Streamer to kill the big boys for under $250

Blizzard

Banned
Sep 30, 2015
3,049
3
0
Since GG is using Amanero USB, only Window NAA is able to output DSD256?

At the moment, I'm only able to output DSD128 through LINUX based NAA.

Yes with the GG it can only do DoP over Linux. Max on any OS with DoP is DSD 128. DoP requires double the bandwidth of Native DSD. It also doesn't sound as good. It's a lossless process, but adds an extra layer of processing on the sending and receiving end. So this streamer may not be the best for the GG, unless you install Windows 10 on it. Better yet, swap the Amanero for the JLsounds interface, and use this streamer with linux and Native DSD 256. Then you will really hear what the GG can do.It's an easy drop in replacement. It uses way better clocks than the amanero, also has galvanic isolation and reclocker onboard. The Amanero was the best solution in 2012, this is the best solution today. As far as single board stand alone DIY/OEM USB boards anyways.

http://jlsounds.com/i2soverusb.html
 
Last edited:

Blizzard

Banned
Sep 30, 2015
3,049
3
0
Blizz,

I am super interested and definitely want one if it is turnkey solution with flash drive and loaded Linux.

What are you using for a DAC?
 

lateboomer

New Member
Oct 22, 2015
50
0
0
Kuala Lumpur
I have one Bryston BDA-2 dac, one MF M1 DAC, and one M2Tech Young Dac. I going to save up to get one Yggdrasil dac.

I may use my brother's idle HP i7 laptop which I will upgrade it to Win 10 to run HQplayer.
 

Blizzard

Banned
Sep 30, 2015
3,049
3
0
I have one Bryston BDA-2 dac, one MF M1 DAC, and one M2Tech Young Dac. I going to save up to get one Yggdrasil dac.

I may use my brother's idle HP i7 laptop which I will upgrade it to Win 10 to run HQplayer.

Is it the M2tech young DSD? If not, looks like all your DAC's don't support DSD. I suppose you can use it for DSD to PCM conversion then. The Bryston should be plug and play with Linux. I'm assuming it's your best DAC. It uses an older gen Xmos chip. Looks like 1 gen before the U series.
 
Last edited:

Blizzard

Banned
Sep 30, 2015
3,049
3
0
My M2Tech Young does not support dsd. But I am thinking to upsample PCM to play and I will also be looking for one dsd dac.

What you would do then is just set it to output everything at 24/192. Find you favorite filter/modulator, and you're good to go. I would seriously consider going with a chipless DSD DAC. As it will provide you with the best sound from HQplayer. The new Bryston BDP 3 would also be good.

The streamer/DAC/amp I'm working on will be optimized for the best sound possible from HQplayer. 100% IC free signal path, DAC output stage/preamp section 100% discrete running in pure class A. True DSD 1028 capability. No onboard down sampling.
 

lateboomer

New Member
Oct 22, 2015
50
0
0
Kuala Lumpur
Could your working-on dac serve as standalone dac? Will it compromise the sound as pure dac as you have more circuit for the amp side? Beside GG, what are other chipless DSD DACs?
 

Blizzard

Banned
Sep 30, 2015
3,049
3
0
Could your working-on dac serve as standalone dac? Will it compromise the sound as pure dac as you have more circuit for the amp side? Beside GG, what are other chipless DSD DACs?

Well it can if you want a raw PCB on your equipment stand. Everything is being built to be optimized to work together. Just think Devialet, but 100% discrete circuitry, DSD 1028 capability, Ravenna Ethernet interface, ultra low jitter clocks (each clock individually tested and matched) and 600w per channel Hypex Ncore amplification. There will be other unique attributes as well. But not going to disclose them.

There's a few chipless DAC's around. You'll have to look for commercial offerings. Not sure what's out there.
 

wisnon

Well-Known Member
Dec 12, 2011
3,536
640
1,200
Since GG is using Amanero USB, only Window NAA is able to output DSD256?

At the moment, I'm only able to output DSD128 through LINUX based NAA.

