Ars prepares to put “audiophile” Ethernet cables to the test in Las Vegas

[Jud]

 

USB is a straight connection to the clocking buffer on a DAC.

 

 

With async USB audio, the receiver calls for data as needed to keep the buffer from emptying (among other functions).

[plissken]

It appears that you don't know how USB audio works in modern DACs that use asynchronous USB operation or equivalent.

 

What I posted has nothing to do with Async or Synchronous USB.

 

What I'm telling you is a USB DAC doesn't know what mechanism filled the buffer that it is fetching out of.

 

Not Ethernet not a USB bridge between computers or local HD.

 

Your the one that just publicly stated that there are shades of grey to what constituted real time.

[plissken]

With async USB audio, the receiver calls for data as needed to keep the buffer from emptying (among other functions).

 

Again USB is a straight connection to the clock on the DAC. Ethernet is not.

 

USB DAC's aren't sensitive to the timing variance of Ethernet. Don't ask me.

 

Go over to What's Best Forum and ask AmirM since he is the head engineer at Microsoft that spearheaded the change from KMixer to WASAPI.

 

I'm not the one that keeps trying to take USB packet jitter and hang it around the neck of Ethernet. They are two different mechanisms.

 

They even linked to Archimago's blog as some sort of example. Which is patently dumb. I thought it would be better to link to Archimago's intercontinental streaming test.

 

"Of interest and importance is that distortion results (THD, IMD) were essentially the same and certainly no worse than having a server in the next room. Again, these results are for a 24/96 "high resolution" audio test."

 

And:

 

"(BTW - ever wonder why some audiophiles and reviewers obsess over a few feet of digital cable - especially blaming jitter despite really no evidence to show that it makes a significant difference? Yet these days, with the advent of digital streaming and advertising dollars from places like Tidal, nobody talks about the "potential" for jitter when data is being streamed from miles and miles away? Instead there's just isolated talk about ethernet cables this and that to sell expensive stuff yet no consideration for the truly big picture! Obviously there's something wrong with this whole cable "industry".)"

 

Now I will just wait while the people insisting that Archimago's packet USB measurements hold over for Ethernet now bash the same source for actually going apples to apples.

 

So we have very low THD and IMD #'s. So where is all that transceiver noise I keep seeing people spout off about?

 

Where is that 60Hz noise?

 

Where is the transformer leakage?

[Tony Lauck]

Again USB is a straight connection to the clock on the DAC. Ethernet is not.

 

USB DAC's aren't sensitive to the timing variance of Ethernet. Don't ask me.

 

Go over to What's Best Forum and ask AmirM since he is the head engineer at Microsoft that spearheaded the change from KMixer to WASAPI.

 

I'm not the one that keeps trying to take USB packet jitter and hang it around the neck of Ethernet. They are two different mechanisms.

 

They even linked to Archimago's blog as some sort of example. Which is patently dumb. I thought it would be better to link to Archimago's intercontinental streaming test.

 

"Of interest and importance is that distortion results (THD, IMD) were essentially the same and certainly no worse than having a server in the next room. Again, these results are for a 24/96 "high resolution" audio test."

 

And:

 

"(BTW - ever wonder why some audiophiles and reviewers obsess over a few feet of digital cable - especially blaming jitter despite really no evidence to show that it makes a significant difference? Yet these days, with the advent of digital streaming and advertising dollars from places like Tidal, nobody talks about the "potential" for jitter when data is being streamed from miles and miles away? Instead there's just isolated talk about ethernet cables this and that to sell expensive stuff yet no consideration for the truly big picture! Obviously there's something wrong with this whole cable "industry".)"

 

Now I will just wait while the people insisting that Archimago's packet USB measurements hold over for Ethernet now bash the same source for actually going apples to apples.

 

So we have very low THD and IMD #'s. So where is all that transceiver noise I keep seeing people spout off about?

 

Where is that 60Hz noise?

 

Where is the transformer leakage?

 

Goodbye.

 

Here's the deal. Real simple. I rank people on two axis: The first is their competence at what they do, say, or claim to do, i.e. their usefulness to a task at hand. The second is how easy they are to get along with, how enjoyable one's interactions are, etc.. The group of computer networking researchers that I managed for 20 years had more than a few brilliant prima donnas who contributed valuable knowledge while being difficult to manage. It contained a few people who weren't so skilled, but were easy to work with, pleasant to be around, and who could be counted on to deliver what they promised. It contained no people who were neither technical giants nor pleasant company.

