Router and Ethernet Switch suggestions

[davide256]

On 4/23/2017 at 8:56 AM, Confused said:

'....Another point to note is that with the set-up mentioned above, you can remove the Ethernet cable feeding the microRendu, and music continues to play for a second or so..."
 

Correct. All audio devices connected to a network have to have a buffer sufficient to avoid playback gaps  due to momentary traffic  congestion or traffic reroutes.

[davide256]

5 hours ago, Al Jones said:

I just spent a couple of weeks conducting a head-to-head shootout between the Aqvox AQ-Switch-8 network switch (€398) and the more recent upgraded version Aqvox AQ-Switch SE (€798). Owners of the original can have it updated to SE by paying the difference. For clarity's sake, I will refer to them in the following as the original and the SE.

 

As always, the usual disclaimers apply. I have no affiliation with Aqvox except as a satisfied customer.

 

My initial impressions of the original can be found here:

 

 

Aqvox tries to describe the improvements introduced in the SE without creating a virtual DIY recipe: newly developed phase correction, new, optimized oscillator for a factor 10 more precise clock generator, additional interference suppression components, carefully optimized internal power with additional Ultra-Fast capacitors, metal housing lined with butyl rubber...

 

All my listening was done in the system described in my March 11 review of the original, with a silver connection from switch to Entreq grounding box and an Akiko RJ-45 Tuning Stick inserted in an unused port on the switch. With more than 8 months experience living with the original, I had no great expectations for what the SE could bring to the party at twice the price, but the short answer to that turned out to be simply "even more of the same."

 

Skeptics can scoff all they want, but these modified switches do bring out more music with substantially more room information and natural ambience, resulting in a sonic image with enhanced dimensionality and solidity.

 

The enhanced background clarity and intertransient silence of the SE provide worthwhile improvements that make music listening so much more effortless and relaxed for hours on end. Apparently, the SE reduces noise and eliminates background distortions and disturbances that the listener probably wasn't even consciously aware of.

 

The improvement is subtle, but consistent across the board, enhancing top-to-bottom coherence, instrumental timbre and the sheer musicality and "physicality" of the listening experience.

 

If you're still running your system on a standard switch, you owe it to yourself to audition an Aqvox. If you're a satisfied user of the original Aqvox and have the discretionary income, look into upgrading to the SE.

 

In my system, the SE is definitely a keeper.

 

I guess I'm trying to understand why you went this more expensive route vs. using fiber optics?

[davide256]

6 hours ago, R1200CL said:

Well, I would think the data stream has already passed the switch, and the “damage” is already done ?

 

As it would through USB as well. Did you pull the usb plugs, usb HD’s ? Power ?

I think you use WiFi. (To avoid the switch ?)  Pull the power of the WiFi, I suppose same happens. 

 

So so how would  USB/SPDIF converts sound different, or maybe they don’t ?

 

Do switches have jitter ? 

Does jitter and other pollutions (power, noice, ) most likely occurs in the digital domain. 

 

Why rule out switches ? And not DCC’s and more...

you are confusing apples and oranges jitter

 

Switch/router  data is encapsulated, has no clock in the frame or packet envelope. The frame/packet arrives  at the destination with no supervision for transmission time delay and can even arrive out of sequence if multiple paths are involved. Network jitter is measured by variation in packet delay... matters for VOIP and other real time applications where we expect a full duplex session at human speeds (push the button?/ yes, now). Recorded audio sent across the network is not a real time application... buffer a few seconds at the receiving end and all should be well even if your streaming source is across the globe with high packet jitter (think driver download from Taiwan). All packets/frames are error checked and if"damaged" retransmitted

 

The jitter that matters is the jitter inside the receiving device after frame/packet envelope is stripped, data sequenced, audio data identified and put into buffer. At this point you are "live",  and any issues  in correctly timing audio output for data  retrieved from the buffer are critical. This is audio jitter vs network jitter

[davide256]

1 hour ago, plissken said:

Dave, this is going to fall on deaf ears most likely. Even though you are correct.

 

Music playback isn't real time, there is no timing data in the packet. GBe speeds can literally deliver a 16/44.1 track in a split second and the player application (Tidal and JRiver can do this) cache the entire track in RAM.

 

Some are behaving like there is some form of stored jitter in a file that took .5 seconds to transfer. Move up to 10GBe and you are now talking about a 0.05 second transfer.

 

This talk about TCXO and other tweaked out switches is pure lunacy.

Agree on no need to improve switch clocks . Just clean up the Ethernet connection so it doesn't inject electrical noise into the audio device. Worry about clocks where the data is in analog form.

[davide256]

1 hour ago, Albrecht said:

Yes,

Very true, but irrelevant to the subject at hand: which is the quality of what is loaded into the buffer.

All data sent across a network has a packet/frame checksum, this allows  the distant end to

verify what was sent was not altered before its loaded into buffer. And if loading into a buffer was subject to data error, the machine you are working on would not work,

there would be no PC's, Mac's or mainframes. Trillions of dollars in transactions ride across networks every day, banks  are not happy with machines that corrupt data in any way..

 

If you want to look for problems with network audio devices look (a) to how well electrically behaved its Ethernet attached neighbor is ( b) how well behaved the device processing audio is (b) at what happens when a & b are interconnected and the data is extracted from the buffer, turned into an analog signal output that has little to no error correction.

 

Ethernet ports are basically "least cost" devices intended for switches, routers and PC's... I trust them not for being built to the spec needed to prevent 2 devices from electrically interacting thru leakage voltages at a level that would cause an audio device to degrade for sound.  Fiber optics are the cheapest way to tackle the transmission source electrically interfering with the receiving system.