Ethernet Cables in for evaluation

You mentioned in your cable shootout that you took time to reboot your devices between cable swaps. Could you explain the reasoning behind that or what the difference would be had you swapped them live?

 

It literally is, tho.

It's like this. Let's say your friend Peter wants to send you a message. Except you and Peter have agreed to a standard (ethernet!) by which Peter has to translate that message into a series of numbers and give it to his friend Alice. Alice then sends those numbers to her friend Bob, who is smart enough to be able to decode the numbers Alice gives him and present Peter's message, in its original form, to you. What a plan!

Except... oh, no! Alice and Bob can only communicate over the phone, and the connection is garbled and bad because the analog phone wire between them is of low quality! Fortunately, the standard everyone has agreed to (ethernet!) has a great solution for this. After Alice reads each sequence of numbers to Bob, she also tells him what the product of that series of numbers is. Bob quickly does the math. If the numbers Alice has read to him don't total the product, he asks her to read the entire sequence again. If she tries again and he still can't make the math work, he comes to you and says: "Sorry. Transmission of Peter's message failed."

Let's try an example. Say Peter wants to transmit to you: "CAT EATS FOOD". Simple enough! But you've all agreed on a standard, by which (let's say) each capital letter is encoded into an ordinal number on the scale of 1-26. So, when Alice reads that message to Bob, she translates it as:

3, 1, 20 (60)
5, 1, 20, 19 (1900)
6, 15, 15, 4 (5400)

Again, though, as we all know, phone wires are analog and of dubious quality. What happens if some noise or jitter or other badness happens when Alice is saying "three" in the first word - and Bob either mis-hears it or doesn't hear it at all? Well... he'll quickly notice that the product of the numbers in her first sequence does not equal sixty. And he'll ask her to repeat. The second time, he'll either hear "three" correctly and ask her to move on to the next word. Or, tragically, he'll conclude that the analog phone connection is not of suitable quality to transmit the message, and he'll tell you it failed.

Either way, you, as the recipient of the message, are guaranteed to get one of two results: "CAT EATS FOOD", or an error message indicating that the transmission failed.

Nothing in between.

That is how ethernet works, and is the basis of literally all modern communication, from the datacenters owned by Amazon and Microsoft that buy literally thousands of miles of ethernet cable every year (and commodity cable at that!), down to your local network connection between your streamer and DAC.

Ethernet is either bit-perfect or it doesn't work at all.

But again, as I said months ago earlier in this thread: if say you can hear a difference, you can. Mazel tov!

It just has nothing to do with ethernet.

 

@gkg2k perceptions are so easily influenced by our cognitive biases- the placebo effect and confirmation bias are real things! This is why blind testing is always a good idea to confirm what our ears are telling us when trying to make distinctions between small differences in audio reproduction.

These 2 statements can not be simultaneously true in reference to Ethernet. Either it's bit perfect or it isn't.

 

Yes, most probably I could swap them live.

Anyway I preferred to "reset" the two devices (server and renderer) at each swap. Was it useless? Maybe!

 

You didn't read my post carefully. It's not a matter of bit perfection, the matter is the Ethernet cable is a cable.

Gianluca

 

I see you don't even read my posts.

I certainly won't measure the signal being bit-perfect after the Ethernet link (it was even when live-reading a scratched CD), but I'll try to testimony that different-quality/different-construction/different-geometry/different-material/different-shielding/... cables provide different analog audio waves out of my speakers.

Gianluca

 

Good points and i agree,

 

I'll need such high quality microphones and complementary electronics that I don't know if I'll be able to afford it!

 

Your claims are nothing but contradictions. Here are some of your claims after waiting 2 minutes between listens:
"strong differences"
"palpable differences"
"rock solid"
"3D"
"you can see the sound"
"definitely hear marked differences"

You keep saying that the differences are large and obvious, but now now you follow it up stating that the differences are so slight that you need prohibitively expensive equipment to be able to measure it? Your entire stance is a comedy of really entry level errors.

 

Dear Subzro,

you domn't impress me at all with your posts. Actually you produce some of the less impressive posts on the entire forum scene, imho.