 

In your case I am limited to what you post and the way you post it. My comments are not directed at you personally, but at the persona I imagine who is behind your posts. "The medium is the message." "On the Internet, nobody knows you are a dog."

[plissken]

Goodbye.

 

Here's the deal. Real simple. I rank people on two axis: The first is their competence at what they do, say, or claim to do, i.e. their usefulness to a task at hand. The second is how easy they are to get along with, how enjoyable one's interactions are, etc.. The group of computer networking researchers that I managed for 20 years had more than a few brilliant prima donnas who contributed valuable knowledge while being difficult to manage. It contained a few people who weren't so skilled, but were easy to work with, pleasant to be around, and who could be counted on to deliver what they promised. It contained no people who were neither technical giants nor pleasant company.

 

In your case I am limited to what you post and the way you post it. My comments are not directed at you personally, but at the persona I imagine who is behind your posts. "The medium is the message." "On the Internet, nobody knows you are a dog."

 

That's ok. I've worked with bosses in the past who were clueless.

 

Any data you want to bring to this party is welcome anytime. I don't have any obligation to extend courtesy to wanton ignorance.

 

Speaking to delivering what they promised you let me know when you have something in regards to Ethernet packet noise and measurement of some analog outs of a DAC. Either scope or measurement mic.

 

That's where you stand right now. I'm glad conjecture gets you paid in your day job. It certainly doesn't make it around my place.

[mayhem13]

That's ok. I've worked with bosses in the past who were clueless.

 

Any data you want to bring to this party is welcome anytime. I don't have any obligation to extend courtesy to wanton ignorance.

 

Speaking to delivering what they promised you let me know when you have something in regards to Ethernet packet noise and measurement of some analog outs of a DAC. Either scope or measurement mic.

 

That's where you stand right now. I'm glad conjecture gets you paid in your day job. It certainly doesn't make it around my place.

 

Ignorance is quite uncalled for given the context, don't ya think?

 

What Tony is saying is that there's a rational and reasonable way to debate these topics if you possess the neccessary understanding of the mechanisms at hand. Tony wants you to accept that his conjecture and speculations are 'possible'. .....of course anything is possible. In turn, you can ask him for evidence to support proboble other than the mass hysteria displayed by audiophiles here and just about everywhere they roost.

 

But if you really wanna be a lamb, just say yeah, it's possible!......if enough do the same in turn, SandyK can reference you in every post going further as an example of the countless audiophiles who experience unexplainable phenomena.

[sandyk]

Ignorance is quite uncalled for given the context, don't ya think?

 

What Tony is saying is that there's a rational and reasonable way to debate these topics if you possess the neccessary understanding of the mechanisms at hand. Tony wants you to accept that his conjecture and speculations are 'possible'. .....of course anything is possible. In turn, you can ask him for evidence to support proboble other than the mass hysteria displayed by audiophiles here and just about everywhere they roost.

 

But if you really wanna be a lamb, just say yeah, it's possible!......if enough do the same in turn, SandyK can reference you in every post going further as an example of the countless audiophiles who experience unexplainable phenomena.

 

They may be definitively unexplainable currently, but they are damn easy to demonstrate to anyone who hasn't a severe case of Expectation Bias, and doesn't use additional software to destroy worthwhile subtle differences that the high resolution formats give to those who haven't had their hearing destroyed by too close a proximity to commercial sound reinforcement speakers at a few too many hearing destroying, overloud concerts.

As usual, your views in general are at odds with the vast majority of members who appreciate the high resolution LPCM and Double DSD formats. I would suggest that a few members who appreciate their new DSD recordings and the improved recent gear to play this new format , try using additional Digital Room EQ and see if their DSD still sounds markedly better than a decent RB CD recording.

[kumakuma]

They may be definitively unexplainable currently, but they are damn easy to demonstrate to anyone who hasn't a severe case of Expectation Bias, and doesn't use additional software to destroy worthwhile subtle differences that the high resolution formats give to those who haven't had their hearing destroyed by too close a proximity to commercial sound reinforcement speakers at a few too many hearing destroying, overloud concerts.

 

You forgot to insult his DAC, dog, and mother.

[sandyk]

You forgot to insult his DAC, dog, and mother.

