Jump to content
IGNORED

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


Message added by The Computer Audiophile

Important and useful information about this thread

Posting guidelines

History and index of useful posts

Most important: please realize this thread is about bleeding edge experimentation and discovery. No one has The Answer™. If you are not into tweaking, just know that you can have a musically satisfying system without doing any of the nutty things we do here.

Recommended Posts

8 hours ago, Confused said:

The on-paper specs of the sCLK-OCX10 look superb, no question.  One key differentiator is that the REF10 is square wave and the sCLK-OCX10 is sine wave.  I am sure both will be excellent performers, as to which is best, or what the audible differences might be, I have no idea.  It would be a fascinating comparison.

 

It's a good point. Theoretically, a square wave signal has a much higher frequency spectrum, which would suggest that cable bandwidth, quality, as well as for the connectors would be more of a factor for a square wave generator. I don't know if this is in fact the case in practice.

Link to comment

Hi Guys, I am having trouble connecting Synology NAS to Roon using a bridged mini-mac config and full trifecta solution. I can see the NAS in the mini-mac finder and can get to NAS via the web. However, when I try to add it as a folder via "add network share" feature, it says path invalid. Any help would be really appreciated... thx

Link to comment
7 hours ago, Summit said:

 

I presume you perceive some SQ gain with a shunted switch directly followed by a modded switch. But do you have any insight to why and can’t the modded switch be shunted instead or is the reason for having one switch followed by another switch not to get the benefit of a shunted switch?

 

This convoluted setup was prompted by John Swenson's findings with regards to switches, SMPSes, and ground shunting. He has several postings across various threads that cover this, but this one from my bookmarks is probably a good one for an overview in regard to switches:

https://www.computeraudiophile.com/forums/topic/30376-a-novel-way-to-massively-improve-the-sq-of-computer-audio-streaming/?do=findComment&comment=730832

 

And this one gets into his observation that only some switches seem to block noise through the magnetics in each port:

https://www.computeraudiophile.com/forums/topic/35129-a-novel-way-to-massively-improve-the-sq-of-the-microrendu-ultrarendu/?do=findComment&comment=723196

 

The bottom line is that we don't know the universe of switches that block noise correctly. John has tested and confirmed that the Netgear FS/GS 105/108 switches are good. 

 

Given this data, the question for me was - what about my modded Zyxel switch? Obviously, if I could get John to test it on his measurement setup, I'd know. But John has made it clear he can't spend his entire waking life testing switches, and who can blame him! 

 

In considering the choices, it was actually John who came up with the suggestion of preceding the unknown (leakage characteristics) switch with a known one like the Netgear. 

 

As for SQ, when I added the Netgear prior to the modded Zyxel, I wasn't sure I heard much of an SQ difference at all. In keeping with my practice of periodically revisiting my optimizations to confirm they still matter, I will redo the experiment with and without the Netgear in my latest setup to see if it actually helps SQ or not.

Link to comment
1 hour ago, Johnseye said:

 

Given the price difference between the two, is it a $2000 improvement?  Understanding that if $ is no object one would buy the Ref 10.

Put another way, is there another place in someone's system where $2k would be put to better use and the OP-14 will suffice?

 

 

You know what I'm going to say, right! It depends. To some, absolutely, to others, absolutely not.

 

Key points:

  • With components of this quality, we are well into diminishing returns territory.
  • Ref 10 does have more features:
    • both 50 Ω and 75 Ω outputs
    • 8 outputs. This means an excellent distribution amp, which Kenji @ Cybershaft has told me would cost well north of $1000 if/when he makes one.

For me, I want to wait and see how this space shakes out a bit more. Also, want to see how the SOtM clock compares.

Link to comment
27 minutes ago, austinpop said:

 

You know what I'm going to say, right! It depends. To some, absolutely, to others, absolutely not.

 

Key points:

  • With components of this quality, we are well into diminishing returns territory.
  • Ref 10 does have more features:
    • both 50 Ω and 75 Ω outputs
    • 8 outputs. This means an excellent distribution amp, which Kenji @ Cybershaft has told me would cost well north of $1000 if/when he makes one.

