Preferred aes/ebu 110 ohm digital cable you have tried

[ar-t]

Maybe to keep this as above-board, as possible, should we consider all interested parties send you the cables? That way, I can measure them cold (or blind, if you prefer), without any bias.

Won't matter to me who meets spec, and who blows smoke. You guys can sort all of that out.

I can see it now:

"Your test method is bogus/BS/disreputable/etc."

So, we need to have an agreement: if you send it, you accept our method. And the result. No kvetching if you bomb. You can crow if you pass. Again, won't matter to us, in either case. Really, it won't.

 

[ar-t]

Because "digital audio cables" don't work at audio frequencies. They work at SPDIF frequencies. Meaning the lowest frequency one will ever see is 1.411 MHz. And on the high end, frequencies approaching 100 MHz, depending on how nice of a square wave you want it to look like.

Which is the whole point I am trying to make. Some, maybe even a lot, but certainly not all, companies who make "digital audio cables" use the same rationale. (I know, because some actually employ us to measure their stuff. And argue when they don't like the results. Still others send us their competitor's crap, to measure. But, I digress.)

And you guys wonder why they sound so different. There is your answer.

 

[jap]

Interesting thread.

Wonder if the SR ANACONDA XLR digital cables measure better than their Cobra XLRs or a $60. Mogami cable.

"The definitive results demonstrate that the ANACONDA DIGITAL offers performance that rivals and exceeds that of any competitive design regardless of price - bringing digital-system playback closer to analog than ever before."

How about it Caelin, are you willing to send some cables to Pat?

 

[BlueFox]

Well, time for me to ruin this thread.................

A lot of the AES/EBU cables that we have measured are NOT 110 ohms. Including some very famous names.

They should try their hand at USB cables.

Over and out.

So, what is the point of all these posts regarding measurements? Are you saying any digital cable that measures 110 ohms sounds better than digital cables that measure 109 or 111 ohms? What is your point? Is the impedance of the cable the only characteristic that determines its sound quality? If so, at what deviation from 110 ohms does it become apparent?

 

[Rufus McDufus]

And is the equipment at either end perfectly calibrated to operate optimally with a "perfect" 110 Ohm cable or are there significant margins of error there too?

 

[ar-t]

Good questions.

First, the cable.

If it is supposed to be 110 ohms, and if it measures 90-95 ohms (yes, a famous brand does measure that far off), it is not likely to sound good. (BTW, it makes a really great USB cable! For obvious reasons.) I even seen SPDIF cables that are 50 ohm and 93 ohm passed off as 75 ohm.

Granted, a few ohms won't kill you. May not sound as good as the correct impedance, on a really good system, but if you want the best, then you should know what has a chance to really be the best. (Isn't that the raison d'etre of this forum?)

Now, the other point, and this is where things get messy.

The gear. Not only is it supposed to be 110 ohms, it is supposed to measure 110 ohms, without a lot of reactive components, over a very wide frequency range. How the impedance of the source, receiver and cable interact is not easy to predict.

Without going into a lot of technobabble, our research indicates all have to be matched within 3% or so, of ideal. Once you reach that level of measured parameters, all the sonic differences tend to go away.

We have encountered many examples of gear where the impedance is off by as much as 20-30%. (Yes, measured on a TDR! And a VNA, as well. Both are valuable, and give slightly different insights as to what is wrong.)

Anyway, short description of why digital cables sound different, and may not yield repeatable results, on different gear. Yes, works the same way on interconnects and speaker cables. Here is one instance where you can reasonably predict which ones can really be the best.

 

[microstrip]

IMHO the debate of exact cable impedance is useless if you do not correlate its effect with sound quality - something I have never seen.

SPDIF and AES/EBU systems have an output impedance (source component) and a receiving input impedance (DAC input). We should remember that most of the time it will not be the exact specified value, and these impedances can show significant variations versus frequency.

For more technical information related to this subject see this link. http://www.avtechpulse.com/appnote/techbrief12/

I have no doubt that cable manufacturers use their cables to improve perceived sound quality - if their target was just passing a pulse without reflections all cables should sound the same. And you can be sure that their marketing departments will be always hiding the real technical truth from us.

BTW, I just got a Transparent Audio XL AES/EBU cable for my system - it sounds great compared to Madrigal MDC1 I was using. Perhaps when I have the time I will I check its impedance using a Philips 1nS rise time balanced pulser, a Tectronix 10 GHz scope and a few calibrated impedances. But just to please my curiosity.

