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

[seeteeyou]

34 minutes ago, romaz said:

Yes, this uses a better board than my own as it uses a Celeron with a larger secondary cache which should result in less latency.  It also utilizes DDR4 RAM instead of DDR3.  At this point, I do not know which type of RAM is lower latency.

 

Not sure about the latency part but I compared DDR3 from different companies back in January

 

https://www.computeraudiophile.com/forums/topic/30376-a-novel-way-to-massively-improve-the-sq-of-computer-audio-streaming/?page=225&tab=comments#comment-762083

 

Just like the relatively low current draw of Intel SLC drives, we could stick with the ones with lowest values but it's just a matter of finding single 8GB sticks that don't draw THAT much current to begin with.

 

Unfortunately BIOS settings from NUC boards might not give us anything to run our RAM at a lower speed. If that were the case, we'll have to figure out how to control the speed with specific software instead.

[RickyV]

10 hours ago, jean-michel6 said:

Ś

 

Mumetal is a blend of nickel and iron  and I really do not understand what is mucopper described by the holland shielding company. 

 

I think the copper shield advertised are effective for RFI interférences for EMI that is another story as you have to use a mumetal foil which is usually quite expensive. 

Yes I found that one too but I see no specs so how effective is it in comparison with the mu copper.

[seeteeyou]

17 minutes ago, romaz said:

I think my point with separating server and renderer has more to do with distributing tasks (such as library management, DSP, upsampling, rendering, etc) so that a single CPU doesn't have to do all the work.

 

Have you looked into NUMA by any chance?

 

https://www.computeraudiophile.com/forums/topic/38729-a-crowd-funded-motherboards-for-audiophiles-part-1/?tab=comments#comment-791428

 

Running dual Xeon at a much lower speed could be fun to experiment with.

[romaz]

On 8/8/2018 at 9:36 PM, ElviaCaprice said:

Roy actually sold the Innous SE and prefers the NUC so I'm confused what his current setup is in both rooms at this time.  The Apple mini for the main PC I guess in place of the Innous SE.  Regardless does it matter, Roy is going to keep trying and changing things around.

 

Nothing has been finalized.  I would love to evaluate the new Innuos Statement and even the Antipodes CX + EX and see how far off I am (if at all) but obviously, it would be very hard for either of these units to compete with what I have from a value standpoint.  

 

Whether I upgrade my NUC with an sCLK-EX board or not, I haven't decided because what I have now is sounding so good although I suspect at some point, I will want to do it so that I can take further advantage of the REF10.  

 

The beauty of this simple inexpensive setup is that I could run RoonServer off a centralized QNAP NAS with SSD cache.  As their NAS units have multiple LAN ports that are bridgeable, this NAS could then feed multiple NUCs.  Of course, each NUC will need to be preceded by its own switch but this whole thing shouldn't cost that much.

[seeteeyou]

2 minutes ago, romaz said:

Whether I upgrade my NUC with an sCLK-EX board or not, I haven't decided because what I have now is sounding so good although I suspect at some point, I will want to do it so that I can take further advantage of the REF10.

 

Since your modified Oppo UDP-205 (with "lowly" OCXO that's placed inside the same chassis) should sound at least as good when compared to your ZENith SE + tX-USBultra + Habst + REF10 + Paul Hynes SR7 combo, what if we're applying the same "direct-path" tricks with an excellent OCXO like these guys instead?

 

http://newclassd.com/index.php?page=200&hv=1

 

Given a wide range of frequencies could be chosen, maybe we could find something that's gonna be able to match what NUC would actually need.

[romaz]

21 hours ago, HeeBroG said:

 

Thanks Roy,

 

Can I ask how all this complexity compares with Chord BluMkII playing a CD or have you managed to surpass this long ago?

 

What I'm talking about isn't necessarily complex (server > switch > renderer) but certainly, it can't compare to the simplicity of spinning a CD with a Chord BluMk2.  With regards to SQ, the BluMk2 CD transport has served as a very good benchmark to compare against and it has not been easy to surpass CD playback since when you spin a CD, you don't have to worry about the impact of computer noise, power supplies, cables, etc.  However, depending on the quality of the master, the quality of the rip, and the bitrate of the file, I have in many cases matched and even surpassed CD playback SQ.  When it comes to the convenience of being able create playlists off millions of tracks from within my library and from Tidal, there's simply no comparison.

