I went out to WGutz last year with a 100 foot, $13 cable from Amazon.
Compared to his Cryo treated 15 foot boutique cable he only hit 60% reliability.
Yep, a cable that was 660% longer, $0.13 a foot.
Later in the day when listening to Tidal he couldn’t hear when the cable was in the system or removed.
We queued up an ~11 minute piano solo and I showed him that then entire track after the first 10-15 seconds had been playing with the cable removed the entire time. |
Just as a matter of fact, ethernet doesn't always use
TCP/IP because there isn't really a "TCP/IP" protocol. Those are two
different protocols for two different layers of the software interface.
In all likelihood, it's using TCP/FTP.
You have a gross misunderstanding. There is indeed a TCP/IP protocol. It's transmission control protocol over internet protocol. UDP/IP is also a protocol. Ethernet isn't any one protocol. It's made up of many moving parts and the abstraction it provides up/down stream layers is why it's so robust, reliable, and elegant. There is no such thing as TCP/FTP because FTP doesn't live at layer 3. It's layer 7 (application). |
It seems to me that what has been largely overlooked in this discussion (with the exception of the brief post by Markalarsen) is the fact that 100% of the energy of an electrical signal, especially one that as in the case of Ethernet contains spectral components at very high RF frequencies, does not necessarily go only where it is supposed to go. Experienced designers of high speed digital circuits (of which I happen to be one) will recognize that.
And given that a number of members here who are highly respected and highly experienced And I encourage and invite every last one of you to, in your own setup, to evaluate some spec meeting, but very inexpensive, CAT5e/6 to your own esoteric CAT5e/6 cabling. I’ll provide a client, server, layer 3 managed switch with a LAG setup for dynamic LACP. Of course you will not be allowed to know which cable is being used. |
@jinjuku what do you mean by 60% reliability?
what kind of tests did you run? @acepilot71: William chose the testing format. I randomly generated a sequence of cable swap out. William went with either listening to an entire track or part of a track. He only obtained 60% accuracy of stating what cable was in use. Which means he was also in error 40% of the the time. |
When someone tee's up a track in Tidal on their 100Mb/s cable modem and they pull the Ethernet cable and the song still plays what is actually happening from a cable perspective at that point?
|
@ burgh
I wired up heavy machining plants in Wixom, Mi, Cleveland, Oh, Youngstown, Oh, Mentor, Oh etc...
Didn't matter if it was Ethernet or 485 Serial it was always shielded because all the 3 phase power and large inductive motors.
Homes don't have these challenges. So the shielded designs don't help, but they could hurt if the shield ties end points to chassis and creates a ground loop. Floated shield would be fine however and the costs are minimal if it makes the audiophile feel better.
No one is saying don't use quality cabling. But the fact remains that the highest performance Ethernet cable out there is Belden's 10GX and it's about $2/foot terminated.
You can get their normal bonded pair CAT6 for ~$1.40 a foot.
I've personally had Ethernet cabling from $27 a foot to $233 a foot and compared directly to 315 foot of BerkTek CAT5e. No difference. |
Such as to D/A converter circuits, where timing jitter amounting to far less than one nanosecond is recognized as being audibly significant. (See the section entitled "Jitter Correlation to Audibility" near the end of this paper).
This is where we have a problem. Where in the playback system are you referring to this jitter? The entire point that, and I will keep with Tidal as example, is that once local buffer is filled up, and buffers are indeed static storage, that any timing variance ceases to exist. It's why I can watch Netflix 4K streamed with no issues. The fact of the matter, and it is indeed FACT, is if I pull the network cable for 1 second I've introduced
1000000000ns of jitter but some how the playback system has managed to deal with this and deal with it to the point that if you are blinded you couldn't tell me if your life depended on it. While this timing difference may be in the DA converter circuits, that's not the same as Ethernet which is burst in nature and asynch.
In addition to the effects of shielding on radiated emissions, shielding would presumably also affect the bandwidth, capacitance, and other characteristics of the cable, in turn affecting signal risetimes and falltimes (the amount of time it takes for the signals in the cable to transition between their two voltage states), in turn affecting the spectral composition of RF noise that may find its way past the ethernet interface in the receiving device.
If your playback equipment is susceptible to standards compliant Ethernet cabling affecting playback I would say your equipment is defective. Again using Tidal: When I was at WGUtz place the 100', $13 cable allowed Tidal to cache the entire track just as quickly as the 15' boutique cable. Modern PHY's put the interface into a low or no power state when not transmitting. An 11 minute song was cached, in it's entirety, in about 15 seconds.
