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

[rafa]

19 hours ago, Johnseye said:

 

No it does not have to be connected.  I've read the manual.  Point out where it says it "has" to be connected.  I do not have it connected and I power it though an external source.

 

You missed what I posted earlier from May at SOtM.

 

"The tX-USBexp can take 12V internal power, but when you use external power to tX-USBexp, it can’t be 12V, it should be max.9V.

You can power to tX-USBexp externally(9V) and power to sCLK-EX via external 9V-> tX-USBexp-> sCLK-EX directly without SATA connection.

and If you use 12V IDE power to tX-USBexp, yes, you can use this power to sCLK-EX via the J403(tX-USBexp) to  sCLK-EX’s Power input connector(Check the page 2 on the manual), the sCLK-EX should set to accept 12V."

Sorry if I mist something.

 

But "has to be" is when says:

 

4.If you connect the IDE power cable inside of PC to the IDE power input connector of tX-USBexp while the tX-USBexp get powered by the external power input, it can prevent the damage of the internal parts of tX-USBexp or PC when the external power suddenly cut off

 

For me it is as must be.

[Johnseye]

6 hours ago, rafa said:

Sorry if I mist something.

 

But "has to be" is when says:

 

4.If you connect the IDE power cable inside of PC to the IDE power input connector of tX-USBexp while the tX-USBexp get powered by the external power input, it can prevent the damage of the internal parts of tX-USBexp or PC when the external power suddenly cut off

 

For me it is as must be.

 

"IF" is not "has to be" or must be.  However, you are free to do as you wish.  Just don't inform people they must do something when they do not.  So for you, if it must be, then great.  For me and others who choose not to, it is an option.

 

I prefer to have one clean power source to that card.  It works, sounds great and I've had no issues.

 

[mikicasellas]

On 15/2/2018 at 1:35 PM, Johnseye said:

 

ROCK is the OS unless you're referring to a NUC sold by Roon, which I'm not aware if they are selling those or not.

 

There are many ways to skin this cat.  If you want simple and great sound, get a Zenith.  If you want to build your own for great sound you can use ROCK.  If you just want simple, buy a standard NUC with ROCK on it.

 

Thanks Johnseye:

 

So much BUZZ about ROCK, i thought that yes it is a ROON OS but in a way that makes it better/dedicated

 

So if i build a NUC with ROCK, it should be the same like having ROON on my Mac mini?

[Johnseye]

3 minutes ago, mikicasellas said:

 

Thanks Johnseye:

 

So much BUZZ about ROCK, i thought that yes it is a ROON OS but in a way that makes it better/dedicated

 

So if i build a NUC with ROCK, it should be the same like having ROON on my Mac mini?

 

ROCK OS is Linux based and Roon customized it for Roon music listening.  Whether it sounds better or not is up to your ears and what else you're doing. For example if you want to use HQ Player you can't install it on ROCK. You also can't bridge NICs with ROCK. 

 

The Mac and NUC hardware are different and will have their own influence on the sound. 

 

There are a lot of variables to consider. I had been using ROCK for a few months and went back to Windows. I can use HQ Player with Roon and bridge my NICs again eliminating the switch. I think it sounds slightly better. 

[mikicasellas]

14 minutes ago, Johnseye said:

 

ROCK OS is Linux based and Roon customized it for Roon music listening.  Whether it sounds better or not is up to your ears and what else you're doing. For example if you want to use HQ Player you can't install it on ROCK. You also can't bridge NICs with ROCK. 

 

The Mac and NUC hardware are different and will have their own influence on the sound. 

 

There are a lot of variables to consider. I had been using ROCK for a few months and went back to Windows. I can use HQ Player with Roon and bridge my NICs again eliminating the switch. I think it sounds slightly better. 

 

Excellent thank you for clearing that for me!!

 

Cheers!

[mikicasellas]

Hello Guys,

 

For those with more and vast experience than me (mostly all of you)

 

What do you Guys feel is the better first step for upgrading components if you were in my shoes? (DAC or SERVER)

 

I am considering upgrading my DAC (OPPO 205) and my Mac mini as well: 

 

But i need to do it one step at the time: either begin with DAC upgrade or Server Upgrade...

 

If going with DAC in first place:

 

I could mod my OPPO or get another DAC like DenafripsTerminator or a used Bricasti M1SE (more investment)

 

1.-NEW Denafrips Terminator: $4,400

2.- Used Bricasti M1SE: $5,500

3.-Sending my OPPO 205 for Modification ($850):

   1.- Linear Power supply SE
   2.- OCXO clock mod & TCXO clock mod
   3.- DAC mod

 

--- DAC BOARD: "The main stereo balance and unbalanced output should be modified mainly.
The electrolytic capacitor is now used, they should be replaced with OSCON(SANYO or Panasonic) and Wima film capacitor ( MKS2 1uF/63V)

Coupling capacitors are a very important part where the sound signal passes directly. The function is to cut off DC voltage and pass only music signal, which is very directly related to sound quality."