 

I will consider remedying that next time he feels like needlessly mentioning me in a thread, just to take yet another potshot at me, and the same applies to you too. (grin)

[jriver]

My comments are not directed at you personally, but at the persona I imagine who is behind your posts. "The medium is the message." "On the Internet, nobody knows you are a dog."

I remember that cartoon, but I thought your comments were very personal. Plissken has the same right to an opinion as others here.

[jabbr]

That's ok. I've worked with bosses in the past who were clueless.

 

Any data you want to bring to this party is welcome anytime. I don't have any obligation to extend courtesy to wanton ignorance.

 

Speaking to delivering what they promised you let me know when you have something in regards to Ethernet packet noise and measurement of some analog outs of a DAC. Either scope or measurement mic.

 

That's where you stand right now. I'm glad conjecture gets you paid in your day job. It certainly doesn't make it around my place.

 

I appreciate your desire to be precise and to avoid conjecture, but where is it written that Ethernet PHY is noise free? If you have measured PHY under various circumstances, then what is the limit of your ability to measure? We generally think in SNR rather than zero noise. If you haven't personally measured these numbers, then I assume your statements are based on what you've either read or been told. Why do you believe these readings/sayings? Are you certain? If you are requiring other people to post proof, then why not you too? Aside from your offer to do a demo (which I'm not going to consider proof of anything regardless of the results -- and I will leave it to you and/or the readers to consider why I have this viewpoint).

 

I any case, I do really appreciate your desire to be logical (and share your suspicion of "audiophile cables" so I think we are agreeing on the basic premise of this discussion). Its really just that you, and everyone, should place your own views under equal scrutiny as you do those of others. Don't stop questioning!

[Tony Lauck]

I remember that cartoon, but I thought your comments were very personal. Plissken has the same right to an opinion as others here.

 

We each have a right to our own opinion. We do not have a right to other people's attention or respect, although it is reasonable to expect that if one is friendly and helpful or, if one lacks knowledge demonstrates a willingness and ability to learn from the attention.

[plissken]

I appreciate your desire to be precise and to avoid conjecture, but where is it written that Ethernet PHY is noise free?

 

Anything generating a signal will have an SNR associated with it. The S part of SNR is key here. What do you think the signaling rate is on a PHY? You answer that and then you will know why it's of no concern.

[jabbr]

Anything generating a signal will have an SNR associated with it. The S part of SNR is key here. What do you think the signaling rate is on a PHY? You answer that and then you will know why it's of no concern.

 

I don't have the same assumption you are making. You can make a logical argument that your assumption is correct, but proving this is more difficult than you are assuming. I would say that the onus of proof is on the assumption. To start out I don't assume the system is linear with independent frequency "channels" -- I am not assuming that high frequency trash does not have audible effects -- indeed the counter proof is that I hear clicks when my equipment is sitting next to my 4K monitor -- and regarding the infallibility of noise rejection from TP: HDMI and DisplayPort are both LVDS/TP. So again.

[plissken]

I don't have the same assumption you are making. You can make a logical argument that your assumption is correct, but proving this is more difficult than you are assuming. I would say that the onus of proof is on the assumption. To start out I don't assume the system is linear with independent frequency "channels" -- I am not assuming that high frequency trash does not have audible effects -- indeed the counter proof is that I hear clicks when my equipment is sitting next to my 4K monitor -- and regarding the infallibility of noise rejection from TP: HDMI and DisplayPort are both LVDS/TP. So again.

 

What would 'My Equipment' be? And how, other then anecdotal (non evidence), does it pertain to 25Mhz signaling of Ethernet PHY?

 

What on earth would make you say that the noise you are hearing is ingress via the HDMI or Display port cable when it could be any number of single ended circuits in a device picking up that noise those cables just happen to be plugged into?

 

Grab a video camera and tape you getting close to the 4K monitor. Now remove the HDMI and DP cable and try it again. I would be curious if the noise is still present in your video.

[jabbr]

What would 'My Equipment' be? And how, other then anecdotal (non evidence), does it pertain to 25Mhz signaling of Ethernet PHY?

 

What on earth would make you say that the noise you are hearing is ingress via the HDMI or Display port cable when it could be any number of single ended circuits in a device picking up that noise those cables just happen to be plugged into?

 

Grab a video camera and tape you getting close to the 4K monitor. Now remove the HDMI and DP cable and try it again. I would be curious if the noise is still present in your video.