 

This highlights another significant impact of the switch.  For some time now, as I have been using a reclocked switch for the past year in my chain, I have not noticed any difference between local playback and playback from a NAS or off Tidal.  Provided we are talking about identical masters, even when I still had my Zenith SE, I could detect no meaningful difference when playing back from the SE's internal SSD or from my NAS.  With the anticipated arrival of Qobuz here in the U.S. (and hopefully, it gets incorporated into Roon), it's possible I will have no incentive to buy music in the future, certainly not CDs since Qobuz will be streaming hi-res files.

[romaz]

16 hours ago, Johnseye said:

I’ve been thinking through the outcomes of this post for the past two days.  Here’s my takeaways and resulting questions.

 

1.       The OS drive has played a significant impact in his tests.  Mac mini with Mac OS on SD card, Mac mini with Windows OS on NVMe SSD, drive of the Zenith SE, TLS Linux OS on RAM, NUC with TLS OS on eMMC.  Of all these options the NUC with TLS Linux OS on the eMMC drive that sounded the best to Roy.  My results with the Optane also blend well with these findings.

2.       The Celeron processor on that NUC could not upsample DSD, but if there is a similar board with i7 proc this could be possible.  I suspect an i5 or i7 will still require a fan.

3.       I suspect this is early in Roy’s testing, but this board could be modified by SOtM to utilize their sCLK-EX.  This would also allow the use of a REF10 for the board.

4.       The bridge connection with endpoint does bring us back full circle.  Many of us have or still do this.  Question is what the NUC brings to the table over another endpoint.

5.       Roy found little difference in the upstream source, comparing his Mac mini with the Zenith SE, especially with the modified clock switch in between it and the endpoint.  I my own listening experiments with Rajiv we both could hear a difference when the server was a noisy PC vs a customized sCLK, low power PC.  The difference was very subtle.  Roy’s comparison with a $7k Zenith PC reduces that device’s value.

6.       The ethernet and modified clock switch provides isolation, or filtering from upstream noise.  As Roy mentions, this has been known for some time.

7.       Switch versions with better clocks and power sources are undergoing rapid changes with the boutique audio companies.

@romaz why use the NUC as an endpoint when the sMS-200ultra or the ultraRendu provide the same functionality?  Was it because there wasn’t one of those devices to compare against the NUC?  Are the hard drives on those devices noisier than the eMMC on the NUC?  The sMS and uR can be powered by an LPS-1.2 or SR7/4 with less power.

I’d like to know how the TLS Linux OS compares to Roon ROCK when Roon is upsampling.  Likewise with HQPlayer.  Finally, both streaming to an endpoint.

 

John, the significance of the eMMC drive is to avoid having an SSD anywhere in the renderer where it's negative impact will be heard, however, the significance of being able to fully run the OS from RAM is the more important point, I believe, since this results in massively less latency compared against an SD card or Intel X25-E.  This scheme essentially provides you lower latency than even an Optane with noise comparable to an SD card.

There are Intel NUC boards with i7s that you can use if you feel you want a small single chassis server and would like to upsample, however, none of the units I saw had eMMC drives.  Generally speaking, I have only seen eMMC drives with Celeron NUCs.  Since these NUCs are quite small and inexpensive, it would probably be more ideal to buy an i7 NUC as a server and something like the NUC I am using which has an eMMC drive as a renderer and if desired (if space is a premium), stack one on top of the other.  The key will be the switch and of course what you use to power the switch and the renderer.

 

As for the need to spend lots of money on a server if you are using a good switch between server and renderer, I think this is where a switch becomes extremely valuable.  I certainly couldn't justify using something like a Zenith SE as a server in this situation, however, as a standalone server/renderer, the Zenith SE is still an excellent unit and deserving of all the praise that has been heaped upon it.  For those who don't want to deal with the complexities of multiple devices, the Zenith SE is tough to beat.  Furthermore, Innuos is a solid company who offers first rate support.  As I prepared to sell my SE, I had some operational issues with my unit and Nuno was very quick and gracious to help me out even though he knew full well that I was selling my unit.  Their customer support is some of the best I have experienced.

 

I, too, would like to know how my cheap NUC compares against something like a Nucleus that uses ROCK.  Hopefully, I will be able to make this comparison soon.