Linux ALSA for Amanero coming very soon. 256 enabled.
 

wisnon

Well-Known Member
Dec 12, 2011
3,536
640
1,200
Could your working-on dac serve as standalone dac? Will it compromise the sound as pure dac as you have more circuit for the amp side? Beside GG, what are other chipless DSD DACs?

Even the much cheaper Lampizator Level4 is chipless DSD.
 
Last edited:

wisnon

Well-Known Member
Dec 12, 2011
3,536
640
1,200
Well it can if you want a raw PCB on your equipment stand. Everything is being built to be optimized to work together. Just think Devialet, but 100% discrete circuitry, DSD 1028 capability, Ravenna Ethernet interface, ultra low jitter clocks (each clock individually tested and matched) and 600w per channel Hypex Ncore amplification. There will be other unique attributes as well. But not going to disclose them.

There's a few chipless DAC's around. You'll have to look for commercial offerings. Not sure what's out there.

So, Blizz, why not build all that electronics into the speakers? Active/DSP route...or is that your plan already?
 

wisnon

Well-Known Member
Dec 12, 2011
3,536
640
1,200
Hi Guys,

http://www.amazon.com/Intel-Thin-Ca..._UL160_SR135,160_&refRID=0T8Q57HXD0SBW0WF9V0Q


http://www.amazon.com/Crucial-PC3-1..._UL160_SR160,160_&refRID=19M2XHTV5WPMWQK34TM4


Those 2 items is all the hardware required. for even better performance, a LPS can be used to power the streamer. For the OS, if interest is shown, I can talk with the creator of the real time Linux system I'm using to see if we can offer turn key USB sticks for $90 each.

That's all there is to it. If anyone is interested in this, let me know. It really is the real deal!!

What are the LPS power requirements for this and the dimensions of the power leaad going into the NUC?
 

Blizzard

Banned
Sep 30, 2015
3,049
3
0
So, Blizz, why not build all that electronics into the speakers? Active/DSP route...or is that your plan already?

That's a different system.
 

Blizzard

Banned
Sep 30, 2015
3,049
3
0
What are the LPS power requirements for this and the dimensions of the power leaad going into the NUC?

12-24V, 25w is plenty. Using this system the quality of the server computer is irrelevant, as long as it has enough processing power. Could be a Lampi Komputer or $600 off the shelf HP, both will sound identical.
 

amirm

Banned
Apr 2, 2010
15,813
38
0
Seattle, WA
This is the Linux version I'm using. It's configured specifically for audio. Whatever the NAA Daemon does, it improves the sound. And when running on Audiolinux, it sounds better than Windows.

https://translate.google.ca/translate?hl=en&sl=it&u=http://www.tophifi.it/&prev=search
Thanks for the link. But there is nothing there to improve audio fidelity. Or at least nothing on purpose :).

This is a much longer topic. I will give an introduction here but perhaps we should have a new thread on this.

The way an audio or most devices work in an operating system like Linux, Windows and Mac works is that you give them something to do, they go and do that and meanwhile the operating system runs other jobs. When they are done, the devices ring a bell like a front desk bell asking for attention of the operating system. That way, the CPU is not wasting time just waiting for them to do something. Playing for example a chunk of audio samples in a DAC could take milliseconds. The OS could execute millions of instructions during that time, running for example your browser and email instead of waiting to feed the DAC the next chunk of data.

Another characteristic of audio playback is that it is "real time." If you are playing CD, you need to feed the DAC 44.1 samples of two bytes each for each channel. Do less of this and the DAC will run out of data to play and will pause. That pause if short sounds like a glitch and if long, literally silence. If you take a very old PC with little memory and hook it up to your DAC and play something and then start say a game, you can easily see this happening. A dedicated CD player doesn't have this issue because it is only a CD player. An operating system is "multi-tasking" and can in theory run behind.

Going back to your Linux distro, it addresses the interrupt latency (first issue) by boasting how much better it is than Windows. It provides these measurements, first for their Linux "running in real-time mode":



We see lightning fast response in that the latency, or the time taken for the CPU to respond to a device is down to just a few microseconds. In contrast they show this fro Windows (different program but same idea):



In Windows we see latencies or time to respond to a device of nearly 1000 microseconds or about 500 times worse than the Linux example above.