For me, I want to wait and see how this space shakes out a bit more. Also, want to see how the SOtM clock compares.

My plan is to squeeze out a bit more performance from the OP-14. First step is to find a high quality 50Ohm BNC cable for a relatively sensible price.

Link to comment
10 minutes ago, mozes said:

My plan is to squeeze out a bit more performance from the OP-14. First step is to find a high quality 50Ohm BNC cable for a relatively sensible price.

Let us know if you find something reasonable. For now I am going to order this one from Japan https://item.rakuten.co.jp/auc-treevillage/8-2-bncp-bncp-06/ as I have not found anything else under several hundred dollars.

HQplayer - NAA - Devialet D-800 - YG Acoustics Carmel + dual ELAC sub-2090

Link to comment

Both SOtM dCBL-BNC50 and dCBL-BNC75 should cost $700 per meter, though we really need something like 0.5m instead

 

http://www.sotm-audio.com/sotmwp/english/shop/dcbl-bnc/

 

Still a good deal if they're performing even better than those top dogs from Shunyata Research.

 

BTW, it's gotta be mighty interesting to compare €700 Habst @ 0.5m versus $700 SOtM @ 1.0m IMHO. Both of them are terminated with 75Ω BNC connectors, though Habst should benefit from its separate ground while SOtM just happened to have a better filter as usual.

Link to comment
18 minutes ago, seeteeyou said:

Both SOtM dCBL-BNC50 and dCBL-BNC75 should cost $700 per meter, though we really need something like 0.5m instead

 

http://www.sotm-audio.com/sotmwp/english/shop/dcbl-bnc/

 

Still a good deal if they're performing even better than those top dogs from Shunyata Research.

 

BTW, it's gotta be mighty interesting to compare €700 Habst @ 0.5m versus $700 SOtM @ 1.0m IMHO. Both of them are terminated with 75Ω BNC connectors, though Habst should benefit from its separate ground while SOtM just happened to have a better filter as usual.

Thanks, still at $700 it is around 50% of the cost of the OP-14.

Link to comment
23 minutes ago, seeteeyou said:

Both SOtM dCBL-BNC50 and dCBL-BNC75 should cost $700 per meter, though we really need something like 0.5m instead

 

http://www.sotm-audio.com/sotmwp/english/shop/dcbl-bnc/

 

Still a good deal if they're performing even better than those top dogs from Shunyata Research.

 

BTW, it's gotta be mighty interesting to compare €700 Habst @ 0.5m versus $700 SOtM @ 1.0m IMHO. Both of them are terminated with 75Ω BNC connectors, though Habst should benefit from its separate ground while SOtM just happened to have a better filter as usual.

The non-boutique 50 ohm cable recommended by Mutec is RG400.  Keep it short, .5m or shorter, and order from Pasternack with 50 ohm connectors, $50-$60 US range depending on connectors.

 

Warning, when I ordered from Pasternack, ship to Canada, UPS charged me $40 brokerage fee to collect $7 GST.

Link to comment
4 minutes ago, d_elm said:

The non-boutique 50 ohm cable recommended by Mutec is RG400.  Keep it short, .5m or shorter, and order from Pasternack with 50 ohm connectors, $50-$60 US range depending on connectors.

 

Warning, when I ordered from Pasternack, ship to Canada, UPS charged me $40 brokerage fee to collect $7 GST.

Thanks for the info and the warning. Did you have a chance to

compare it with the stock cable?

Link to comment
54 minutes ago, mozes said:

My plan is to squeeze out a bit more performance from the OP-14. First step is to find a high quality 50Ohm BNC cable for a relatively sensible price.

 

Why not make a custom build and let Ghent Audio do it for you? Have a look at QED Reference coaxial spdif cables (with dual center core wires) and just run an additional wire from the shield (negative) to a 2.1/5.5mm DC cable to connect to the Silver Minimus? It could be worth the gamble if you ask me! ?

