AudioQuest Diamond "directional" CAT-5 ethernet cable. 3 meters for $3500.

 

This is real. It exists. I had to look it up.
Quote from the site: "The sound opens up like a flower to the sun."

Edit: I tried to find out where the bs price comes from. I still don't know, but I have a brilliant idea for a small business since there are apparently people on this planet willing to pay that kind of money for a cable.

 

Last edited by a moderator: Apr 21, 2024

then the song is played through phone speakers in mp3 128kbps format

 

Bargain... My mixing will be so much better after installing those...

 

First and foremost, the price is due to the fact that silver costs 95 times as much as copper. Silver is the better conductor compared to copper because silver has a lower resistance.
Another factor could be the structure of the conductor. We are talking here about microscopic differences in the conductor core and on its surface, which can have a measurable influence on signal transmission. Furthermore, cable production and the size of the consumer market have an influence on the price.

Personally, I think the price is a little too high. Something between half and a third of the advertised price would probably be more appropriate for what it is.

Do high quality silver cables have an audible advantage in audio applications over high quality copper cables? Probably not. And if I had the choice between silver and copper for a digital transmission signal, I would choose fiber. For analog transmission - assuming money is no object - I would go with silver because it is the better conductor. But I'm just a poor audio engineer.

 

Last edited: Apr 21, 2024

Really? :D Silver sure is expensive, but I still smell bullshit and an extra premium margin. The AudioQuest guys maybe just want a fresh pair of lambos.
But maybe there is a real world application for that type of cable. Someone out there buys them.

Edit: I saw a video by AudioQuest. It's for "audiophiles" with too much money. The difference according to AQ is better isolation. Technically that is probably true. Maybe I should start a business like that and charge 10k a meter , because my cables have an extra layer of isolation which is clearly superior to AQ.

 

Last edited by a moderator: Apr 21, 2024

Nope. Not even close. Digital versions of analog wave forms are ALWAYS stair-stepped (square-wave) facsimiles of the analog input. In fact given what we know about square wave synthesis there are analog harmonics generated each time a square wave transitions from "on" to "off" and vice versa. So not only do you get a stair-stepped digital facsimile of an analog signal you get all the harmonics that are generated due to the square wave transitions. Of course there are methods (most commonly dealt with via low-pass filters) to deal with the harmonics but in essence the D-A (or A-D) conversion process is not as cut and dry as one would be led to believe. Just because you don't see it on the scope at the same resolution as its analog counter part doesn't mean the steps aren't there... Kick up the trigger on the scope and zoom in on any part of that digital signal and you'll see the stepping for real...

 

Last edited: Apr 21, 2024

Sine waves are the simplest and easiest to predict. Their reconstruction is straightforward. Therefore, they cannot serve as a reference for estimating the range of reconstruction for complex waves.

 

Complex waves are made up of... sine waves. All sound is made up of ... sine waves. It's like trying to explain that a circle is indeed round and not 'made up of a finite number of edges'.

 

Examples?
I posted video earlier.
Or just resample sine to way higher sample rates. Create a 21k sine at 44.1 and resample to 192k. And you will see it becomes way rounder from its original ugly view. In analog it is round always.

Squares should be squares as well. Or what? O_o

 

Simply joining the points would result in "polygonal approximation" It involves approximating the curve by connecting consecutive points along the curve with straight line segments, creating a polygonal path that closely resembles the original curve. It results in a lower fractal dimension compared to the original curve. So the line segments smooth out the curve's irregularities, resulting in a simpler geometric structure (= loss of detail + artifacts).
Of course, an analog current source produces an analog output, and numbers don't simply fly through the air. What's at stake here is the fidelity of the reproduced physical phenomenon to the original — in other words, the degree of distortion. When attempting to recreate a disproportionately larger number of physical elements with a limited set of data, we venture into interpolation, essentially guessing and masking gaps in the data. It's important to note that an ideal reconstruction in DACs is unattainable, leaving sine waves to the realm of energetics, the study of AC currents, and electric power transfer.

 

When you send a sine wave through a non-linear process, harmonics (sine waves) are added to the fundamental. The result is a complex wave. However, this complex wave consists of the sine waves f0+f1+f2+f3+f4+...
As long as the frequency of these sine waves is not higher than Nyquist, the waveform can be constructed correctly, if we ignore filter distortion for the moment.