Obviously noise that may find its way to circuitry of the receiving device that is downstream of its ethernet interface, as a consequence of the signal it is receiving, will be eliminated. On the other hand, airborne RFI may increase since the cable would no longer be connected to a termination that would absorb the signal energy. Which of those effects may have audible consequences, if in fact either of them does in some applications, as I indicated in my previous posts figures to be highly component and system dependent and to have little if any predictability.
Then everyone is screwed. I don't believe that to be the case. I can build a world class client / server setup for $700 to feed a DAC.
I don’t doubt your experience. However, I also don’t doubt experiences that have been reported by members such as DGarretson, Bryoncunningham, Grannyring, and others here who are similarly thorough when assessing a change.
You should doubt me and everyone else. The difference being I actually showed up at a members house and we went through this where the perceived changes disappeared once sighted bias was controlled for. I have no problem doing this elsewhere. |
That’s interesting. Can you please tell us more about your comparison? Was it based solely on measurements? If you conducted listening tests, can you please tell us how they were conducted? Which specific cables did you evaluate?
Nordost Heimdall 2 one meter $699. BerkTek Hyper 5e, 98 meters, $90. Cary Audio DMS-500 RME UFX A/D duties fed to my laptop ARTA for measurements. Nothing in spectral components, nothing in the 11.25Khz jitter, nothing in the linearity testing, nothing in the noise floor. I also did a cascade plot with my measurement mic. Nothing changed other than room ambient noises from one run to the next. This would even be slightly different using the same cable on multiple runs. I also captured the tracks and posted them elsewhere for people to download and evaluate. During playback I swapped out the cabling and simply asked for people to tell me when the cabling was swapped. Only a handful tried and all failed. |
I am referring to timing jitter at the point of D/A conversion. And I am referring to the possibility that cable differences may affect the characteristics of RF noise that may bypass (i.e., find its way **around**) the ethernet interface, buffers, etc. and **to** the circuitry that performs D/A conversion. At best that’s a design issue of the connected hardware. Not a cabling issue, IMO, where the cable meets or exceeds spec. The reason behind my thinking is that RF noise generated, by say impedance mismatch in a cable, is due more to length and twisted pairs not staying in mechanical balance than differences in 12-15 foot typical patch cable. Not to mention the horizontal run is most likely some junk CCA. At worst, and if we take your interpretation, I would say from what I’ve seen, most of the incredibly expensive CATX cabling doesn’t pass IEEE / TIA spec and it introduces noise and audiophiles don’t understand what they are enjoying is a degradation of their playback chain. That’s a stretch for me though. Bottom line it would be measurable as the outputs of a DAC are voltage output devices. Regarding disconnection of the cable, putting aside the possible significance of airborne RFI doing so would of course work in the direction of reducing noise that may be coupled from the input circuit
Again this is measurable. Also why I like WiFi. It’s low latency, high throughput, no measurement (either instrumented or human) shows harmonic component’s of RFI frequencies showing up. I think you just said that removing the plug from the back of the client would work in the direction of reducing noise... So with that said I would encourage a blinded evaluation session where the Ethernet cable is removed during playback of a track and the listener successfully is able to indicate that removal or insertion. Let’s even paint a scenario where that’s actually the case. That noise component is most likely going to be buried in the noise floor (or a component of) the DAC, in the -130dB range on a competently designed piece of gear. You can’t hear anything that low even if it’s there. And if it’s not. It’s not. Putting it all very basically, responses by those claiming ethernet cables won’t make a difference nearly always focus just on the intended/nominal signal path. The basic point to my earlier posts is that in real world circuitry parasitic signal paths also exist (via grounds, power supplies, parasitic capacitances, etc.), which may allow RF noise to bypass the intended signal path to some extent, and therefore may account for some or many of the reported differences. Your response sounds like a guess. I have another theory for all the differences and it’s sighted bias or really poorly designed, often expensive, equipment. I’ve been recommending either WiFi (ubiquiti) of wired (Intel Server PCIe) NIC’s (left and right they are available NIB or as new but pulled) for for ~$25. They seem impervious to what cabling I’ve thrown at it. |
Bottom line is I went out to Williams’ house. We all had a good time and food for thought. Even brought a bottle of Michters' single barrel Bourbon.
He didn’t know that Tidal would cache the entire song and you could pull the plug.
People will often take something trivial and make as big of a deal of it as suites their agenda. Nothing I, nor anyone else, can do about that.
At that point it’s the agenda speaking and not anything else. |
You can make what ever mountains out of molehills you like. Nothing I can do to change that.