 

 

Link to comment
5 minutes ago, Cornan said:

 

Why not make a custom build and let Ghent Audio do it for you? Have a look at QED Reference coaxial spdif cables (with dual center core wires) and just run an additional wire from the shield (negative) to a 2.1/5.5mm DC cable to connect to the Silver Minimus? It could be worth the gamble if you ask me! ?

You read my mind :) I have the Entreq in my head to use as a ground since my OP-14 has only one output. In fact I can try grounding it with the cable I have and see if it improves SQ. 

Link to comment
1 hour ago, seeteeyou said:

Both SOtM dCBL-BNC50 and dCBL-BNC75 should cost $700 per meter, though we really need something like 0.5m instead

 

http://www.sotm-audio.com/sotmwp/english/shop/dcbl-bnc/

 

Still a good deal if they're performing even better than those top dogs from Shunyata Research.

Two things. 

For one, if you are in the EU, anything from outside the EU will cost 30-40% on top of the list price. In this case SOtM cables will be close to $1000 (I am almost sure that they will be sold around 800 euros) which makes them equally expensive as Habst. Shunyata is much more, needless to say.

 

Second, I bought a tx-USB ultra the other day as the last piece of the puzzle. I opened it to change the voltage to 12V. I noticed that the clock cable that goes from the input connector (inside of the BNC) to the board is curled up. I measured it and it is 40cm. By looking at it, it looks like more industrial than something in the league of high-end clock cables. I wonder how much our effort (both keeping short lengths and high quality) matters in this case when the last 40cm is something like that. 

Then again, given that SOtM is as good as it is (both products and as a company), my only conclusion is that they don't think that this matters that much. But why do they sell the first 1m for $700? As you can see, I am a little confused here.

IMG_0764.JPG

HQplayer - NAA - Devialet D-800 - YG Acoustics Carmel + dual ELAC sub-2090

Link to comment
29 minutes ago, zoltan said:

But why do they sell the first 1m for $700? As you can see, I am a little confused here.

 

I'm more than just a little confused here as well, why do Shunyata Research sell the first 1m for $1,000 (Alpha Clock 50) and $2,000 (Sigma Clock 50) respectively?

 

BTW, they do have distributors throughout Europe and let's see if they're charging more than how much SOtM would ask for

 

https://www.sotm-audio.com/sotmwp/english/support/distributors/

 

Maybe blame EU or whatever officials who decided to set those rules up for imported goods? Or you could persuade SOtM to relocate their business instead. LOL

Link to comment
1 hour ago, zoltan said:

Two things. 

For one, if you are in the EU, anything from outside the EU will cost 30-40% on top of the list price. In this case SOtM cables will be close to $1000 (I am almost sure that they will be sold around 800 euros) which makes them equally expensive as Habst. Shunyata is much more, needless to say.

 

Second, I bought a tx-USB ultra the other day as the last piece of the puzzle. I opened it to change the voltage to 12V. I noticed that the clock cable that goes from the input connector (inside of the BNC) to the board is curled up. I measured it and it is 40cm. By looking at it, it looks like more industrial than something in the league of high-end clock cables. I wonder how much our effort (both keeping short lengths and high quality) matters in this case when the last 40cm is something like that. 

Then again, given that SOtM is as good as it is (both products and as a company), my only conclusion is that they don't think that this matters that much. But why do they sell the first 1m for $700? As you can see, I am a little confused here.

IMG_0764.JPG

 

Hi Zoltan, i hear you on both counts that is why i went the jcat femto usb and with newclassd clocks. Newclassd subtract there vat (Denmark). 

Clocking cables should be as short as possible, this is a waste of clock quality. I guess they work with “standard” length cable.

My longest clock cable is 9,5 cm, but i use  multiple clocks.

Meitner ma1 v2 dac,  Sovereign preamp and power amp,

DIY speakers, scan speak illuminator.

Raal Requisite VM-1a -> SR-1a with Accurate Sound convolution.