 

Send only main board and 2 channel stereo DAC board .
 then It will be mod to those boards for OCXO & TCXO clock mod and dac board mod. ---

 

If going with SERVER in first place and keep the OPPO in the meantime while i save founds for either modding the OPPO or getting another DAC and using the OPPO as a transport (CDS and SACDS):

 

Zenith MKII ($3,400):

 

So what are your thoughts: SERVER or DAC?

 

Cheers!!

 

[AmusedToD]

48 minutes ago, Johnseye said:

 

ROCK OS is Linux based and Roon customized it for Roon music listening.  Whether it sounds better or not is up to your ears and what else you're doing. For example if you want to use HQ Player you can't install it on ROCK. You also can't bridge NICs with ROCK. 

 

The Mac and NUC hardware are different and will have their own influence on the sound. 

 

There are a lot of variables to consider. I had been using ROCK for a few months and went back to Windows. I can use HQ Player with Roon and bridge my NICs again eliminating the switch. I think it sounds slightly better. 

 

In my experience, ROCK on a NUC in a fanless case powered by linear PSU sounds way better than a "heavy" Windows 10 machine. The main difference is in the clarity and resolution department. And yes, you can do the direct connection under ROCK, as a matter of fact that's how I am using it. So, to clarify:

 

NUC - usb to ethernet adapter - network switch (internet connection needed for controlling Roon)

NUC - lan output - Roon endpoint (direct or bridged connection)

 

No need for any additional settings, this is just plug and play. 

[Johnseye]

1 hour ago, AmusedToD said:

 

In my experience, ROCK on a NUC in a fanless case powered by linear PSU sounds way better than a "heavy" Windows 10 machine. The main difference is in the clarity and resolution department. And yes, you can do the direct connection under ROCK, as a matter of fact that's how I am using it. So, to clarify:

 

NUC - usb to ethernet adapter - network switch (internet connection needed for controlling Roon)

NUC - lan output - Roon endpoint (direct or bridged connection)

 

No need for any additional settings, this is just plug and play. 

 

How did you bridge the NICs in ROCK?  Are you certain they are actually bridged?

[Johnseye]

I ask because I do not use an endpoint but am able to directly connect my NAS with a bridged connection to my server. That NAS needs to be accessible from other devices on my network. 

[AmusedToD]

1 hour ago, Johnseye said:

 

How did you bridge the NICs in ROCK?  Are you certain they are actually bridged?

 

Just for the purpose of clarification, the connection between the NUC/ROCK and the Roon endpoint (dCS NBR in my case) is direct, meaning no router or switch in between, just an ethernet cable. According to Danny of Roon Labs, such connection provides theoretical benefits by being electrically isolated from the network.  

 

Internet connection to the NUC/ROCK is provided through a simple USB to ethernet adapter (network switch -> ethernet cable -> AmazonBasics USB to ethernet adapter plugged into a USB port on the NUC). 

 

In this configuration, the NUC has internet access which is necessary for remote control of Roon software (as well as updates).

 

The endpoint does not have internet access. As mentioned, this provides electrical isolation of the RAAT communication protocols from the network traffic.

 

Everything is just plug and play, no additional settings are necessary.

[rickca]

Several people have long reported that providing clean power to the OS disk (Windows 10 Pro in my case) makes a significant difference.

 

I moved my OS from a Samsung 950 Pro M.2 NVMe SSD to an old Intel X25-M SATA II SSD and got an SOtM SATA II filter.  I am powering the X25-M with a Paul Hynes SR4 using a DC jack to SATA power connector cable from HDPLEX.

 

My first impression is really positive.  Better clarity, better dynamics and really authoritative bass.  I'm listening to Nikolai Lugansky play the Rachmaninov Preludes.  It's a wonderful album.

 

I don't get it.  This is just the OS disk.  Why does it make such a remarkable improvement?  Glad I tried it.

[Johnseye]

38 minutes ago, AmusedToD said:

 

Just for the purpose of clarification, the connection between the NUC/ROCK and the Roon endpoint (dCS NBR in my case) is direct, meaning no router or switch in between, just an ethernet cable. According to Danny of Roon Labs, such connection provides theoretical benefits by being electrically isolated from the network.  

 

Internet connection to the NUC/ROCK is provided through a simple USB to ethernet adapter (network switch -> ethernet cable -> AmazonBasics USB to ethernet adapter plugged into a USB port on the NUC). 

 

In this configuration, the NUC has internet access which is necessary for remote control of Roon software (as well as updates).

 

The endpoint does not have internet access. As mentioned, this provides electrical isolation of the RAAT communication protocols from the network traffic.

 

Everything is just plug and play, no additional settings are necessary.

 

You are not bridging then.  The endpoint has no Internet access.  This will not work in my scenario.

 

[Johnseye]

32 minutes ago, rickca said:

Several people have long reported that providing clean power to the OS disk (Windows 10 Pro in my case) makes a significant difference.

 

I moved my OS from a Samsung 950 Pro M.2 NVMe SSD to an old Intel X25-M SATA II SSD and got an SOtM SATA II filter.  I am powering the X25-M with a Paul Hynes SR4 using a DC jack to SATA power connector cable from HDPLEX.

 

My first impression is really positive.  Better clarity, better dynamics and really authoritative bass.  I'm listening to Nikolai Lugansky play the Rachmaninov Preludes.  It's a wonderful album.

 

I don't get it.  This is just the OS disk.  Why does it make such a remarkable improvement?  Glad I tried it.

 

Glad to hear you got the SSD and filter in line.  The SSD generates noise.  That X25 generates less and the filter removes even more.

[AmusedToD]

1 hour ago, Johnseye said:

 

You are not bridging then.  The endpoint has no Internet access.  This will not work in my scenario.

 

 

Why do you need internet on the endpoint side?

[Johnseye]

8 minutes ago, AmusedToD said:

 

Why do you need internet on the endpoint side?

 

I don't use an endpoint. As I mentioned before my NAS is directly connected through my server. Internet access to by NAS allows access to my music from anywhere. LAN access to my NAS.... 

[austinpop]

18 minutes ago, AmusedToD said:

 

Why do you need internet on the endpoint side?

 

Well, an endpoint has a web UI for configuration, management, software updates etc. How do you access it otherwise?

 

Do you know for a fact that ROCK is not bridging? You'd have to look at the IP configurations (address, subnet, gateway) on both of the ROCK interfaces as well as the endpoint to know definitively. Or post them here!

[seeteeyou]

2 hours ago, rickca said:

I don't get it.  This is just the OS disk.  Why does it make such a remarkable improvement?  Glad I tried it.

 

Sam Laufer actually went all out to do stuff like this

 

http://www.hiendy.com/hififorum/forum.php?mod=viewthread&tid=114571&page=1#pid2745381

Quote

write in sequential ,  and read in sequential ..... will NOT spread around NAND flash ,  it is their US Patent.

 

https://patents.justia.com/inventor/sam-laufer

 

System and method for suppressing jitter
https://patents.justia.com/patent/8451701

 

System and method for suppressing jitter in digital data signals including image, video and audio data signals
https://patents.justia.com/patent/8995243

 

32-core Memory Player would set you back $32,950 and that really wasn't a typo by any means

 

https://www.thememoryplayer.net/the-digital-transport-

http://www.lauferteknik.com/product-descriptions-the-memory-player/

 

It would be even more interesting if we're booting off OS disk with only one or maybe two NAND flash memory chips. 64GB ones cost $45 or so

 

https://www.aliexpress.com/item/eVtran-high-4K-performance-460MB-s-32G-64G-USB3-0-Flashdisk-USB3-0-SLC-flash-drive/32846479966.html

https://item.taobao.com/item.htm?id=562670493852

https://item.taobao.com/item.htm?id=563376880489

 

128GB ones would cost $85 for M.2 and about $100 for USB 3.0 drives respectively, obviously those prices implied SLC chips rather than MLC

 

https://item.taobao.com/item.htm?id=525911902038

https://item.taobao.com/item.htm?id=549521154761

 

They also have 250GB M.2 drives for $150

 

https://item.taobao.com/item.htm?id=522026308068

 

Each SLC chip (largest ones should be 512Gb or 64GB a piece) really doesn't consume much power because even the maximum would be merely 50mA

 

https://www.micron.com/products/nand-flash/slc-nand/512Gb

 

MLC ones seemed to have fairly similar numbers

 

https://4donline.ihs.com/images/VipMasterIC/IC/MICT/MICTS04810/MICTS04810-1.pdf#page=129

[Johnseye]

3 hours ago, austinpop said:

 

Well, an endpoint has a web UI for configuration, management, software updates etc. How do you access it otherwise?

 

Do you know for a fact that ROCK is not bridging? You'd have to look at the IP configurations (address, subnet, gateway) on both of the ROCK interfaces as well as the endpoint to know definitively. Or post them here!

 

It actually may be bridging when the IPs are set statically.  I've since uninstalled ROCK so can't test, but if anyone else can this would be good to know.  I'd love to know if ROCK actually does this natively, but I'm starting to think not.

 

Here's a post on the topic.  Danny's more concerned about isolation.  At one point he mentions its possible, but I'm not sure if he was referring to something else.

 

https://community.roonlabs.com/t/bridging-network-interfaces/19233/26

 

[AmusedToD]

8 hours ago, austinpop said:

 

Well, an endpoint has a web UI for configuration, management, software updates etc. How do you access it otherwise?

 

Do you know for a fact that ROCK is not bridging? You'd have to look at the IP configurations (address, subnet, gateway) on both of the ROCK interfaces as well as the endpoint to know definitively. Or post them here!

 

You access it by unplugging the ethernet cable that comes from the NUC and plugging another one that runs from the switch. Literally a 1 second job.

 

Yes, Danny was pretty straight about this. He said bridging doesn’t make any sense unless you electrically isolate the direct connection from the server to the endpoint.  

[str-1]

Following up on a very helpful pm, I discussed with May what upgrade options might be available when ordering a tx-USBultra.  She confirmed that I could order with either 7N UPOCC copper or 7N UPOCC silver internal wiring upgrades, and also with an EMI absorption sheet.  Has anyone been able to compare the stock tx-USBultra with one that has upgraded internal wiring, or who has been able to compare with and without the EMI sheet (which I am assuming I would have to cut up and place rather than it being already attached in the best places)?  Thanks for any feedback on this.

[chauphuong]

Danny is in the right position to say no to bridging simply because his Roon ROCK just doest NOT support bridging. Experience it yourself. Anyhow I think this topic (especially its 1st post) becoming possibly the most hectic here at CA is for a reason.

[lmitche]

58 minutes ago, chauphuong said:

Anyhow I think this topic (especially its 1st post) becoming possibly the most hectic here at CA is for a reason.

And what reason is that?

[AmusedToD]

1 hour ago, chauphuong said:

Danny is in the right position to say no to bridging simply because his Roon ROCK just doest NOT support bridging. Experience it yourself. Anyhow I think this topic (especially its 1st post) becoming possibly the most hectic here at CA is for a reason.

 

Linix does support bridging, doesn’t it? But Danny specifically stated he won’t allow bridging under ROCK (which is a special version of Linux).

 

Anyways, what is the benefit of direct bridged connection (with internet access on the endpoint side) vs. direct non-bridged connection (without internet access on the endpoint side) in terms of sound quality? I would say none. In fact, Danny says the non-bridged direct connection should have SQ benefits from being electrically isolated from the noisy network.

 

My point is, he can probably enable it under ROCK, but he refuses to do so. I asked him and he specifically said no.

[Johnseye]

6 minutes ago, AmusedToD said:

 

Linix does support bridging, doesn’t it? But Danny specifically stated he won’t allow bridging under ROCK (which is a special version of Linux).

 

Anyways, what is the benefit of direct bridged connection (with internet access on the endpoint side) vs. direct non-bridged connection (without internet access on the endpoint side) in terms of sound quality? I would say none. In fact, Danny says the non-bridged direct connection should have SQ benefits from being electrically isolated from the noisy network.

 

My point is, he can probably enable it under ROCK, but he refuses to do so. I asked him and he specifically said no.

 

What I find ironic is that I stated ROCK couldn't bridge and you replied, and I won't quote it all here but, "NUC - lan output - Roon endpoint (direct or bridged connection)".  But now you're stating you asked Danny and he said no.

 

Danny may know Roon but that doesn't mean he knows everything.  If you go back and read through this thread you'll see the countless people who have experienced an audible benefit from bridging.

 

If you can completely eliminate a switch, and if you've read through this and other threads here you'll know a switch and its power supply introduces noise, then you eliminate its noise.

 

I think you've read enough here to know that upstream sources have an influence on SQ.  Do it if you want, or don't do it at all, but you've just stirred up a pot of nothing for no reason.  If you like ROCK use it.  I happened to like what I heard when I switched back to Windows, using HQPlayer and Roon.  If I could eliminate the switch as I can with Windows I'd probably give ROCK a try again.

[Johnseye]

26 minutes ago, lmitche said:

And what reason is that?

 

I think he's saying that it's because so many people have experienced a benefit from bridging, as well as a myriad of other tweaks and improvements that have made this thread into a valuable source.