 

Since HQP makes an app for renderers that will communicate with HQP installed on a separate server, there's no reason to believe what I am doing can't apply to an HQP setup but I have not done this testing to know for sure.

 

Regarding the sMS-200ultra and ultraRendu, look at my response to Larry a few posts back.  If I thought I could get them to sound as good as a Zenith SE, I would never have bought the Zenith SE.  As stated, my NUC setup now surpasses my SE.  I'm not quite sure what you mean by hard drives on the sMS-200ultra or ultraRendu since these units don't have hard drives.

[romaz]

9 hours ago, Bricki said:

Also, @romaz, did you try the new sotm switch outside of the direct connection and did the direct connection still impact the sound with the new sotm switch?

 

Yes, the impact when used outside of this direct connection is still significant and quite worthwhile, however, it's impact within the direct connection is noticeably greater.

 

There are other important applications of a good switch.  Having tried the SOtM switch connected to my Oppo BluRay player using a standard connection to my router, I was quite surprised by how much more briskly apps like Vudu and Netflix would load their video offerings and how much cleaner and sharper the video was.  In some ways, the improvement with video is even more impressive.

[seeteeyou]

https://fit-iot.com/web/products/fitlet2/

 

fitlet2 J3455 Barebone     × 1 = $161
https://fit-iot.com/web/product/fitlet2-j3455-barebone/

 

16 GB eMMC module for fitlet2     × 1 = $14
https://fit-iot.com/web/product/16-gb-emmc-module-for-fitlet2/

 

Not bad for $175 plus shipping?

[seeteeyou]

On 8/8/2018 at 7:29 PM, romaz said:

I have no scientific explanation for so many differences that I hear but as best as I have been able to figure out, good digital amounts to 3 things:  (1) low noise, (2) low impedance, and (3) low latency.

 

Ain't that kinda like those Blue Zones: (1) low salt, (2) low sugar, and (3) low fat?

 

Seriously, does low latency affect our music servers as much as High Frequency Trading and High Performance Computing?

 

Ultra low latency Ethernet (UCS “usNIC”): questions and answers

https://blogs.cisco.com/performance/ultra-low-latency-ethernet-questions-and-answers

 

Ultra-Low latency for Everybody

http://www.mellanox.com/blog/2017/05/3877/

[lmitche]

7 hours ago, Johnseye said:

 

The ethernet cable between the switch and PC is long.  Probably about 20 ft.  Between the NAS and switch is about 1 ft.

John, sorry not to be clear, it is the physical distance between the server and nuc that matters. 20 feet may not do it.

[BigGuy]

14 hours ago, auricgoldfinger said:

 

The AA Cable Shield is made by the company that manufactures MuMetal.  The shield is made from Co-NETIC, which I believe uses the same alloys as MuMetal, but you may want to double check before buying.  The WBS-250 is $11.35/ft. and the WBS-375 is $12.60/ft.

Magnetic Shield Corp holds the copyright for "MuMetal", "Co-NETIC", and other tradenames.  Co-NETIC = MuMetal in composition.

In addition to some braided shield I ordered some 0.002" foil which I thought might be more flexible and certainly less bulky than the braid.

[seeteeyou]

Could this be even better than eMMC or what? Simply boot the VHD image off USB disk and then unplug that afterwards

 

Diskless Windows 10 PC Setup Procedure
https://www.computeraudiophile.com/forums/topic/26705-diskless-windows-10-pc-setup-procedure/

 

randytsuch's audio page: Running Windows from RAM
http://randytsuch-audio.blogspot.com/2016/02/running-windows-from-ram.html

 

Install and Boot all versions of Windows from VHD using Firadisk
http://agnipulse.com/2017/01/install-boot-all-versions-windows-vhd-using-firadisk/

 

Windows on a RAM disk / Geek magazine
http://geek-mag.com/posts/185172/

 

138 - Directly boot from a Windows 7/8/10 VHD file with grub4dos
https://www.rmprepusb.com/tutorials/138ntboot

 

Someone actually put Windows PE on his NAA and obviously that USB drive could be unplugged once the OS was good to go

 

https://post76.hk/thread-263931-1-1.html

[soares]

4 hours ago, romaz said:

 

If you're referring to SOtM's new switch, they are awaiting parts and so no official release date or price that I am aware of.  If you're referring to Uptone Audio's switch, I would like to know the answer to that one myself.