 

[Orb]

And then there is also the receiver chips and associated circuitry-design/transformer coupled (if)/etc.

Cheers
Orb

 

[dallasjustice]

Less jitter is always more desirable than more jitter.

 

[microstrip]

Less jitter is always more desirable than more jitter.

It is no as simple as that - jitter has a spectra. The question is always how you value the different components of complex thinks. It is why simple minds always prefer to say thinks are inaudible!

 

[dallasjustice]

Fewer digital cable reflections is always better than more digital cable reflections.

 

[ar-t]

IMHO the debate of exact cable impedance is useless if you do not correlate its effect with sound quality - something I have never seen.

So, you have never seen it, and that means it doesn't exist. Oh, now I get it. Maybe you should look for it. You might be pleasantly surprised.

SPDIF and AES/EBU systems have an output impedance (source component) and a receiving input impedance (DAC input). We should remember that most of the time it will not be the exact specified value, and these impedances can show significant variations versus frequency.

Exactly. But that means cable designers should just about, willy-nilly, and design whatever maximizes profits? How else will they pay for their fancy cars and expensive watches?

I have no doubt that cable manufacturers use their cables to improve perceived sound quality - if their target was just passing a pulse without reflections all cables should sound the same.

Can you think of a better starting point? Other than how much they can sell cables for. I mean, if it is just a matter of getting the geometry right, would you want to just accept whatever impedance it comes out to be?

"Uh, yeah, New England Wire Co.? Yeah, we want you guys to make us an AES/EBU cable, but here is how we want you to do it..............(snip)...............because we think that would be cool. Impedance? Well, I dunno. I guess it would be best if it was 110 ohms, but if it comes out to be 95 ohms, or 128 ohms, I guess that would be ok. Still have only a 5000' minimum? Cool! When can we get a reel?"

Wouldn't "the best" cable be the one that sounds good, on all systems? Or would you prefer one that sounds great on Brand X, but not so good on Brand Z?

Remind me to tell you guys the story of the company (who shall remain nameless, even though they are long since gone) that made the 93 ohm SPDIF cable. (Hint: it was because some "designer" thought that 75+28=75. Only if 28=0, and last time that I looked, it didn't.) See if you guys can connect the dots, and figure out why they came up with, and promoted, a 93 ohm coax.

What a radical idea................you can actually make a cable that sounds the same, on just about any piece of gear. How are we going to pay for our fancy cars and expensive watches, if it is only that simple? Marketing, man, that is where it is at. Coming up with creative marketing, to cover up the fact we don't care about impedance, or pulse dispersion. Or maybe that we don't know as much as we pretend we do. No, can't have that. None of this "making a better product by using real engineering" crap.

And you can be sure that their marketing departments will be always hiding the real technical truth from us.

And the fact they don't brag about how tightly they control the impedance speaks volumes.

BTW, I just got a Transparent Audio XL AES/EBU cable for my system - it sounds great compared to Madrigal MDC1 I was using. Perhaps when I have the time I will I check its impedance using a Philips 1nS rise time balanced pulser, a Tectronix 10 GHz scope and a few calibrated impedances. But just to please my curiosity.

Only 1 nSec? C'mon man, we have a 40-year old TDR with 150 pSec. (Just joking.) Seriously, now that you have heard them, you have the chance to see if there is any correlation between how they sound and how they measure.

Now that your engineer curiosity is piqued, should be tonight when you get around to it, right?

 

[ar-t]

And then there is also the receiver chips and associated circuitry-design/transformer coupled (if)/etc.

Cheers
Orb

Yes! Yes! Yes!

Man wins a cigar. Thank you.

 

[ar-t]

Fewer digital cable reflections is always better than more digital cable reflections.

Is that something like "Bleeding head, good; healed head, bad"?

No, probably best I don't even try to explain that one.

 

[audioarcher]

ar-t, this a bit off topic but what are your thoughts on what makes an "ideal" usb cable?

 

[ar-t]

Another good question.

One of the most baffling things I have run up against. Wish I knew, but nothing we came up with seemed to stick. And we tried. Just to satisfy our engineering curiosity.

Part of it might be the interface is not terminated, on the RX end. They know there will be reflections, and take that into account, in the specs. Which is where the 26 nSec round trip limit comes from. The idea is the reflections will die down, far enough, that they will not interfere with the next pulse.

If you are going to make an interface that is unterminated, at least they knew enough to come up with a way to handle the reflections.