 

The equipment which exhibited that doesn't matter -- I just moved my equipment away -- I'm not trying to prove anything (honestly!). I use this example as a counter example to any other single system which doesn't exhibit some behavior. I am not even saying that this is caused by Ethernet -- my own bias is that any so-called audiophile Ethernet cable would not be able to improve upon Belden mediatwist with 2 good connectors -- all I am saying is that noise issues can be complicated and can challenge one's assumptions. Now look, you've assumed that my Ethernet PHY is 25mhz! why did you make that assumption?

[plissken]

The equipment which exhibited that doesn't matter -- I just moved my equipment away -- I'm not trying to prove anything (honestly!). I use this example as a counter example to any other single system which doesn't exhibit some behavior. I am not even saying that this is caused by Ethernet -- my own bias is that any so-called audiophile Ethernet cable would not be able to improve upon Belden mediatwist with 2 good connectors -- all I am saying is that noise issues can be complicated and can challenge one's assumptions. Now look, you've assumed that my Ethernet PHY is 25mhz! why did you make that assumption?

 

It's not an assumption about the 25Mhz signalling rate. Read some T.I, National Semi, Broadcom implementation guides.

 

http://www.ti.com/lit/ds/symlink/dp83865.pdf

 

I understand your the point you are trying to make. But I haven't seen anyone here take what you said you had a problem with and show that it was the Ethernet PHY.

 

In your scenario you could have, with minimal effort IMO, ran the issue down a bit further. The issue I perceive is you explicitly mentioned low voltage differential twisted pair cables and hearing the noise. I simply asked in response if you did or could do any further isolation.

 

It's a fair question IMO.

[jabbr]

It's not an assumption about the 25Mhz signalling rate. Read some T.I, National Semi, Broadcom implementation guides.

 

http://www.ti.com/lit/ds/symlink/dp83865.pdf

 

You are assuming that the 25 Mhz clock rate is the signaling rate. 25 Mhz clocks are pretty standard on motherboards and ethernet devices but don't by themselves determine signaling rate (when 'overclocking' a CPU, you don't modify the 25 Mhz clock). Now 100 Mbs Ethernet with 4 tp get's you 25 Mhz but we are probably talking 1Gbe generally, and there's 10G fiber with a duplex connection, and 40g with 4 of these or 100g with 10 -- assuming multimode, but singlemode uses a single pair -- and now there's 25g and 50g Ethernet so it gets pretty complicated.

 

I understand your the point you are trying to make. But I haven't seen anyone here take what you said you had a problem with and show that it was the Ethernet PHY.

 

In your scenario you could have, with minimal effort IMO, ran the issue down a bit further. The issue I perceive is you explicitly mentioned low voltage differential twisted pair cables and hearing the noise. I simply asked in response if you did or could do any further isolation.

 

It's a fair question IMO.

 

No really the point I am trying to make is:

 

1) Challenge your assumptions -- at least beware easily accepting these as fact

2) As Tony's father said: "Not seeing a car at an intersection is not the same as seeing that there isn't a car"

3) Don't believe everything you read -- I probably wrote some of that and I was probably making some of what I wrote up.

4) This should be fun:)

 

5) If I ran every issue I had down, I would never leave my house in the morning:)

[Paul R]

I can see your point of view.

 

I think it may be a little overly critical, but I certainly can understand the reasons for that too.

 

Give the folks a break, they are excited about this stuff, and it will take a while for them to learn enough to have a deeper understanding. Most folks take the time to learn if they are really interested, but do remember, learning is not a boring exercise. There are all these "gee whiz" and "wow!" moments along the way.

 

 

-Paul

 

 

Only speaking for myself I have seen too many people read a white paper and come away with a gross misinterpretation that actually did them more harm than good.

 

There is a really good example at Whats best forum where a member there asked about Ethernet Jitter being stored in a file on HD.

 

He read a paper and didn't realize that the paper was referring to ether RLL or MFF endoding algorithms used on a hard drive.

 

I think this is the link: Does a file with jitter keep the jitter when saved to a hard drive?

 

The issue is that despite others telling him differently he, for several pages, dug into a position he wanted to be in.

 

I've seen this at other sites also.

 

Like here on the issue of packet loss.

 

1. Why is it being constantly paraded in this thread? I can only think the person doesn't realize how little packet drop is an issue in a home LAN environment. CRC32 will only miss 1 in 4 billion bits.