I was referring to the Uptone Audio's switch... 

[seeteeyou]

Does it actually make sense or not?

 

Ethernet or WiFi – which is better for high-end audio streaming?
https://darko.audio/2018/08/ethernet-or-wifi-which-is-better-for-high-end-audio-streaming/

Quote

 

“The network router, the NAS and other network device are directly attached to the router using Ethernet cable which can be quite noisy. If the streamer is attached to the router using Ethernet cable then the noise can easily be transmitted to the streamer over this physical link. The regular CE, FCC and other EMC standards only guarantee the device will work stably, but when we are talking about sound quality, we are actually talking about EMI noise that is 1/1000 of the EMC’s standard, so it affects [sound quality] a lot. That’s why audiophiles say different NAS or different routers sound different.

 

The Ethernet cable is [also] a big problem. It runs long and can pick up low-frequency noise which cannot be filtered out. This noise will introduce jitter at low frequencies which no PLL circuit can get rid of and it will affect sound quality a lot.

 

WiFi will sound better if:

 

The engineer that designed the streamer knows how to design the WiFi module right. ‘Right’ means the module does not introduce noise to other audio circuits sharing the same ground circuit;

 

The WiFi signal between streamer and router is strong and stable. This is because if you have a poor WiFi signal, the WiFi protocol will try to fetch the data again and again due to large error rate (Yes, we use TCP to transfer music data which has error correction). If there are too many retries, it will add additional load to the processor and potentially generate more EMI noise.

 

The reason that many audiophile companies claim Ethernet to sound better is that they are not capable of designing a good WiFi circuit and they have no such background knowledge (poor WiFi network). All our streamers use existing Intel WiFi modules because they work well on a Linux system and they are well designed with minimum EMI noise generation and, most importantly, we know how to design the circuit around it correctly.”

 

 

[lmitche]

6 hours ago, flkin said:

 

In my MuMetal conduit (solid conduit, not braid) experiments on USB, I am finding that it affects the sound enormously adding unbelievable details and tightening bass. Unfortunately it compresses the soundstage and makes the brighter sound further away and less organic. After listening for longer periods of time, I find that without MuMetal conduits is better. Now working on shorter conduits that only cover part of the USB cable.

 

I have the conduits over my 12V PoE DC cables supplying my router but they don't seem to make that significant a difference but I still have to confirm this again when I find some time.

Flkin,

 

First off, thanks for sharing your experience with the conduit. Your positive experience helped me pull the trigger on this experiment.

 

With the 3/4 inch braid over the jssg 360 lush cable there is no impact on sound stage size here so I am surprised to learn you have an issue. The lush cable carries DC to my USB powered DAC and that power comes from a lt3045 chain.

 

Is your DAC powered via USB? If so, how? Could you could have an impedance or power  issue that is related with the increased dynamics? I have to admit that I anticipated an issue with the mumetal such as your sound stage compression and am pleased to have dodged that problem. Maybe the braid is a better solution?

 

My application of the conetic braid here is as simple as possible. I'm am just placing the cable in the braid avoiding an electrical connection at either end.

 

After listening to the braid on the lush cable the addition of the braid to the Gotham power cable adds another level of refinement, more detail and density in the treble and bass. For example I can hear whole cymbals in many more tracks now.

 

Two more Gotham cables will be treated with the braid today. One that carries power from an lps-1 to my USB card, and the second powering my router/switch. I expect a big impact from the first, and little from the second, but who knows?

 

Stay tuned!

 

 

 

 

[lmitche]

6 hours ago, romaz said:

For an upsampler such as yourself

Hi Roy,

 

LOL! Thanks for your reply and yes indeed I am an upsampler in a world full of (what's the word?) non-samplers (no), un-samplers (not right), regenerators (already taken), or samplers (makes sense I guess but boring).

 

Your point on using a direct (bridged) connection makes sense, and may not be the only way to get this benefit. My 5 volt $18 trendnet wired router allows me to create a music subnet and effectively filters broadcast traffic from the rest of the house from the music server. That is likely the largest benefit of the bridge setup you describe as broadcast traffic is filtered by the bridge software. And this may explain why the audio switch alone doesn't do the same. Having said this, there may be some additional electrical noise filtering with higher quality dual nics as others such as Alex C have conjectured.

 

Thanks to Rickca and yourself for the pointer to the 7 series nuc. I may try that instead of the 6 series. I did notice it is a dual core machine, not a quad core like yours so I am a bit torn.

 

Also great to know I am on the right track with audiolinux. Piero has done a great job with audiolinux and his support is first class. Archlinux is extremely well documented. For these reasons it is my go-to linux solution so I am happy to see it's your choice.

[mourip]

26 minutes ago, seeteeyou said:

Does it actually make sense or not?

 

Ethernet or WiFi – which is better for high-end audio streaming?
https://darko.audio/2018/08/ethernet-or-wifi-which-is-better-for-high-end-audio-streaming/

 

 

My take-away from his article was just that this choice is not black or white. It is about implementation as always.

 

I have always avoided Wi-Fi close to my audio equipment as much as possible, including using ethernet for my nearby TV and streaming boxes. I use ethernet to connect my audio PC to my Dante based Focusrite Rednet D16 ethernet to AES converter and have ethernet coming into my audio PC for remote control of JRMC and for copying music files. I have experimented with fiber media converters for connections between my audio PC and other devices but found that they did not help much, preferring working on using LPS's for power and the grounding scheme that John Swenson has described here.

 

For me the answers have been found in wired ethernet, LPS power, grounding, isolation transformers, and simple signal paths.

[RickyV]

I had a thought yesterday with analog signals we are trying to avoid ferro/ magnetic metals in signal connectors isn't it. Also near analog cables?

In mu metal alloy  17% or something  is ferro metal. What does it mean for digital signal?

Just a thought.

Somebody else any thoughts on this. 

[BigGuy]

More info on the MuMetal and Cu-containing shields...

 

Extract from CA <

Also from a post on that forum...

 

Look into TI shielding. Copied from the Micheal Percy catalog:

 

TI SHIELD: Unique composite shielding material from Texas Instruments for blocking EMI & RFI interference. Conventional shielding materials excel either in electrical conductivity or magnetic permeability, but not both. Steel is not conductive enough to be effective at blocking high frequencies, while copper and aluminum are not effective at absorbing low frequency EMI. No single material met both of these requirements until TI developed this copper/alloy 49/copper composite material. TI Shield is effective from below 10Hz up to 10GHz and is currently available only in .014" thickness (28-29ga). Counterpoint utilized sim- ilar technology in construction of their chassis' and found significant audible benefits were to be derived from surrounding audio circuitry with materials that shield against both EMI & RFI. Unlike Mu-Metal, this material may be cut and bent into conven- ient shapes without requiring annealing to restore its shielding properties.

12" x 12"/$32.50, 12" x 24"/$64.50, 12" x 48"/$127.00

[auricgoldfinger]

6 hours ago, jean-michel6 said:

Thank . The co-Netic is apparently  a proprietary alloy where mumetal is also an other proprietary alloy but more common source .

 

Where have you tried this Co-netic shield ? Which construction do you use  ( connected or not ? ,JSSG ? ....) ? Which SQ benefit did you get ?

 

Here in France there is a company HiFicable which is applying a double layer of mumetal very thin plasticized film to all his top of the line cable . The SQ benefit is usually a darker background .

 

See comments by @lmitche several posts above on this page regarding SQ benefits.  I have been waiting for someone with the experience of the tinned copper braids to compare to the Co-NETIC shield, and no one is better qualified to opine than the originator of the JSSG360 technique himself!

[lmitche]

46 minutes ago, auricgoldfinger said:

 

See comments by @lmitche several posts above on this page regarding SQ benefits.  I have been waiting for someone with the experience of the tinned copper braids to compare to the Co-NETIC shield, and no one is better qualified to opine than the originator of the JSSG360 technique himself!

Thanks Aurogoldfinger, actually we should thank @Bigguy who first made us aware of mumetal braid. Also I have not compared the mumetal braid to tinned copper, as I have laid layer one over two others. I may do this experiment later on if I can find a way ro accomplish it without having to deconstruct something else.

[Johnseye]

5 hours ago, romaz said:

 

John, the significance of the eMMC drive is to avoid having an SSD anywhere in the renderer where it's negative impact will be heard, however, the significance of being able to fully run the OS from RAM is the more important point, I believe, since this results in massively less latency compared against an SD card or Intel X25-E.  This scheme essentially provides you lower latency than even an Optane with noise comparable to an SD card.

There are Intel NUC boards with i7s that you can use if you feel you want a small single chassis server and would like to upsample, however, none of the units I saw had eMMC drives.  Generally speaking, I have only seen eMMC drives with Celeron NUCs.  Since these NUCs are quite small and inexpensive, it would probably be more ideal to buy an i7 NUC as a server and something like the NUC I am using which has an eMMC drive as a renderer and if desired (if space is a premium), stack one on top of the other.  The key will be the switch and of course what you use to power the switch and the renderer.

 

I had been following your path on the single sever approach with system, lan, tx-USBexp and tx-USBultra clocked off an sCLK.  REF10 master.  I'm using a TLS switch isolating upstream, including NAS from the server and usb components.  This NUC as an endpoint has some potential for improvement.  It still requires a USB out to the DAC.  I wonder to what extent if at all the tX-USBexp could be leveraged if it were able to be somehow connected to the NUC.  Perhaps it's not even necessary.  While you do say that your system sounds good already, I can't help but think what a REF10 supported sCLK replacing the system, lan and USB clocks in that NUC could do.

 

5 hours ago, romaz said:

 

I, too, would like to know how my cheap NUC compares against something like a Nucleus that uses ROCK.  Hopefully, I will be able to make this comparison soon.

 

Since HQP makes an app for renderers that will communicate with HQP installed on a separate server, there's no reason to believe what I am doing can't apply to an HQP setup but I have not done this testing to know for sure.

 

More fun to come.

 

5 hours ago, romaz said:

Regarding the sMS-200ultra and ultraRendu, look at my response to Larry a few posts back.  If I thought I could get them to sound as good as a Zenith SE, I would never have bought the Zenith SE.  As stated, my NUC setup now surpasses my SE.  I'm not quite sure what you mean by hard drives on the sMS-200ultra or ultraRendu since these units don't have hard drives.

 

It sounds like no direct comparison, but a chain of events leading to the conclusion that one would be better than the other.  Understandable and I'm sure the outcome would be the same.  This does change the endpoint game and it sounds like we don't know exactly why yet.  Perhaps reduced latency through more capable hardware.

 

Alex corrected me earlier.  I was loose with the hard drive term and should have been more specific as to whatever containers are holding the OS and running the OS.

 

Thanks for sharing your continued exploration down the path of improved SQ.  It certainly inspires everyone in the camp.

[lmitche]

@romaz, one last question please. I am curious, have you tried booting from the SD card slot in the nucs and how does it sounds compared to the emmc storage? I believe slc SD cards are available. Thanks!

[BigGuy]

1 hour ago, BigGuy said:

More info on the MuMetal and Cu-containing shields...

 

Extract from CA <

Also from a post on that forum...

 

Look into TI shielding. Copied from the Micheal Percy catalog:

 

TI SHIELD: Unique composite shielding material from Texas Instruments for blocking EMI & RFI interference. Conventional shielding materials excel either in electrical conductivity or magnetic permeability, but not both. Steel is not conductive enough to be effective at blocking high frequencies, while copper and aluminum are not effective at absorbing low frequency EMI. No single material met both of these requirements until TI developed this copper/alloy 49/copper composite material. TI Shield is effective from below 10Hz up to 10GHz and is currently available only in .014" thickness (28-29ga). Counterpoint utilized sim- ilar technology in construction of their chassis' and found significant audible benefits were to be derived from surrounding audio circuitry with materials that shield against both EMI & RFI. Unlike Mu-Metal, this material may be cut and bent into conven- ient shapes without requiring annealing to restore its shielding properties.

12" x 12"/$32.50, 12" x 24"/$64.50, 12" x 48"/$127.00

More info on TI Shield...

 

Product is no longer available from either Percy or Gaven Industries.  Rep at Gaven indicates that replacement material is called Cu430 since TI too expensive to produce.  Apparently both were/are only available in 0.012" which will very likely be an issue to use as flexible shield?!  Regardless, I have asked for pricing and availability (Min qty) to us hobbyists. 

Stay tuned...