A novel way to massively improve the SQ of computer audio streaming

[Qhwoeprktiyns]

Here for Linux : https://wiki.debian.org/BridgeNetworkConnections

[Qhwoeprktiyns]

I set it up in Linux, with an Intel board running RoonServer and a RaspberryPi running RoonBridge. Took me two minutes.

 

Both have fixed IP addresses set on my router.

 

On the Intel board (Ubuntu in my case):

apt-get install bridge-utils

 

then in /etc/network/interfaces:

auto lo br0

iface lo inet loopback

iface br0 inet dhcp

bridge_ports eth0 eth1

 

Done ! Works, but have not spend time listening yet

[Qhwoeprktiyns]

Is it now safe to assume that this 'novel way' is only for those those using the microRendu and sMS-200 as a Roon Endpoint?

 

So:

- Can't be used with the device as a NAA for some reason.

- Irrelevant/no improvement for AirPlay.

- Irrelevant/no improvement for Squeezelite.

- Irrelevant/no improvement for DLNA/MPD & even possibly can't be used anyway if your UPnP/DLNA controller app is on a wireless device (WiFi devices only accesssible via UPnP from the router's DHCP server handled devices, ie, can't be accessed from another subnet).

 

Why would this solution not be relevant using another sbc as a roon endpoint?

[Qhwoeprktiyns]

I did say it was just for @hopkins

 

I set up a bridge between RoonServer and RoonBridge on a RaspberryPi2 equipped with a Kali Reclocker (with I2S output going directly into a "Full Digital Amp"), but went away on a business trip for two weeks before I could sit down and have a serious listening session - am eager to be able to compare it when I get back.

[Qhwoeprktiyns]

For those struggling with their machines because of lack of dual native LAN ports or for those looking to start from scratch and are looking for a relatively inexpensive and simple solution to run Roon (and possibly HQP), this device might be an elegant solution:

 

https://www.amazon.com/Fanless-industrial-processor-wireless-Windows/dp/B01CQL9GKK/ref=sr_1_8?ie=UTF8&qid=1484513553&sr=8-8&keywords=qotom+i7

 

This device is interesting to me because of a few features:

 

1. It utilizes an i7 4500U processor with a fairly light electrical footprint (TDP 15w) and yet, it has a very desirable 4MB secondary cache which is ideal for Roon. Because it sports a dual core CPU capable of 3GHz speeds, it could be handy for HQP for those with more modest upsampling ambitions.

 

2. You can power it with a 12V LPSU.

 

3. It is housed in a fanless chassis and what is hopefully a relatively resonant-free aluminum chassis.

 

4. Compact form factor = short signal paths = low impedance

 

5. Price includes 64GB SSD which should be sufficient for any OS + RoonServer + Roon database

 

6. Incorporates 2 mini-PCIE slots which is huge for me because storage on the PCIE bus (whether it be OS or music storage) results in better SQ compared to SATA (possibly due to lower latency). Also, this obviates the need for expensive audiophile-grade SATA cables (which do make a difference with SATA drives). In my own listening comparisons, PCIE SSDs result in improved detail retrieval, greater immediacy and potentially less noise compared with a SATA SSD. Of course, SATA SSDs have the potential for greater capacity (which isn't a big deal for an OS drive).

 

I have a similar model (but i5 not i7) and can confirm it works well. I use an SSD disk and was wondering whether, for a server, it really matters whether the OS is on a SSD or msata?

[Qhwoeprktiyns]

Thanks. I started out with a 500 gb SSD with a partition for the OS and roon, and my library on the rest

As I am ripping more CDs I am quickly running out of space. I may get a msata soon anyway, put my library on a large USB drive, and have the SSD for travelling. I think that would make sense. The server also has an SD card slot, however, and I was contemplating using that as am alternative... Not sure it could boot from the SD card however. Too many options!

[Qhwoeprktiyns]

Maybe a stupid question, but would it not make sense to use these clocks on a single "reclocking" card (such as a FIFO buffer) just before the DAC input, and not have to bother with the rest of the chain?

[Qhwoeprktiyns]

I wish that this was true and if it were, then replacing the clock in the sMS-200 should have sufficed but in my system, with each clock upgrade, the improvements continued.

 

In my simplistic way of looking at this, there is no clock that is perfect as even the finest atomic clocks will have some level of phase noise (clock jitter) and instability over time and so the best that a clock can do is to not cause harm to a signal but in reality, all clocks will degrade a signal.

