.

 

Important to note the very obvious though, that these "increasing returns" observed are also highly subjective...

 

And that's ok of course. I enjoy reading all the observations of all the various tweaks we all share (myself included).

[asdf1000]

34 minutes ago, lmitche said:

only because we have no objective measurement tools but our ears.

 

Absolutely - the ears connect to the brain and that's where it all becomes subjective (for now).

 

To repeat, I enjoy reading all the observations of all the various tweaks we all share (myself included). So there's no discrediting of subjective observations from my end.

 

I just enjoy the discussion of the potential technical mechanisms at play I guess - whether they are accurate or not. In the end I just have to trust me own ears too, like everyone.

 

[asdf1000]

3 hours ago, jabbr said:

in conclusion: spend your time obsessing elsewhere

 

Agreed. Like RF affecting the analogue components/electronics.

[asdf1000]

19 minutes ago, beerandmusic said:

 

Will your streamer selection make a difference here?

 

Possible? They can be noise/RF sources, no?

[asdf1000]

7 minutes ago, PeterSt said:

But yes, it surely reads like that -> Nothing can make a difference anyway.

 

Actually, I think many would agree that lots can make a difference. I just wonder (sometimes) if what some believe to be a massive improvement in SQ, is actually possibly technically worse... i.e. enjoying the effects of slightly more RF (caused by whatever mechanism/s)... Everyone is free to enjoy whatever of course. In the end we trust our ears and enjoy the music.

 

But at the same time, I personally like learning about potential technical mechanisms that MAY be involved. So the (free and open) discussion is always interesting to me.

 

[asdf1000]

1 minute ago, PeterSt said:

There's so much going on, you don't want to know ...

 

Ha. I like this.

[asdf1000]

18 minutes ago, PeterSt said:

Would be quite fruitful for many. But it doesn't work like that. The one camp posts a 10,000 posts in a followers thread and the other camp calls that crazy by means of one post. Both do not lead to discussion.

 

Agreed! So let's discuss.

 

What do you think is happening with numerous people thinking an Intel NUC + AudioLinux sounds "better" than dedicated (also Linux based) ultra low noise output networked USB sources? I say "better" because they claim better. I have no doubt it can sound "different" . 

 

Is it possible they're enjoying the extra RF of the (noisier) Intel NUC, resulting in 'more detailed' sound? Or something else might be at play? Or 10,000 things may be at play?

 

Or do you think they may be hearing something technically better? Better than a custom designed ultra low noise networked USB source (for example)?

 

Your theories (guesses) here are fine - your guesses are based on knowledge and lots of practical experience.

 

[asdf1000]

6 minutes ago, diecaster said:

 

Keep in mind that Peter has a vested interest in his PC and software and that his ideas and products do not mesh with the server/endpoint model.

 

Of course. But he's a DAC designer too. I've had these conversations with a few DAC designers (including the designer of your DAC). They all have some bias in some way or another - naturally, so that's fine.

 

It's still always an interesting conversation and you will often find several of the most famous DAC designers tend to agree on certain topics, more than disagree...

 

I would rather see his opinions/theories, than not... otherwise it makes for a boring thread with nothing much left to discuss ?

 

[asdf1000]

6 minutes ago, lmitche said:

Peter can afford to put his money where his mouth is, buy a NUC/AL have a first hand experience and then come back to talk to us about his theories. 

 

Respectfully, if you only want discussions around first hand experience, you already have a thread for that ?

 

Please don't shut down discussion (even if theoretical) here. There's already a thread for that too.

 

As per the opening post here:

 

"The purpose of these thread is to allow free and open discussion about some the ideas and solutions floated and used in the "A novel way to massively improve the SQ of computer audio streaming" thread. In that thread, you aren't allowed to make posts that question the validity of what is being done there."

 

[asdf1000]

1 hour ago, PeterSt said:

If that other camp only would listen too, then we could really have a discussion. Like "how the heck is it possible that ...". Would be quite fruitful for many. But it doesn't work like that. The one camp posts a 10,000 posts in a followers thread and the other camp calls that crazy by means of one post. Both do not lead to discussion.

 

Well, it seems you and I are the only ones that are interested in discussing potential technical mechanisms.

 

I guess we give up on this thread and if I visit Europe next year, we discuss these things over a beer maybe! ?

 

[asdf1000]

6 minutes ago, diecaster said:

@Em2016 @Jud I am just pointing out that Peter has a set of products and to expect him to be biased toward them.

 

Noted. And that's obvious and very natural. As I noted above, I've had these discussions with a few well known DAC designers. No surprise, all of them have something to sell... so there is always some bias. It should go without saying.

 

When discussing potential technical mechanisms though, for certain things, you will often find most of the well renowned designers agree more than they disagree (as I mentioned before)...

 

So the discussions are always interesting... to me... maybe I'm alone here lol

[asdf1000]

4 hours ago, jabbr said:

so there is no reason to think that audiolinux running NAA would be better than Jussi's linux images including NAA

 

Agreed, I tried to hint that in the other thread. Unless AudioLinux knows how to better optimize Linux for HQP NAA than Jussi himself?

 

6 hours ago, lmitche said:

Dev tells us that Intel engineering is monitoring the "Novel Thread". We know that Jussi  works for Intel. 

 

Is it possible that Intel did something to purposely increase the SQ of the NUC product line? If so, what could that be?

 

 

Jussi also  recommends the UpBoard Gateway for NAA, running his bootable NAA image.. Intel is a backer of Up...

 

I have one. Do I think it’s massively “better” than an ultraRendu? No... It’s a tad brighter sounding to my ears but some may say this is better “detail”.

 

But it’s superb value for the dollar.