Under development:

NUC7i7dnbe, Euphony Stylus, Qobuz.

Modded Buffalo-fiber-EtherRegen, DC3- Isoregen, Lush^2

Link to comment

Today i succesfully implemented a clock operation on my mobo. I replaced the network nic oscillator with a newclassd neutrino, 25mhz.

It is burning in now, but it sounds promising.

IMG_1075.thumb.JPG.8785ce0b3b54dc37e543dde837994ba7.JPG

 

IMG_1080.thumb.JPG.37b49c6d271848f9892fb62f1ea015af.JPG

 

the copper box above the mobo is the new clock.

As you can see copper is my go to material.

Meitner ma1 v2 dac,  Sovereign preamp and power amp,

DIY speakers, scan speak illuminator.

Raal Requisite VM-1a -> SR-1a with Accurate Sound convolution.

Under development:

NUC7i7dnbe, Euphony Stylus, Qobuz.

Modded Buffalo-fiber-EtherRegen, DC3- Isoregen, Lush^2

Link to comment
7 minutes ago, RickyV said:

Today i succesfully implemented a clock operation on my mobo. I replaced the network nic oscillator with a newclassd neutrino, 25mhz.

It is burning in now, but it sounds promising.

IMG_1075.thumb.JPG.8785ce0b3b54dc37e543dde837994ba7.JPG

 

IMG_1080.thumb.JPG.37b49c6d271848f9892fb62f1ea015af.JPG

 

the copper box above the mobo is the new clock.

As you can see copper is my go to material.

Great job you did there and absolutely love the copper boxes.

Link to comment
2 hours ago, zoltan said:

I noticed that the clock cable that goes from the input connector (inside of the BNC) to the board is curled up. I measured it and it is 40cm. By looking at it, it looks like more industrial than something in the league of high-end clock cables.

 

I am a little confused here.

IMG_0764.JPG

 

Wow, that is unsettling! I have yet to open up mine, but this would bum me out. If yours is a new unit, I would suggest sending it back and insisting they trim the cables down to size.  Or consider asking them to upgrade all the clock cabling from the board to the backplane to silver. I seem to recall some others doing that.

Link to comment
41 minutes ago, spotforscott said:

Man I wish I knew that you have to have a master in computer science to know how to connect a NAS drive to a bridged macmini. Guess I will explore external hard drives or sell the macmini and get a server with enough built in storage...

 

Scott, is this the same question you PM'ed me?

 

I've never run Roon Core on a Mac, but I find it weird that if you can "see" and navigate to your music folder on the NAS in Finder, there is no reason why you should not be able to do so in you Storage settings in Roon.

 

The only other gotcha for Roon with the sMS is to make sure you've enabled the Roon Ready app in the sMS. You should be able to connect to the sMS by typing http://eunhasu.local (on a Mac) in your browser. Make sure your Roon Ready app is "Active." You should also go through the settings - both the general settings, and the Roon settings.

Link to comment
7 hours ago, limniscate said:

@austinpop forgot to add that I was doing the tests blind.  I thought I knew what some of the changes were because I could kind of hear him connecting/disconnecting, but it turned out I was wrong!  E.g. the silver DC cable versus the copper one, I assumed it was a power supply change!

 

BTW - the impact of the silver cable seems to matter what component it's for. In my system, I have always found the PSU (and now the cable) that powers the tX-USBultra immediately prior to the DAC to be the most critical.

 

All the (blind) experiments I did with Eric with PSUs and cables were with the tX-USBultra.

 

Last night, I also experimented with the modded switch further upstream. There too, I had replaced the LPS-1 with the LPS-1.2, and noticed an SQ benefit, although not as pronounced as on the tX-USBultra. On the modded switch, I did not find the silver cable to make as much of a difference, and I'm not convinced I could reliably tell in a blind test.

 

Given that I already have a silver cable on my SR-4 for the tX-USBultra, I am still debating whether to buy a silver cable for my switch PSU. Has anyone found a good priced cable? Zenwave has one for $150 for 0.3m (1 foot). 

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...