Seems like the case is made or is it? Well, it isn't :). Let's go back to the hotel example. You ring the bell and the person shows up in 10 seconds. You would care if it is 10 seconds or 100 seconds. But what if he came back in 10.01 seconds instead of 10? You would not notice or care, right? Same is here.

Audio devices are very slow. And they employ buffering which means they hold on to data they are playing. So we only need to give them attention once in a while compared to the lightning speed of a CPU. Let's make an extreme case of giving one second worth of data to the DAC at a time. In that case, being able to respond to the DAC in 1000 *micro* seconds that Windows does is lightning fast. Note how the DPC checker in the second graph for Windows says clearly that the system is functional enough for audio!

We have proof of this working as countless people use their standard Mac/Windows machine for audio playback and we don't have reports of glitches and pauses. Yes, it can happen. Lee had a DAC that was doing this and turned out the problem was the visualization (random graphics playing with music) that his media player was generating. Turned that off and the problem was gone.

Importantly all of this impacts *functionality* not fidelity. If these problems exist, you will know it because of the pops and glitches. It will not manifest itself in audiophile qualities. Either you can keep up with the conveyer belt of audio samples going to the DAC or not. If you don't the parts, or audio samples, fall on the floor. The quality of the parts won't be impacted one way or the other. In that sense, nothing in the feature list of that Linux distribution that aids in fidelity department.

Now there could be a fidelity difference due to system activity but that has huge dependency on hardware and at any rate, nothing is documented there to demonstrate any improvements are there.

You might ask but what is the down side of having better real-time performance anyway? Well one main one: you get locked into a strange operating system that may not be supported in the future or with the devices you use. And a secondary one is security flaws. There are vulnerabilities discovered in Linux as well as Windows and Mac. The latter two come with frequent updates to plug them. But your little embedded computer sitting in your home network with full view of all other computers is not patched or updated by these companies. Is the risk high? No. It is actually very small but should it happen, the consequences can be high.

Now, I don't want to pour cold water over this :). If you are a person like Mike who likes to do this kind of thing, then is is a cool option. But if you are not the type, I don't think it is a wise path. I think you will be better off with a Mac Mini or a silent Windows machine.
 

Blizzard

Banned
Sep 30, 2015
3,049
3
0
Thanks for the link. But there is nothing there to improve audio fidelity. Or at least nothing on purpose :).

This is a much longer topic. I will give an introduction here but perhaps we should have a new thread on this.

The way an audio or most devices work in an operating system like Linux, Windows and Mac works is that you give them something to do, they go and do that and meanwhile the operating system runs other jobs. When they are done, the devices ring a bell like a front desk bell asking for attention of the operating system. That way, the CPU is not wasting time just waiting for them to do something. Playing for example a chunk of audio samples in a DAC could take milliseconds. The OS could execute millions of instructions during that time, running for example your browser and email instead of waiting to feed the DAC the next chunk of data.

Another characteristic of audio playback is that it is "real time." If you are playing CD, you need to feed the DAC 44.1 samples of two bytes each for each channel. Do less of this and the DAC will run out of data to play and will pause. That pause if short sounds like a glitch and if long, literally silence. If you take a very old PC with little memory and hook it up to your DAC and play something and then start say a game, you can easily see this happening. A dedicated CD player doesn't have this issue because it is only a CD player. An operating system is "multi-tasking" and can in theory run behind.


Going back to your Linux distro, it addresses the interrupt latency (first issue) by boasting how much better it is than Windows. It provides these measurements, first for their Linux "running in real-time mode":



We see lightning fast response in that the latency, or the time taken for the CPU to respond to a device is down to just a few microseconds. In contrast they show this fro Windows (different program but same idea):



In Windows we see latencies or time to respond to a device of nearly 1000 microseconds or about 500 times worse than the Linux example above.

Seems like the case is made or is it? Well, it isn't :). Let's go back to the hotel example. You ring the bell and the person shows up in 10 seconds. You would care if it is 10 seconds or 100 seconds. But what if he came back in 10.01 seconds instead of 10? You would not notice or care, right? Same is here.