 

Ideal for me would be Jussi’s bootable NAA image on the Rendu’s (not supported by Sonore of course because they have their own Linux based OS) but that’s a thing for another day.

 

[asdf1000]

1 hour ago, Johnseye said:

Jussi did not include ramroot until I asked for it the other day.  When I side by side compared Jussi's NAA with Audiolinux running HQPlayer NAA, both in ramroot, I had a very difficult time hearing a difference if there even was one.

 

Noted.

 

On my UpBoard (recomennded NAA by Jussi) , even before I used the ramrooted NAA image, I heard no difference between AudioLinux and Jussi’s bootable image. Both sounded the same and both a tad brighter than my ultraRendu. Brighter for me may be more detailed for others. I don’t know which is technically better but the brighter option causes me a little more listening fatigue over longer durations.

 

I then loaded the ramrooted NAA image that you asked for and the differences were  the exact same as the above, for me.

 

The ultrarendu stays for me. The same way the ISO REGEN + LPS-1.2 stays for @lmitche after his NUC?

 

[asdf1000]

Just now, lmitche said:

I'll put a pumpkin between my NUC and DAC if it raises SQ. I'm not wedded to anything.

 

Hehe I like that. 

[asdf1000]

11 minutes ago, barrows said:

Brighter usually means more RFI influencing sound quality.  This can create "false detail" in some systems which may be perceived as "better" (again in some systems), but is actually an artifact.

 

Ya. I shared a Rob Watts quote about that in the other thread and they didn’t like it.

 

What they don’t appreciate is I’ve been told the same by a few other famous DAC designers in private..

 

So any tweak in my digital chain that makes things sound more “detailed”, I avoid..

 

Just a personal thing. I won’t make any claims but everyone is free to discuss with their own DAC’s designers also, the same way I have. 

[asdf1000]

8 minutes ago, Johnseye said:

 

Different hardware influencing things maybe.  I think it was you who pointed out to me that Jussi was providing a bootable NAA.  I've only tried the ramroot version.  There may be no, or little difference but can only speak to what I've tried.

 

Noted John.

 

Yes I pointed you to the bootable image and you got the ramrooted option made! Helping each other to do all this fun tweaking.

 

Whether we agree which is technically better, worse, same whatever - it doesn’t matter in the big picture of life - none of this is life and death of course.

 

But it is interesting to share and read all these observations and play armchair expert (we all do it) to guess what may be going on.

 

[asdf1000]

Here is the fanless Up Board that Jussi recommends for NAA:

 

https://up-shop.org/home/81-up-gws01w4g-memory32g-emmc-boardwo-vesa-plate.html

 

Works very well indeed with DSD512 and PCM768kHz rates.

[asdf1000]

23 minutes ago, PeterSt said:

So are we talking the same or are we. Brrr.

 

Didn't know you were a Cardi B fan Peter ?

 

 

[asdf1000]

7 hours ago, lmitche said:

I have personally built and installed seven NUCs in seven different systems for seven different people. Within seconds, one of the first things people say is, "hold-on I have to lower the volume", usually followed by some form of OMG. Over here, my average listening volume is four db lower than before the NUC.

 

Thanks for sharing this observation. When I compare a direct iMac USB connection to DAC vs ultraRendu connection to DAC (as an example), for me there is a greater desire to crank UP the volume knob when using the lower noise networked USB source (ultraRendu, for example only).

 

There is a greater desire (for me) to turn down the volume knob with a noisier USB source (iMac, for example)...

 

 

[asdf1000]

11 minutes ago, lmitche said:

Anything I write will be sliced to death.  Screw it, here goes.

 

 

This is not life and death. Thanks for replying.

 

11 minutes ago, lmitche said:

I agree with the statement above. This is different.

 

What I hear is like a flashlight out of focus. With the NUC there is perception of an increase in intensity (volume) as the source of the sound becomes more focused. This happens especially in the field between speakers.

 

Cool thanks, I respect this observation. Most of my critical listening is done with headphones - no room effects at play. No leakage loops at play (when I do critical listening).

 

But  I know very well some of the shortcomings of headphones, compared with speakers.

 

I'm going to try the NUC7PJYH if I can get one cheap and will try it in a speaker setup too. If it doesn't replace my ultraRendu, it'll make itself useful in a 2nd setup for casual, non-critical listening.

 

[asdf1000]

3 minutes ago, lmitche said:

You will enjoy that setup for sure. Power is critical. Put your best power supply on the NUC. That model will just make it using an Lps1.2 with no monitor or keyboard.

 

Cool, thanks. I have an LPS-1.2 already...

 

I asked you in the other thread but never got a reply... have you had an ultraRendu + LPS-1.2 in your system (ultra not microRendu)?

 

To compare with your current chain of NUC+LPS-1.2 + ISO REGEN+LPS-1.2 ?

 

[asdf1000]

51 minutes ago, lmitche said:

I would not be surprised that DACs get the bits correctly, but can we prove it one way or the other? Psaudio has some software that does this in their DACs.

 

Very easy. There's a very good and simple bit perfect test... play DSD over PCM (DoP) to a USB DAC that supports DoP... to it's highest supported rate if possible or whatever rate you can.

 

You will know if playback is not bit perfect... 

 

 

[asdf1000]

8 minutes ago, Dev said:

 

How many times folks have heard not-bitperfect playback over Dop ?

 

Not sure what you mean?

 

A few DAC designers have told me DoP is a great bit perfect test... you will know (audible dropouts..).

 

[asdf1000]

Just now, Dev said:

 

If you don't hear any drop outs, does it imply bitperfect playback ?

 

They (a few DAC designers I've spoken to) say DoP is a great test.. so yes I guess this is what they are saying.

 

Do you believe it's not a good test?