 

2. Jitter/Interpair Skew/Crosstalk etc. These are issues for any structured electrical cabling and some of it even effects optical. But they are well engineered for and buffering takes care of this for the audio reproduction side.

 

I just haven't seen any direct evidence in support. No scope, no mic, just conjecture. I'm all for a data driven approach.

[firedog]

And back to the original topic: does anyone else find it curious that over a week has gone by and no results of the "test" have been published? "Preparing the video" shouldn't take that long if you are serious about your test and informing the public.

[esldude]

And back to the original topic: does anyone else find it curious that over a week has gone by and no results of the "test" have been published? "Preparing the video" shouldn't take that long if you are serious about your test and informing the public.

 

I would give them a little longer. Hoping it might be a feature article Monday morning. If not this week, then yes, I would begin to wonder why?

[One and a half]

And back to the original topic: does anyone else find it curious that over a week has gone by and no results of the "test" have been published? "Preparing the video" shouldn't take that long if you are serious about your test and informing the public.

 

Video editing takes forever. Remember the videos of RMAF are recorded in October, they don't post until a few months later.

 

Could also be they stuffed up the votes...

[plissken]

You are assuming that the 25 Mhz clock rate is the signaling rate. 25 Mhz clocks are pretty standard on motherboards and ethernet devices but don't by themselves determine signaling rate (when 'overclocking' a CPU, you don't modify the 25 Mhz clock). Now 100 Mbs Ethernet with 4 tp get's you 25 Mhz but we are probably talking 1Gbe generally, and there's 10G fiber with a duplex connection, and 40g with 4 of these or 100g with 10 -- assuming multimode, but singlemode uses a single pair -- and now there's 25g and 50g Ethernet so it gets pretty complicated.

 

I'll have to refer back the paper but nothing is mentioned about a clock divider.

 

The point not to be missed are a few. Ethernet only works because it is well isolated. It does a great job at noise rejection. No one can hear 25Mhz (or higher) and no one, outside of mere conjecture, has proven in any way, shape, or form, that it negatively impacts what happens on the USB buss.

 

Lumping packet based Ethernet and packet based USB together is like saying because diesel fuel doesn't flash to open flame then you are equally safe around gasoline since they are both petroleum products.

[plissken]

And back to the original topic: does anyone else find it curious that over a week has gone by and no results of the "test" have been published? "Preparing the video" shouldn't take that long if you are serious about your test and informing the public.

 

As a long time Ars reader (2000) they take their time and it's worth it IMO. They've done a multi-part series on Commodore Amiga and that's been years in the making and worth it.

[Paul R]

I'll have to refer back the paper but nothing is mentioned about a clock divider.

 

The point not to be missed are a few. Ethernet only works because it is well isolated. It does a great job at noise rejection. No one can hear 25Mhz (or higher) and no one, outside of mere conjecture, has proven in any way, shape, or form, that it negatively impacts what happens on the USB buss.

 

Lumping packet based Ethernet and packet based USB together is like saying because diesel fuel doesn't flash to open flame then you are equally safe around gasoline since they are both petroleum products.

 

It may be we are oversimplifying this. If your Ethernet connection happens to be over a HDMI connection for instance, you bet electrical noise flooding into the computer can be a serious and audible issue.

 

If we are talking only UTP Ethernet then there is probably no reason to assume that electrical noise is going to flood the computer or affect the DAC.

 

Unlike a USB cable, where electrical noise on the cable that can seep into the DAC is almost a given these days.

 

Now, down at the fine edge of possible, but unknown exactly how (im)probable, you have things like UTP emitting RFI, acting as an antenna, conducting electrical noise from the switch, etc.

 

And all of that is an attempt to find a reason why Ethernet cables have sounded different to some folks. Which is hardly questionable, but deucedly difficult to explain. Have to get to the bottom facts on all this before it will really make sense.

 

And unfortunately, it may turn out that the differences are not measurable at all - which makes it very possible they are not 'objectively real'.

 

Though that has nothing at all to do with whether they are actually heard or not.

 

It is also possible that the measurements were not sensitive enough, or failed to measure some factor or combination of factors.

 

 

Again, most of the things mentioned in the last five paragrahs are definitely on the fine edge of possible - as far as making an audible difference. They will probably remain right there too.

 

It is very expensive to do the kind of testing necessary to measure for those possibilities, let alone prove or disprove them.