 

Where a really good clock seems to make a difference is in recovering some of the damage made by a bad clock that preceded it. Can it completely undo damage by a bad clock (or a series of bad clocks)? My guess is no but that with repeated reclockings, it would appear that each reclocking can further refine or restore the signal and there are many posts here on CA of how people have reported improvements by placing several USB Regens in series. I used to own a TotalDac d1-monobloc DAC and this DAC includes a very good reclocker and this DAC seemed to benefit as well from having several of these reclockers in series. As such, TotalDac's best "twelve" DAC actually has 2 reclockers.

 

What I am thinking is that it would be best to have as few bad clocks as possible in your chain making it less important to have so many great clocks at the end of the chain to rescue the signal timing. Lastly, I am guessing that if you have an entry level DAC with a mediocre clock, all of these efforts may not make as much of a difference because it should be the very last clock that matters the most. When you look at your digital front end, I believe strongly that the DAC is the most important piece but your DAC can only be as good as the quality of the signal it is fed.

 

Thanks for your input.

I guess someone needs to try chaining several FIFO cards

[Qhwoeprktiyns]

I don't know if anyone has noticed this relatively new "audiophile" switch:

best audiophile Switch, high end switch - AQVOX Audio Devices

 

400 euros for a switch! No thanks... This is getting ridiculous.

[Qhwoeprktiyns]

This should make for an interesting product: Raspberrypi hat with isolation, reclocking, and high quality regulators:

 

http://www.diyaudio.com/forums/digital-line-level/305583-digione-rca-bnc-output-rpi.html

 

Spdif out only. 

[Qhwoeprktiyns]

Agreed. I have their Kali Reclocker and Boss DAC. While I think that statement is over the top, they do deliver very good quality products at reasonable prices. 

 

Since you mentioned that pcie card, I thought I would mention this one, as it is in the same vein, but for a sbc. 

 

[Qhwoeprktiyns]

1 hour ago, R1200CL said:

 

Shuld I understand this product (with the Pi) as an Ethernet to SPDIF converter ?

 

No. The Pi runs a "player" software (squeezelite, MPD, RoonBridge - you name it).

[Qhwoeprktiyns]

51 minutes ago, alligatorman said:

So this is an alternative to a HifiBerry Digi+?

Exactly 

[Qhwoeprktiyns]

2 hours ago, romaz said:

 

This is a fair question and I suspect you won't get a definite answer no matter how many times you ask.  Ask an engineer, IT specialist or a typical poster on ASR and they might offer some explanation that improving latency should offer no SQ benefit but those that offer these responses often do so based on a finite understanding of digital audio and not actual listening.  They may show you measurements that there is no impact on jitter or SNR but are these the only measures of SQ?

 

Go ahead and disconnect the USB, SPDIF, I2S, or Toslink cable between your player and your DAC for 30 seconds while playing back music.  Although extreme, this physical disconnection is a form of latency, is it not?  If your DAC is like mine, you will notice that playback stops fairly quickly, typically within just a few seconds and so your DAC's buffer that you speak so highly of isn't that large and it doesn't take long before the DAC is indeed starved for data.  Nonetheless, look at all the buffers you have upstream of your DAC (RAM, CPU buffer, disk buffer, Ethernet buffer, etc.).  With such buffer redundancy, how could latency possibly be an issue and so your point is well taken. 

 

If you use Roon, unlike your DAC, you probably know that Roon typically buffers a whole track into memory (depending on the size of the track) during playback.  This is easily proven with Tidal streaming or streaming from a NAS.  During playback, disconnect your Ethernet cable and notice that playback will continue for the length of that track in most cases.  This should suggest that the Ethernet cable, server, NAS, and router that are upstream of your Roon playback device should have no impact on SQ and yet they can and do.  

 

Here's another example.  Even though you use an ultraRendu which in of itself is a very good buffer against a higher noise, higher latency server, try loading the latest version of headless AudioLinux + RoonServer onto the server you are currently using and see for yourself if you can hear a difference.  Whether the difference is large or small, positive or negative, the point that you can hear a difference at all should tell you that a buffer is not a complete firewall for all of your upstream problems and that upstream components, whether it be digital or analog, whether it be a resistor, inductor, or capacitor, seem to have the ability to permanently imprint on the signal.

 

As for your question of how latency directly impacts SQ?  I wonder about this myself.  Is it latency itself or is increased latency impacting some other property?  Is higher latency leading to increased jitter, increased noise or dropped bits?  Unless you are upsampling or applying DSP, most software players claim to be bit-perfect but how does the player know if bits weren't dropped beforehand or dropped somewhere downstream?  Neither USB, SPDIF, Toslink nor I2S employ error correcting protocols.  

 

And so we're left with empirical evidence.  With either Windows, MacOS, or Linux, there are plenty of threads here on CA and elsewhere that suggest that as you trim your OS of unnecessary processes and services, latency improves and SQ improves and this is the foundation for products like Audiophile Optimizer regardless of the DAC and that DAC's buffer.  As you move your OS from a higher latency storage medium to a lower latency storage medium (whether it be a hard drive, an SSD, or RAM itself), there are many who will tell you that SQ improves.  AudioLinux has an "Extreme" mode which prevents CPU sleep and according to the designer, results in exceptional CPU latency.  I can verify that when I use "Extreme" mode on my Mac Pro, this otherwise quiet and cool running machine gets noisier and hotter.  While you would think that this would result in increased electrical noise that would be detrimental to SQ, SQ actually improves.

 

As you've suggested that 99% of what we are hearing on this thread is due to "expectation bias," you should listen for yourself.  Even blind test yourself.  I do this often when I'm not sure if what I'm hearing is real or not.  If you can't hear a difference, then live in bliss with what you have but there's no need to call what people are doing on this thread a "joke."  Respect begets respect.

 

 

 

 

You use a USB DAC, so you have a ground line that could carry "crap" into your DAC. The processing that is done upstream could affect the quality of the signal, and it may not have anything to do with latency, you will probably never know, unless you were to compare the same software with different settings. 

 

Some DAC designers are quite knowledgeable about all the interferences that come into the DAC and those generated by the processing within the DAC itself as well. There are some very interesting projects out there that may render all these experiments with sources moot quite soon, and that will be a huge step in computer audio (not the experimenting you are doing here - don't kid yourself...). 

[Qhwoeprktiyns]

You will not have to wait too long : https://www.diyaudio.com/forums/digital-line-level/79452-building-ultimate-nos-dac-using-tda1541a-post5597901.html

[Qhwoeprktiyns]

4 hours ago, romaz said:

 

Ouch.  Enjoy your new DAC.

 

I experiment as well, and I just thought it would be useful to let some people know here about the advances being made in DACs, and which are not necessarily advertised on CA. It is a different perspective, and I am not criticizing yours. 

 

However, you choose to reply explaining that that specific DAC I linked is NOS and NOS is no good because it misses out on the musical content that you somehow recreate by oversampling..

 

I call BS when I see it. Sorry for being so straight forward. 

[Qhwoeprktiyns]

13 hours ago, romaz said:

 

I never said your DAC is no good because it’s NOS and so if you somehow inferred that, than I apologize.  But my claims are valid and I stand behind them.  As I stated, NOS and oversampling DACs aspire to different things and in the ideal world, I believe we each would desire to faithfully recreate the original analog waveform if we could and not just get the best out of the digital recording.  The problem is with the execution and there are successes and failures on both sides.  I can name plenty of NOS DACs that sound better than oversampling DACs and visa versa.

 

I used to own a TotalDac d1-monobloc, an R2R NOS DAC that to this day is the 2nd best sounding DAC I have ever owned.  NOS DACs are known for their musicality and I imagine that your NOS DAC excels in that virtue, otherwise, you wouldn’t be extolling it.  However, both from a conceptual and experiential standpoint, I prefer to oversample and I believe there are ways to oversample today to gain the advantages of resolution while still maintaining musicality, but that’s just me and so this statement shouldn’t be looked at as some gospel truth.  

 

There is no way to know if someone will like a DAC based on its topology or its stat line alone and so maybe I would be floored by your DAC given the chance to evaluate it but resolution is the inherent limitation of NOS DACs and that isn’t BS.  Ultimately, resolution is only one factor and I concede that in the big picture, musicality, as we each define it, is the more important factor.   Peace.

 

I don't know why you shifted the debate to NOS versus oversampling DACs. It has nothing to do with my original post. If you do want to learn more about that specific DAC I mentioned (which l do not have, I have the Mosaic UV) you can read that long thread on Diyaudio. You will find out how NOS DACs can achieve high resolution. Here is a link to some explanations on one of their models: https://www.diyaudio.com/forums/digital-line-level/79452-building-ultimate-nos-dac-using-tda1541a-post4298065.html

 

That thread offers a wealth of information. They have come up with some innovative solutions which will, I believe, significantly improve digital audio playback. 

 

To get back to my initial point... given the sensitivity of DACs to the incoming signal quality (and don't think a simple galvanic isolation is going to solve that), and the complexity of all the processing that is involved in computer based playback, it is highly improbable that you can achieve breakthrough results using the type of solution you recommend (NUC, LPS, AO). 

 

Sorry for my skepticism... 

