Ethernet clocking myth or fact

[Yelamanchili manohar]

Hi all

This subject was touched upon in a different thread, so thought better to start a new thread without derailing the original.

There are ether regen users swearing by the improvements brough by the addition of an external master clock. However, on Dcs and ASR ( yeah i know ) there have been no reported improvements by this addition.

Thing to bear in mind, is that almost all DCS users are using network feeds for the files and not USB. And from my own reading, there seems to be no clock signal at all in ethernet transmission. It is only data being transmitted, which has to be buffered and then converted to a digital stream. Only once the digital bits are assembled this way do we have any notion of a time signal coming into play. So how come the ether regen is benefitting from clocking.

This got me thinking, and Iam wondering if the users reporting improvements with clocking are using USB inputs and not network inputs by any chance.

In an audiophilestyle site comparison between the clocked ether regen, clocked SOTM SNH-10G and a non clocked Melco S100. The melco came out on top. The ether regen with the addition of an external clock came really close. But the total price with the clock was more than the melco. This got me even more confused.

Any feed back or opinions on this will be much appreciated in deciding my next course of action, as I plan on investing in upgrading my network this summer. So should I go for a switch with clock or not, as there are not many around with clock inputs. Besides, I dont even have a clock at the moment, so that will be another spend on top. Thanks all

 

[amit11]

On this topic, it is my understanding that, it is not the clock timing that affects the output. But it is rather the induced noise of the electrical signal which affects the output.

Yes ethernet connection is galvanicaly isolated, but even galvanic isolation is not 100% immune to noise from the incoming electrical signal, which we call as data.

 

[Yelamanchili manohar]

On this topic, it is my understanding that, it is not the clock timing that affects the output. But it is rather the induced noise of the electrical signal which affects the output.

Yes ethernet connection is galvanicaly isolated, but even galvanic isolation is not 100% immune to noise from the incoming electrical signal, which we call as data.

Thanks Amit

I have'nt read every written word on the subject yet. But my reading so far seems to point to some residual noise and jitter left due to the optical conversion. So there seems to be a trend to use these so called "audiophile switches" even after fiber to clean up the residual noise a bit.

Though Iam still confused about the clocking part. Depending on who you listen to, there are equal advocates on either side of the fence so far

 

[drkrack]

It Makes a Difference, especially if your setup is Decent midfi with predominant use of streaming as source.
Many of us have been enlightened, not groundbreaking difference, but can't go back once you remove the clock.
Bringing to the attention of @IndianEars, @ryanrulez4ever @sajjad4joy, can share their experiences

 

[mustafakk]

Hi all

This subject was touched upon in a different thread, so thought better to start a new thread without derailing the original.

There are ether regen users swearing by the improvements brough by the addition of an external master clock. However, on Dcs and ASR ( yeah i know ) there have been no reported improvements by this addition.

Thing to bear in mind, is that almost all DCS users are using network feeds for the files and not USB. And from my own reading, there seems to be no clock signal at all in ethernet transmission. It is only data being transmitted, which has to be buffered and then converted to a digital stream. Only once the digital bits are assembled this way do we have any notion of a time signal coming into play. So how come the ether regen is benefitting from clocking.

This got me thinking, and Iam wondering if the users reporting improvements with clocking are using USB inputs and not network inputs by any chance.

In an audiophilestyle site comparison between the clocked ether regen, clocked SOTM SNH-10G and a non clocked Melco S100. The melco came out on top. The ether regen with the addition of an external clock came really close. But the total price with the clock was more than the melco. This got me even more confused.

Any feed back or opinions on this will be much appreciated in deciding my next course of action, as I plan on investing in upgrading my network this summer. So should I go for a switch with clock or not, as there are not many around with clock inputs. Besides, I dont even have a clock at the moment, so that will be another spend on top. Thanks all

Hi

If you are using an ether regen the after dark clock makes a marked improvement in the chain.

How I am now using the Melco S10 switch and this is definitely a step up.

 

[Yelamanchili manohar]

Hi

If you are using an ether regen the after dark clock makes a marked improvement in the chain.

How I am now using the Melco S10 switch and this is definitely a step up.

Thanks for sharing the above

The S10 is way beyond what i can stretch to.........but the S100/2 might be doable

Having said the above, Iam planning on trying the cheaper Gustard N18 pro to see if it delivers beyond its price point. Iam sure the chinese have had enough time to procure all the well performing switches to reverse engineer them on the cheap by now. Just hoping atleast they copied well

 

[mustafakk]

Thanks for sharing the above

The S10 is way beyond what i can stretch to.........but the S100/2 might be doable

Having said the above, Iam planning on trying the cheaper Gustard N18 pro to see if it delivers beyond its price point. Iam sure the chinese have had enough time to procure all the well performing switches to reverse engineer them on the cheap by now. Just hoping atleast they copied well

A better option would be to use a decent switch with an excellent power supply will be a great option.

The way the packets are arranged and the software that controls it is what makes the difference.

I have not tried any Chinese switches.

Network Acoustics has a great switch.

 

[Yeraditya]

My Source is Aurender N20 with MSB DAC. So you can call me a “snake oil” peddler. But I have personally heard quite a few switches/Routers and from an audibility point of view I have heard ‘larger and more fixed sound-stage’ and consequent separation of instruments. I finally went with Ansuz D3 .

But to be honest, I would not recommend it for many. Its not the world. The logic given is that it insulates/protects/removes noise etc. and the reason I say its not the world because, its not all that pleasurable to hear so much detail and its very subtle. I found adding a good LPS to a non smsp router and wiring your source/roon core directly does pretty much the same for a fraction of the cost of switch/cables/and power for switch etc.

 

[Prateek-Epitome]

can anyone with experience of Gustard R26 and using clocks or DDCs share their experience of improvement in sound?

 

[MIOM]

can anyone with experience of Gustard R26 and using clocks or DDCs share their experience of improvement in sound?

I'm not going into the deep dive how's and why's but I will say DDCs definitely work.

I bought a Denafrips Iris DDC when I bought my Denafrips Pontus II DAC, but also hooking it up my Teac UD-503 DAC did wonders for the Teac. It sounds much richer and cleaner. Bass is tighter, there's more detail presence, as well as improved soundstage. Music out of the Teac simply sounds better after connecting it to the Iris.

At the end of the day many people will say many things, but the best test is to hear for yourself and go from there. I'd also suggest this video to get an idea what a DDC is about. This is a review of the Denafrips Iris, but you'll have a better understanding of DDCs in general...

Good luck.