Audio devices are very slow. And they employ buffering which means they hold on to data they are playing. So we only need to give them attention once in a while compared to the lightning speed of a CPU. Let's make an extreme case of giving one second worth of data to the DAC at a time. In that case, being able to respond to the DAC in 1000 *micro* seconds that Windows does is lightning fast. Note how the DPC checker in the second graph for Windows says clearly that the system is functional enough for audio!

We have proof of this working as countless people use their standard Mac/Windows machine for audio playback and we don't have reports of glitches and pauses. Yes, it can happen. Lee had a DAC that was doing this and turned out the problem was the visualization (random graphics playing with music) that his media player was generating. Turned that off and the problem was gone.

Importantly all of this impacts *functionality* not fidelity. If these problems exist, you will know it because of the pops and glitches. It will not manifest itself in audiophile qualities. Either you can keep up with the conveyer belt of audio samples going to the DAC or not. If you don't the parts, or audio samples, fall on the floor. The quality of the parts won't be impacted one way or the other. In that sense, nothing in the feature list of that Linux distribution that aids in fidelity department.

Now there could be a fidelity difference due to system activity but that has huge dependency on hardware and at any rate, nothing is documented there to demonstrate any improvements are there.

You might ask but what is the down side of having better real-time performance anyway? Well one main one: you get locked into a strange operating system that may not be supported in the future or with the devices you use. And a secondary one is security flaws. There are vulnerabilities discovered in Linux as well as Windows and Mac. The latter two come with frequent updates to plug them. But your little embedded computer sitting in your home network with full view of all other computers is not patched or updated by these companies. Is the risk high? No. It is actually very small but should it happen, the consequences can be high.

Now, I don't want to pour cold water over this :). If you are a person like Mike who likes to do this kind of thing, then is is a cool option. But if you are not the type, I don't think it is a wise path. I think you will be better off with a Mac Mini or a silent Windows machine.


Regardless of the OS used, this little NUC sounds better than any other computer I've tried. I have 6 computers here. It's way better than my Mac mini. The better sound quality I'm experiencing using the Audiolinux, may be because of the Native DSD capabilities with Linux. Because even when I configure in DoP with Audiolinux, there's a slight degregation in quality. To take advantage of the NAA system, you need a computer running some kind of OS for the NAA. Windows is far too bulky and expensive to run for a stand alone NAA. Linux is the way to go. It's also cheaper. So this is the best solution. As far as hardware, if you know of better hardware for the task for under $155, please let me know.

There's a reason every stand alone streamer uses Linux. This is a very similar implementation that Auralic uses for the Aries, and Aurender as well. So if your paranoid about security, you should be with them as well. The way I have it configured, I don't think a virus could even run on it. It's just command line as basic as it can get for running the NAA daemon.

The reason I like to do this kind of thing is for sound. I've compared all my computers direct using HQplayer, running the NAA daemon on Windows and OSX on several pieces of hardware. Nothing sounds as good. It also allows the server to be out of the audio rack.

Also if you know of a Windows version that costs $90 or less, comes on a USB stick, that you can pop in the USB port, push the power button on the computer, and 10 seconds later have beautiful sounding audio come out of your DAC, please share.
 
Last edited:

amirm

Banned
Apr 2, 2010
15,813
38
0
Seattle, WA
No that is a huge bargain at $155. I have no alternative to offer. My case was twice that price and that was for starters :).
 

Blizzard

Banned
Sep 30, 2015
3,049
3
0
I have $250 of disposable cash to try this. Please point me in the right direction. Thanks.

Edit, New Linux version is free


I'm going to get a hold of him and I'll let you guys know what he says. I have a few sticks, but if it looks like there's lots of interest, I could order more. Maybe as another option I can just offer the image download for $15 less. Then it can easily be burnt to stick using Rufus on Windows.

https://rufus.akeo.ie

It would save on shipping cost. The stick I have tested and confirmed to preform well is this one:


http://www.amazon.com/SanDisk-Ultra-Transfer-Speeds-s-SDCZ48-016G-UAM46/dp/B00KZIXSP6
 
Last edited:

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing