USB cable comparisons

[marce]

5 hours ago, seeteeyou said:

 

 

Maybe signal integrity ain't snake oil after all?

Signal integrity has been an important part of electronics since Oliver Heaviside, he came up with the telegraphers equations, circa 1880..... So its not that new.

[ALRAINBOW]

There’s  atleast one more item not mentioned that maybe worthwhile to say. 

The dac input and usb output have an impendance. This mismatch that we can hear in various forms of jitter or noise is a part of the usb sound tunning effects. I have heard usb Cables sound broken and great in another server same dac.  

I first observed this in a coaxial spidif cable 

I use CIEMS with a dac and amp in a pouch 

at one time I used a DAP as the source and later an iPhone. 

Not owning a real spidif cable I used a audio interconnect wow was there a varying in sound. 

I ended up buying a CUST cable form empiricle  audio made by STEVE Nugent. 

This yielded the best sound I could have. 

This was about 5 years ago.  A dark time in digital with some glimmer of lights glowing in the distance. 

The same remains in a AES cable the bal cable of spidif. It’s truly a shame it’s not used and we are left with usb.  Network or naa or there variants are also good but for me not at the level  Of a good tunned usb setup 

 

[str-1]

On 9/3/2018 at 4:36 PM, str-1 said:

Not sure if this is the right thread for this question but I will ask anyway.

 

When it can be used, will an unpowered usb cable more often than not sound better than a powered version of the same cable?  

 

I realise there will be a lot of “it depends” and “your mileage may vary” in any answer but I would welcome thoughts from anyone with experience of doing this kind of comparison.

Well, I’m pleased to say that I have now learned first hand the benefits that can be gained by using an unpowered cable, though I remain a little frustrated by this handshake business.

 

My chain is Innuos Zenith SE/SOtM tX-USBultra/Chord Blu2/DAVE.  I have all this out of the equipment rack and spread out over the floor at the moment to make it easier to try a few different cable combinations.  Knowing that Blu2 needs 5v power for handshake purposes, I decided to see what could be done between the Zenith and tX-U.  With everything on the floor I was able to get the Zenith and tX-U close enough to connect with an Uptone USPCB, which I think in this (less critical?) link of the chain sounds better than both the 20cm Curious and 70cm Lush cables I have.  I was pleased to find that everything continued to work when I switched off the USPCB’s 5v power and delighted to hear the immediate boost in sound quality, with noise levels dropping and everything becoming much cleaner and more punchy.  Result!

 

The frustration comes with the tX-U/Blu2 connection.  Although failing miserably to gain much knowledge about handshaking over USB by looking on the web, I did often see it said that while many dacs need the 5v power for the handshake this could be turned off after the connection is made.  Well, the tX-U has a switch over each of its two USB outputs that turns off the 5v power.  So with music playing quite happily I turned off the power for the output feeding the Blu2 and immediately lost all sound, only for this to return the moment I switched 5v back on.  Can someone explain to me what is happening here?  I have also asked this question on one of the Chord Electronics threads over on Head-Fi and expect Rob Watts will offer some thoughts from the dac end.

[reverendo]

4 hours ago, str-1 said:

Well, I’m pleased to say that I have now learned first hand the benefits that can be gained by using an unpowered cable, though I remain a little frustrated by this handshake business.

 

My chain is Innuos Zenith SE/SOtM tX-USBultra/Chord Blu2/DAVE.  I have all this out of the equipment rack and spread out over the floor at the moment to make it easier to try a few different cable combinations.  Knowing that Blu2 needs 5v power for handshake purposes, I decided to see what could be done between the Zenith and tX-U.  With everything on the floor I was able to get the Zenith and tX-U close enough to connect with an Uptone USPCB, which I think in this (less critical?) link of the chain sounds better than both the 20cm Curious and 70cm Lush cables I have.  I was pleased to find that everything continued to work when I switched off the USPCB’s 5v power and delighted to hear the immediate boost in sound quality, with noise levels dropping and everything becoming much cleaner and more punchy.  Result!

 

The frustration comes with the tX-U/Blu2 connection.  Although failing miserably to gain much knowledge about handshaking over USB by looking on the web, I did often see it said that while many dacs need the 5v power for the handshake this could be turned off after the connection is made.  Well, the tX-U has a switch over each of its two USB outputs that turns off the 5v power.  So with music playing quite happily I turned off the power for the output feeding the Blu2 and immediately lost all sound, only for this to return the moment I switched 5v back on.  Can someone explain to me what is happening here?  I have also asked this question on one of the Chord Electronics threads over on Head-Fi and expect Rob Watts will offer some thoughts from the dac end.

Some Dacs actually need the power from the 5v, not only for the handshake. Might be your case. 

[barrows]

@str-1:  Usually, DACs which incorporate an isolated USB input implement the interface like this:

 

The USB 5 volt power supply is used to power the USB receiver section, as this provides a completely separate, isolated power supply from the DAC's own internal power supply.  So, in your case, it is very likely that the Blu 2 oversmapler is using USB power to supply the isolated USB input section.  This is as it should be.  One can only have an isolated USB input if that input includes its own, separate power supply (otherwise the "isolation" would be broken by the shared ground connection) and the easiest way to get a separate supply is to use the USB 5 V supply.

 

 

[str-1]

Thanks, @barrowsI’ve learned something there.

[Cornan]

2 hours ago, barrows said:

@str-1:  Usually, DACs which incorporate an isolated USB input implement the interface like this:

 

The USB 5 volt power supply is used to power the USB receiver section, as this provides a completely separate, isolated power supply from the DAC's own internal power supply.  So, in your case, it is very likely that the Blu 2 oversmapler is using USB power to supply the isolated USB input section.  This is as it should be.  One can only have an isolated USB input if that input includes its own, separate power supply (otherwise the "isolation" would be broken by the shared ground connection) and the easiest way to get a separate supply is to use the USB 5 V supply.

 

 

 

Mind you that this does’nt mean that it is optimal. My Brooklyn DAC requires both 5v and GND and still improves immensely by both ISO Regen, BluWave USB to Spdif converter, HPULNs (higher A LT3045) and LiFePO4 powered LT3045s. Easiest and cheapest is really to buy a DAC that does’nt require 5v nor GND! ?

[asdf1000]

6 hours ago, barrows said:

it is very likely that the Blu 2 oversmapler is using USB power to supply the isolated USB input section.

 

Blu2's USB input is not isolated - it's BNC outputs are...

 

https://www.head-fi.org/threads/chord-electronics-blu-mk-2-the-official-thread.831343/page-248#post-14353662

 

[asdf1000]

And it's USB section is not externally powered:

 

"The Blu2 uses the VBUS only as a logic level flag - and it's RF filtered too. So removing it will ensure that Blu2 will not use the USB as a valid input. I can see no logical way that it can possibly change the SQ."

 

https://www.head-fi.org/threads/hugo-m-scaler-by-chord-electronics-the-official-thread.885042/page-73#post-14468015

 

 

Rob Watts is really easy to reach over on Head-Fi if you had any questions...

[Veovis]

18 hours ago, Cornan said:

 

Mind you that this does’nt mean that it is optimal. My Brooklyn DAC requires both 5v and GND and still improves immensely by both ISO Regen, BluWave USB to Spdif converter, HPULNs (higher A LT3045) and LiFePO4 powered LT3045s. Easiest and cheapest is really to buy a DAC that does’nt require 5v nor GND! ?

 

Which is probably why the SQ of the USB input of my Brooklyn benefits so greatly by powering it with a 5 v linear PSU instead of the 5 v from the Mac mini. Still potential issues with GND though I suppose. 

[highstream]

On 9/4/2018 at 7:36 AM, reverendo said:

I got to compare my Lightspeed 10G with the entry-level Gekko Black USB cable recently.

 

By Gekko, are you referring to https://gekkocables.com/products/digital links.html in London?  It's not easy to find with a google search from the States. You mention two choices, but I only see one shown on their site.

[str-1]

I’m find it a little frustrating comparing three upmarket USB cables in my setup (see signature).  

 

Recent changes I have made to the system have cleaned things up considerably and I am now finding it harder to reliably tell the difference between these cables.  It feels like I have already removed further upstream some of the crud that these cables normally have variable success in dealing with.  The difference between these and a 1m long stock cable (which I have put a number of ferrites on) is also at times hard to hear.  

 

I won’t name the cables but the one I expected to perform the best based on reputation also happens to be the longest at 1.2m, although the signal and power lines run in separate shielded cables between the connectors.  The stock cable is 1m and the upmarket cable I expected to sound second best is 0.7m.  I have a third upmarket cable which is a little harder to include in the comparison because it is only 0.20m long.  Does the different lengths of these cables mean this isn’t a level playing field for comparison?  I have heard it said many times that USB cables should be as short as possible but I have no feel for how big an impact USB cable length would/could have on sound quality.  

 

Generally, what differences should I expect to hear with different lengths of the same cable?  In a really good clean transparent system should I expect to hear any difference between a 0.5m and a 1.2m cable?

[Ralf11]

No.

 

If the system is galvanically isolated, reclocks, and the myriad of other possible problems have been taken care of... then there should be no differences in SQ among USB cables.

[One and a half]

19 minutes ago, Ralf11 said:

No.

 

If the system is galvanically isolated, reclocks, and the myriad of other possible problems have been taken care of... then there should be no differences in SQ among USB cables.

Theorist.

[Ralf11]

Theories are complicated constructs of hypotheses which have been tested.  So, yes.

[barrows]

While the concept of a "perfect" interface for USB which "should" make USB cable performance a non-issue is nice in theory, I hear plenty of differences between USB cables in a so called "perfect" set up, where the USB source is very, very clean, and the USB receiver is separately powered with complete galvanic isolation and re-clocking.

My conclusion is that in the real world, USB cables are always gong to matter, they may matter a bit less in a "perfect" set up, but they will still matter.

[xyzzy1]

52 minutes ago, Ralf11 said:

No.

 

If the system is galvanically isolated, reclocks, and the myriad of other possible problems have been taken care of... then there should be no differences in SQ among USB cables.

... if you have a mid-fi or worse audio system. If you have a entry level or better high-end audio system one can hear differences from noise caused by the cables themselves and noise they pass thru/distort which can have cumulative/reactionary effects with your other components. As always, we don’t yet have the science to ascertain the causes and symptoms so you have to go by audiblity and pick the best (subjective) that costs the least per what you can afford. There is no such thing as perfect electrical isolation when your audio or digital signal is passed between audio components using electrons. 

 

hey, this guy started it!

[xyzzy1]

25 minutes ago, Ralf11 said:

Theories are complicated constructs of hypotheses which have been tested.  So, yes.

 

A pretty smart guy said...

 

In theory, theory and practice are the same. In practice they are not. 

[pkane2001]

10 minutes ago, xyzzy1 said:

There is no such thing as perfect electrical isolation when your audio or digital signal is passed between audio components using electrons. 

 

If you believe that, use photons instead.

 

[reverendo]

On 9/10/2018 at 3:25 PM, highstream said:

By Gekko, are you referring to https://gekkocables.com/products/digital links.html in London?  It's not easy to find with a google search from the States. You mention two choices, but I only see one shown on their site.

oh, sorry. the black USB cable is their entry level cable and it's not on their site yet.

the one on the site is what I call the full-blooded version and is the one I want to try to see if it surpasses the LS in its strengths. let's see how that goes

[reverendo]

2 hours ago, str-1 said:

I’m find it a little frustrating comparing three upmarket USB cables in my setup (see signature).  

 

Recent changes I have made to the system have cleaned things up considerably and I am now finding it harder to reliably tell the difference between these cables.  It feels like I have already removed further upstream some of the crud that these cables normally have variable success in dealing with.  The difference between these and a 1m long stock cable (which I have put a number of ferrites on) is also at times hard to hear.  

 

I won’t name the cables but the one I expected to perform the best based on reputation also happens to be the longest at 1.2m, although the signal and power lines run in separate shielded cables between the connectors.  The stock cable is 1m and the upmarket cable I expected to sound second best is 0.7m.  I have a third upmarket cable which is a little harder to include in the comparison because it is only 0.20m long.  Does the different lengths of these cables mean this isn’t a level playing field for comparison?  I have heard it said many times that USB cables should be as short as possible but I have no feel for how big an impact USB cable length would/could have on sound quality.  

 

Generally, what differences should I expect to hear with different lengths of the same cable?  In a really good clean transparent system should I expect to hear any difference between a 0.5m and a 1.2m cable?

there are many theories about lengths of digital cables. Steve Nugent had some interesting theories about coax digital, saying that there was a minimal length that they should have. I am not sure if his musings apply to USB cables, though. I always tried to compare cables of similar lengths to take that variable out of the picture.

[reverendo]

2 hours ago, str-1 said:

I’m find it a little frustrating comparing three upmarket USB cables in my setup (see signature).  

 

Recent changes I have made to the system have cleaned things up considerably and I am now finding it harder to reliably tell the difference between these cables.  It feels like I have already removed further upstream some of the crud that these cables normally have variable success in dealing with.  The difference between these and a 1m long stock cable (which I have put a number of ferrites on) is also at times hard to hear.  

 

I won’t name the cables but the one I expected to perform the best based on reputation also happens to be the longest at 1.2m, although the signal and power lines run in separate shielded cables between the connectors.  The stock cable is 1m and the upmarket cable I expected to sound second best is 0.7m.  I have a third upmarket cable which is a little harder to include in the comparison because it is only 0.20m long.  Does the different lengths of these cables mean this isn’t a level playing field for comparison?  I have heard it said many times that USB cables should be as short as possible but I have no feel for how big an impact USB cable length would/could have on sound quality.  

 

Generally, what differences should I expect to hear with different lengths of the same cable?  In a really good clean transparent system should I expect to hear any difference between a 0.5m and a 1.2m cable?

Differences between cables diminished somewhat once I started using the REGEN in my setup, which is why I emphatically suggest people give it a try (I have no association with Uptone).

still, the differences are there and it depends on how sensitive the person is to those changes to make messing with them worthwhile

[Ralf11]

28 minutes ago, pkane2001 said:

 

If you believe that, use photons instead.

 

 

he posted magical results, so photons won't help

[Sonicularity]

My new DAC came with a generic, certified USB 2.0 cable.  Nothing fancy about it at all.  If the manufacturer of the DAC is not concerned about it, and I can't hear any differences, any certified USB cable should work the same.

[HumanMedia]

1 hour ago, Sonicularity said:

My new DAC came with a generic, certified USB 2.0 cable.  Nothing fancy about it at all.  If the manufacturer of the DAC is not concerned about it, and I can't hear any differences, any certified USB cable should work the same.

 

What is a certified USB cable?

And what makes you think the manufacturer is not concerned about the quality of the USB connection? Because they supplied a bare minimum so that the DAC could be connected?