advertisement


Audiophile Network Switches for Streaming ... really ?

Octavian said:
Threads like these are sometimes criticized for being circular - particularly by friends of subjectivist philosophy. Now I understand why.
Click to expand...

The OP is about audiophile network switches but other related aspects of digital audio have been discussed, namely the effets of noise on clock and D/A chip performance (it's not about the data).
There is the view that digital audio is robust enough (perfect) but reality doesn't always live up to the theory...
I accept that in the grand scheme of things this is a minor issue, bot not one which room interference is able to mask. Does it matter? Depends on who you ask.
 
TheFlash said:
Well clearly I won't be able to convinve you otherwise, but you have no basis other than cynicism for assuming anything about my motives.

And you have no basis to presume you're less smart than me, I can assure you!

I didn't slip anything in by the way, others mentioned this. This thread has nothing to do with any one switch, though the switch-antogonistic have recently started playing the man not the ball. I think I'll leave them to play now.

Good luck with the USB cable :)
Click to expand...
Fair enough, I’m not really a cynic and shockingly haven’t read every post. Good luck in your venture.
 
NRG said:
What is a “properly installed switch”? this is golden, pray do tell, I’m all ears on this fascinating insight. Educate me.

Oh BTW, noise on ethernet, here’s your solution.

https://www.hamradio.co.uk/dx-engineering-iso-plus-ethernet-rf-filters-dxe-iso-plus-2-pd-10944

Twitchy buy now finger?
Click to expand...
Why the bitchy sarcasm? Is this how you talk to people in the office and down the pub?

A properly installed switch simply means one which is installed:
- 0.5m or so from streamer NOT at the router end where they were designed to be installed for port replication
- attached to streamer by an unshielded CAT compliant cable rather than a shielded or non-CAT (ideally 0.5m, max 1m long)
 
tuga said:
The OP is about audiophile network switches but other related aspects of digital audio have been discussed, namely the effets of noise on clock and D/A chip performance (it's not about the data).
There is the view that digital audio is robust enough (perfect) but reality doesn't always live up to the theory...
I accept that in the grand scheme of things this is a minor issue, bot not one which room interference is able to mask. Does it matter? Depends on who you ask.
Click to expand...
And again, against all the measurements in those papers you have posted, no one has provided anything at the same level to illustrated what a switch next to a streamer/DAC does.

Post papers full of measurement, but it kinda defects the object if that level of objective analysis is then shunned by posters in this thread.

Post more, it just adds to the irony.
 
tuga said:
Isn't that what the EtherREGEN's set out to do?

https://uptoneaudio.com/products/etherregen

EtherREGEN_Tech.Highlights_web.jpg
Click to expand...
I like the way the review quote says that it doesn't meet its manufacturer's claim, only that it is cheaper :)
 
chiily said:
And again, against all the measurements in those papers you have posted, no one has provided anything at the same level to illustrated what a switch next to a streamer/DAC does.

Post papers full of measurement, but it kinda defects the object if that level of objective analysis is then shunned by posters in this thread.

Post more, it just adds to the irony.
Click to expand...
It would be interesting to know though whether the TI measurements are what you expect as a bare minimum ie. if any switch manufacturer were to provide same then you'd accept them and move on. Not that I can imagine any such switch manufacturer doing same, just want to know if TI meets the grade.
 
TheFlash said:
It would be interesting to know though whether the TI measurements are what you expect as a bare minimum ie. if any switch manufacturer were to provide same then you'd accept them and move on. Not that I can imagine any such switch manufacturer doing same, just want to know if TI meets the grade.
Click to expand...
The level of measurement would be confined just to the audio band at the output of the DAC. Something like a null tester.

I was more poking fun at the irony of posting stuff full of measurements, yet nothing offered about the impact of switches, or measurements being simply dismissed. To me it seemed strange using measurements to prove one thing, but not willing to trust measurements on the other hand.
 
TheFlash said:
Why the bitchy sarcasm? Is this how you talk to people in the office and down the pub?

A properly installed switch simply means one which is installed:
- 0.5m or so from streamer NOT at the router end where they were designed to be installed for port replication
- attached to streamer by an unshielded CAT compliant cable rather than a shielded or non-CAT (ideally 0.5m, max 1m long)
Click to expand...

Sorry, that was a cheap shot. I will say through when you start trotting out terms like ‘factometer’ and ‘properly installed switch’ you are going to draw push back.

Tuga posted a link to an Analog Devices paper that clearly states on the first page:

“Now what is of interest is that the noise and all the curves are on top of each other. As a result, in this product line, the interface is not of concern, notwithstanding those spurs that may need attention depending on system requirements. Finding that the interface is of little concern leads into the next area of interest: clocking.”

So as the interface is of no concern related to phase noise how does all this talk of removing noise from the interface connection have any affect. What that article does highlight is the need for an accurate clock for the DAC and clean clock supply rails…but this is aimed at DAC implementation not what is put upstream of the DAC.
 
NRG said:
Sorry, that was a cheap shot. I will say through when you start trotting out terms like ‘factometer’ and ‘properly installed switch’ you are going to draw push back.

Tuga posted a link to an Analog Devices paper that clearly states on the first page:

“Now what is of interest is that the noise and all the curves are on top of each other. As a result, in this product line, the interface is not of concern, notwithstanding those spurs that may need attention depending on system requirements. Finding that the interface is of little concern leads into the next area of interest: clocking.”

So as the interface is of no concern related to phase noise how does all this talk of removing noise from the interface connection have any affect. What that article does highlight is the need for an accurate clock for the DAC and clean clock supply rails…but this is aimed at DAC implementation not what is put upstream of the DAC.
Click to expand...

What seems to be surfacing is that the interface (USB, Ethernet) may actually be a problem (actually it's noise), hence endpoints/network-brides, galvanic isolation in USB interfaces, reducing processing and network activity to the bare minimum, etc.
 
tuga said:
What seems to be surfacing is that the interface (USB, Ethernet) may actually be a problem (actually it's noise), hence endpoints/network-brides, galvanic isolation in USB interfaces, reducing processing and network activity to the bare minimum, etc.
Click to expand...

Have you got a paper considering 50-60cm placement length variations because foo dealer mandates 50cm but refuses to explain the technical reasons why x
 
You must log in or register to reply here.


advertisement


Back
Top