Peace indeed!

 

 

 

 

[Qhwoeprktiyns]

1 hour ago, greenleo said:

If your terms are just like massive, terrific, ...  Then I can't see how your statement can be verified or falsified. In this case I find that the terms you used in the original text pretty useless. 

 

Also, anybody's progress can be not massive in your point of view.  This is my point.

 

We are having a hard time understanding each other! 

 

Given the state of affairs regarding DACs, and in spite of what most manufacturers claim, inputs pass garbage to the DAC. The source of this is multiple: power supplies, emi, processing within the computer sources, ethernet, cables, you name it... 

 

This is pretty much accepted, and you can her those differences as long as your DAC does not have so much internal interference that you can no longer distinguish between different sources... 

 

Given the sensitivity of the DACs to these factors:

- to think you can solve them only by meddling with the source is an illusion

- I believe the complexity of factors affecting the signal makes it pretty much impossible to design a perfect source based on PCs

- whatever results you may achieve may not be replicable in someone else's configuration. You may even experiences differences based on the time of day! 

 

Now if you start from a bad source, a really noisy PC, you can apply some of the recipes here and I am sure you will be able to achieve some benefits, but at what costs, and only to a limited extent. 

 

I mentioned some developments on the DAC side which I believe will solve these problems. We will shortly see. Until then, I think this endless quest for the perfect source is just not going to yield any significant results. But be my guest and follow the pack... 

 

 

[Qhwoeprktiyns]

I was just offering a different point of view. I will open another thread. 

[Qhwoeprktiyns]

P. S. That dac designer is not unknown, and happens to be well respected (not just in DIY circles). 

Will not post here any longer... Sorry for the "noise"

[Qhwoeprktiyns]

The thread actually already exists, on the DAC section: EC Designs. 

 

I will reply in it with a brief history of their developments, some links to the more interesting pages in that long thread, and more importantly information on their new DAC, which should be available soon on their website. 

[Qhwoeprktiyns]

On 10/24/2019 at 12:17 AM, austinpop said:

One of the reasons I've been posting rather infrequently here is because I was busy with my review of the dCS Bartók, that went live this morning.

 

I mention this here because the Bartók is the first component I've had in my system that was only modestly improved by all the upstream optimization  we've been discovering on this thread. I'd love to see this kind of resilience (and even better) over time, and at lower price points. 

 

Contrary to what many of our naysayers may think, I suspect most of us would be happy to give up upstream tinkering if one day our DACs became completely immune to upstream effects - both good and bad.

 

+1. But it can only be good. 

[Qhwoeprktiyns]

On 7/4/2020 at 7:58 AM, austinpop said:

I know you were being mostly facetious, but it’s a great question. Trust me, I’ve asked myself that many times, and of late I’m convinced it’s “whenever you decide.” There IS no end, but we all can and should stop or pause and enjoy our systems as they are.

 

Sorry to chime in here,  for a brief comment. The answer is obvious: it will stop when noise can be prevented from distorting digital to analog conversion.

 

Until then, have fun! 

[Qhwoeprktiyns]

On 7/26/2021 at 3:47 AM, romaz said:

Low output impedance is just as important as low noise with respect to SQ when designing a PSU and is more difficult to achieve.  If low noise is all that mattered, then why not just use batteries for everything?  Try battery powering a DAVE with any ol' cheap battery that you can buy from Amazon such as a PowerAdd that are designed for recharging cell phones and what you'll get are soft, slow, and smeared transients, a flat sound stage, and weak dynamic contrasts.  As you start to lower the output impedance from the power source, music starts to take form and comes to life.  It just becomes more 3D and palpable and low output impedance is what differentiates a DC4, SR7, Farad, or LPS-1.2 from the rest.  It's amazing how few ultra-low output impedance PSUs there are that are being sold for audio.  Chances are that if the manufacturer fails to mention "output impedance" or better yet, fails to state the output impedance of their PSU in their ad, that PSU is probably not a low output impedance PSU.  Despite the long wait times for PSUs from Sean Jacobs and Paul Hynes, there's a reason people have queued up for these devices.

 

Hi,

 

I am curious to know if there are any "theoretical" explanations for this, in the context of digital to analog conversion (as I understand you are using these low impendance PS to power a DAC).

 

Thanks

[Qhwoeprktiyns]

12 hours ago, hopkins said:

 

Hi,

 

I am curious to know if there are any "theoretical" explanations for this, in the context of digital to analog conversion (as I understand you are using these low impendance PS to power a DAC).

 

Thanks

 

Question can be dismissed. I got confused by the assertion that there were improvements in "transients".