This one sounds good short...............this other one sounds good long. But if you stick it on that one over there, the shorter one now sounds good. That other one always sounds better long. And this one here sounds about the same, short or long. Which is to say it sounds ok, but not spectacular. Etc.

 

[microstrip]

So, you have never seen it, and that means it doesn't exist. Oh, now I get it. Maybe you should look for it. You might be pleasantly surprised.



Exactly. But that means cable designers should just about, willy-nilly, and design whatever maximizes profits? How else will they pay for their fancy cars and expensive watches?



Can you think of a better starting point? Other than how much they can sell cables for. I mean, if it is just a matter of getting the geometry right, would you want to just accept whatever impedance it comes out to be?

"Uh, yeah, New England Wire Co.? Yeah, we want you guys to make us an AES/EBU cable, but here is how we want you to do it..............(snip)...............because we think that would be cool. Impedance? Well, I dunno. I guess it would be best if it was 110 ohms, but if it comes out to be 95 ohms, or 128 ohms, I guess that would be ok. Still have only a 5000' minimum? Cool! When can we get a reel?"

Wouldn't "the best" cable be the one that sounds good, on all systems? Or would you prefer one that sounds great on Brand X, but not so good on Brand Z?

Remind me to tell you guys the story of the company (who shall remain nameless, even though they are long since gone) that made the 93 ohm SPDIF cable. (Hint: it was because some "designer" thought that 75+28=75. Only if 28=0, and last time that I looked, it didn't.) See if you guys can connect the dots, and figure out why they came up with, and promoted, a 93 ohm coax.

What a radical idea................you can actually make a cable that sounds the same, on just about any piece of gear. How are we going to pay for our fancy cars and expensive watches, if it is only that simple? Marketing, man, that is where it is at. Coming up with creative marketing, to cover up the fact we don't care about impedance, or pulse dispersion. Or maybe that we don't know as much as we pretend we do. No, can't have that. None of this "making a better product by using real engineering" crap.



And the fact they don't brag about how tightly they control the impedance speaks volumes.



Only 1 nSec? C'mon man, we have a 40-year old TDR with 150 pSec. (Just joking.) Seriously, now that you have heard them, you have the chance to see if there is any correlation between how they sound and how they measure.

Now that your engineer curiosity is piqued, should be tonight when you get around to it, right?

Thanks for time you took replying - unfortunately your answer does not add nothing valuable to my post. I would expect to get some links and references of studies, and I get lots of humor, and a suggestion I should look for them. I must say that now I do not feel motivated to measure anything ...

 

[ar-t]

I am not an academic. Peer-reviewed studies are not the sort of thing we place any faith in. I worked in a research lab, and it left me jaded. So much for peer-reviewed studies.

We are not going to disclose the research we do. Other companies actually pay us to do that sort of thing, and we are not about to disclose any of that. Period.

Sorry the humor offends you. And that is your reason why you are going to turn a blind eye to it? Oy vey.

So much for the quest for what is best.

 

[dallasjustice]

Just got back from Pat's house. Using his TDR and another method to verify, we measured a $40 AES/EBU cable. We next measured a "patented" $500 AES/EBU cable. We verified each measurement. The $40 cable measured 120 ohms (pretty good) and the "patented" $500 AES/EBU cable measured 42 ohms (the joke's on you, customer!)

Both manufacturers claim 110 ohm spec. We are going to upload some videos in the near future. We have several more we are going to measure. No manufacturer names will be disclosed. We will only disclose retail price for each cable.

Obviously, we are off to an excellent start!

 

[garylkoh]

Just got back from Pat's house. Using his TDR and another method to verify, we measured a $40 AES/EBU cable. We next measured a "patented" $500 AES/EBU cable. We verified each measurement. The $40 cable measured 120 ohms (pretty good) and the "patented" $500 AES/EBU cable measured 42 ohms (the joke's on you, customer!)

Both manufacturers claim 110 ohm spec. We are going to upload some videos in the near future. We have several more we are going to measure. No manufacturer names will be disclosed. We will only disclose retail price for each cable.

Obviously, we are off to an excellent start!

That's very interesting - what was the qualitative difference in sound between the 120 ohm cable and the 42 ohm cable? It would be great to correlate the sound of digital AES/EBU cables to measured impedance. If you can find another manufacturer's 42 ohm (claimed 110 ohm) cable, and it sounds the same as this particular 42 ohm cable, then we have datapoints we can use.