My question is : is it easy to buy a couple of microphones that are sensitive enough to capture all the details my/our hearing can? A microphone amplifier is also needed and an analog to digital converter with USB output to connect the microphones to a computer for recording. Am I a sound engineer? I am not.

The differences I hear are SO CLEAR (sorry to repeat it, I know it's an eyesore for you) that I'm confident I could succeed. Give me time and let me find the funds for the equipment.

Gianluca

 

It's very strange you don't get the point : the limitations of the recording chain and technique can easily hide/cancel/modify the truth of the sound coming out of my speakers.

Or do you think I should do that with my phone?

Gianluca

 

gkg2k, your repeated insistence that you do actually hear the differences between ethernet cables can only be established as fact in a controlled scientific context, where everyone involved agrees to a set of observational procedures, best suited to allow your perceptions to be counted and catalogued. A bunch of people saying, no, you can't, or that what you hear is merely a psychological response, not repeatable under testing, doesn't constitute a fair evaluation of your own experience. How can it?

You are kinda stuck! Barking back at forum members is no way forward. We need to get you in the lab.......

 

My mom will be sorely disappointed that after all my years of hard work that I am not more impressive on the entire forum scene. I had just told her how well things were going with the scene, and then I get this news! I realize now you are right, and that I am just not getting the point. Let us know when you get your equipment together. I'm eager to hear your findings and hopefully learn new things about ethernet in the process.

 

OK, I can be poked but not to poke.

I'll get in the lab asap...

Gianluca

 

not ETHERNET cables though!
they either work or they don't.
i feel like i might have said this already though heh.
@gkg2k ethernet is fundamentally different. I get why it's hard to believe but it is true. if it wasn't computer networks literally WOULD NOT WORK.

 

ewwww the latency, it burns! i use thunderbolt, apogee element 88.

 

really to prove that it's the ethernet cable making the difference you don't need anything but wireshark.

show that the values in the packets have been modified in transit (and haven't been re-transmitted with the original values intact) and you're a winner

 

You all convinced me Ethernet is marvelously bit-perfect and its CRC system is second to none and doesn't introduce jitter (latter argument to be verified), but I consider it as metal cable between a component to another in an audio system. And being such it interacts with them and the whole system. That's my main point.

Gianluca

 

sure, but the ethernet cable itself can not affect the quality of the sound coming out of your system unless it's defective. and that would cause dropouts, not changes in sound quality.
it can't interact with the rest of the audio system because those components (either hardware or software) interact with the data at much higher layers of the OSI model than the cable. once the data has been handed off to those layers, it's already been checksummed and any packets with modified data have been discarded and retransmitted before your audio software even receives the data from the buffer! if the software can't read data from a buffer properly that's not the cable's fault.

 

The problem of not reading one's posts is a big problem here

I'm not claiming the bits transferred from point A to point B with an Ethernet interface are wrong or different than the origin. Even reading a scratched Audio CD on-the-fly leads to a bit-perfect result, so how could Ethernet fail a much easier task?

But that's just ONE huge easy and clear-to-anyone aspect of the phenomenon we're analisying.

We (you...) may miss the existence of an infinite number of micro implications that happen when using a cable of any kind to connect a device to another.

Don't you think so? Is it just a matter of bits?

Gianluca

 

Based on how you described the Audioquest cane compared to the CSA even a recording done off your phone should indicate some sort of difference between the two.

 

You carefully studied the OSI model white papers and Ethernet is much better than any other digital audio interface including SPDIF, I2S and USB, you sold me.

But is that enough to claim an Ethernet cable CANNOT affect the sound of your system?

Gianluca

 

No Shawn, a phone can't. Let's try to do it well. I'm sure my hearing is better than my phone's $1 microphone.

P.S. : I can't stop thinking to the $1 value, I think it's much less!

 

None of those other cables you mention USE the OSI model! and i've said repeatedly that I myself have heard differences between spidf cables!
so, yeah, it IS enough.
if it wasn't, networks would not work.
I keep sayin' ethernet is fundamentally different, it's not like I'm making this up, yo.

 

That's where we disagree