 

Last edited: Apr 21, 2024

I get that this disscussion is analogue vs Digital but this one thing is getting to me.

Who in their right mind still uses CAT5 data cables? CAT6 became standard internationally nearly 5 years ago and if your worried about signal loss you go CAT7 with shealding.

What makes it worse their website although offering CAT7, when you look into the Cables. These cables are only shielded on the outside.
it's a F/UTP cables, as in foil sheald round unshealded twisted pairs of cable. I know of office blocks with better cables for general I.T. work.

You want good signal then forget burn in, just go CAT7 F/FTP. As in the thickets copper cable you can get with foil shielding round the outside and seperate foil shields on each twisted pair of cables inside.

Sorry it's a little off topic but $3500 for 3m? That's too much.

Also speakers smooth everything out. Cheaper the speaker the more sine wave your square is.

 

really? i know sine wave is the most basic one, but isn't something linear like a triangle wave easier and simpler to predict?

linear regressions are so much easier to calculate than curved regressions

 

I cannot reproduce what you have described. Either my mid-range dac is even better than its reputation, or my oscilloscope is not high enough resolution (in which case I'm not worried about the audibility of those stair-steps), or you have a measurement from some nos-dac.

 

Take Vital or w/e synth you have that lets you look at and draw in harmonics and inspect the following waveforms' harmonics: pure sine, pure square, pure sawtooth, pure triangle.

Then look up what harmonics are. Fuck it, I'll save you the Google. Here's the textbook definition:

- Complex sounds are sounds that contain two or more sine waves.
- The harmonic series is a mathematical combination of sine waves that are perceived by our ears as fusing together into one pitched sound.

Please take some time to understand the implications of above, particularly with regards to the shapes of the synth waveforms.

Now do this. Start with a pure sine wave in your synth. Pencil in harmonics of the ~same amplitude. Do this up to the highest harmonic your synth allows. Pay close attention as you draw to how the preview images changes.

Take some time to understand the implications, especially with regards to the shapes and ripples you're observing when you look at a sawtooth, square, etc.

The principles are so radically different from what you imagine that I'm genuinely unsure if I can help you understand better at all.

There is none; you must fundamentally understand that there is and can be only one solution to reconstructing a bandlimited wave. And that solution is always, without exception, perfect. Distortion is produced not by the principles, the math or the formulas, it is produced by your gear. And that gear, unless broken, will almost universally produce less distortion than any analog equipment has ever been capable of.

See, that's crucial to understand here. The signal is bandlimited (to 24kHz or w/e).

Thus: There. are. no. gaps. There. can. be. no. gaps. The reconstruction isn't just kind of perfect, it is perfect-perfect. Please take some time again to try to understand this.

This is so fundamental and universal that you simply need to accept it, even if it makes no sense to you.

 

$3500? Pfft. Dirt cheap compared to Nordost Odin 2 Speaker Cable at an eye watering £37,350 (approx $46,000) .

https://www.futureshop.co.uk/nordost-odin-2-speaker-cable-factory-terminated-pair?gad_source=1

And that's for ONE metre. 4m will set you back £70,470. Plus £3 for DeOxit enhancer

 

For number one if you're looking at the digital sample @ 44khz it doesn't look "ugly". Dunno where you're getting your info... For number two, the DIFFERENCE is seen by looking CLOSELY (at sample resolution) at the 44khz signal and the 192khz signal. You will see more steps (aka samples) per unit of time on the 192khz signal. As far as sound quality is concerned - the ability for the digital signal to mimic its analog input, the 192khz signal has a better chance at rendering a more accurate representation of that analog input due to the fact it is capable of resolving better due to the higher sampling rate. You still have the same kinds harmonics generated by BOTH digital formats however. That's just the nature of the beast. Resampling introduces its own set of anomalies but
Observe a 1Khz sine wave recorded @ 44khz and 192khz

@ 44khz


and the same signal @ 192khz


keep an eye on the time reference as we zoom in on the 192khz wave. The number of samples has greatly increased making a better representation of the sine, BUT it's still a DIGITAL representation as you can see from the STEPS.


finally the same 44khz signal resampled to 192khz and zoomed in. There is no difference in the stepping/sample integrity - i.e. no "rounding" occurring...

 

Last edited: Apr 21, 2024

Spoiler

That's all I can say to that.

 

Just do sinc or b-spline interpolation. Lol

 

Last edited: Apr 22, 2024