I've also offered to do this at a show. The underlying theme is people and their beliefs are easily separated. |
Of course when you are able to get people "jacked up" on hard liquor it is very possible What? Jacked up on liquor? It was a host gift. No one was jacked up on anything. If four shots total were poured for the entire day I would have been surprised. I simply wanted to bring William something that he couldn’t get elsewhere. Michters is really good bourbon that is not readily available just anywhere. |
I don't understand why folks that choose not to believe in differences
in cabling actually debate with those of us that can detect differences
in sound quality.
WRT to Ethernet cabling it's as simple as putting money where your mouth is. If you don't trust your ears why should anyone else? |
Michters single barrel bourbon? Whoa! Sorry, Charlie! We want tuna that tastes good, not tuna with good taste. I'll bet you know a good Bourbon same as you know when a cable sounds good. By looking at the label as you drink it. Michters is the Bourbon that Pappy Van Winkle noted to get if you can't get theirs. |
Question to self: why is it always the same ones who can hear differences in fuses, differences in directionality or differences in cables but can never seem to demonstrate it blind therefore validating their ears? 😬
|
We are talking about Ethernet cables ( I shouldn't have directly quoted btw ). I’ve already setup with a managed L3 switch, dynamic LACP LAG, the data cabling equivalent of the WW comparator.
No one demanded any up front payment. Links to those posts please. What I remember is that there was an offer of 5:1 and benefiting the person that could hear the difference in Ethernet cabling. |
Sorry, but upfront payment was demanded, along with other prerequisites. The posts were deleted by the moderators. The promoter of the scheme is quite persistent, so it’s likely you’ll see what he calls an "opportunity" promoted again. Just stay tuned.
Upfront payment wasn't demanded. It was also loser pays expenses. This would all be handled after the evaluation. |
If "advocates" of blind testing were serious, they would submit the protocol for advance review here, and the test would be performed in public, which would allow witnesses and an opportunity for others to participate. This could be done at an audio store or audiophile club. But the "advocates" haven’t shown any interest in that; their call for scientific testing is just a red herring.
It's been offered to be done at a show. I've posted a video of the setup in action and it lists all the components and demonstrates the LACP dynamic LAG. It's also been offered to train someone (random pick) to do the cable swap. |
My offer, as it was for William, was to do this in the listeners setup. No money involved.
Seems way more than fair to the claimant. |
No, it tests the claim.
And sighted evaluation, and anecdotal accounts are 100% rigorous. Gotcha.
Guess what happens when you at 1+1 and do it 48 more times over the course of time? You increase the sample size. |
I think you are confused.
If you state that you can jump 10’ straight up from a standstill and either I bring a 10’ high bar or we use your 10’ high bar.
What are we exactly testing, why isn’t the 10’ high bar valid, and why do we need 49 other people to attempt to jump over a 10’ high bar?
Behold everyone where simple questions somehow can't be answered.
|
LOL. Simple questions can’t be answered.
I have the sighted subjectivist crowd DEMANDING scientifically rigorous testing. Oh this is a hoot.
If you can find a single post where I was maintaining that this was supposed to be AES worthy let me know. I'm simply out to test an individuals claim. Nothing less or more.
Also let me know of ANY AES papers that have accepted sighted evaluation. |
I’ve simply pointed out that what you claim is a scientifically valid test
Links to posts please. |
>>>>Uh, but nobody’s making that claim. No one is claiming
he can fly or that UFOs are real, either. That’s what we call a Strawman
Argument. People can think up all kinds of absurd cases that have no
relevance to the actual issue at hand
What? All sorts of zero evidence based claims are in this very thread. I view a claim of a improved bass response due to a change in Ethernet cabling is just as specious as a claim you can flap your arms and fly. I've only seen one person have the intellectual honesty to go ears only. |
I've come to the conclusion that this thread is a Uri Geller convention.
|
Finally, regarding the potential effects of RF noise on analog circuitry, the sonic character of whatever audible consequences may result from effects such as intermodulation of that noise with the audio signal, and demodulation of AM (amplitude modulation) spectral components of the noise, it seems to me could very well manifest themselves in either or both of the two categories you defined.
CAT5E UTP cabling has good noise rejection up to 30MHz and this is quoted from "The Antenna Myth" by Siemon: The Antenna Myth
It is a common myth that screens and shields can behave as antennas because they are long lengths of metal. The fear is that screens and shields can "attract" signals that are in the environment or radiate signals that appear on the twisted-pairs. The fact is that both screens and shields and the copper balanced twisted-pairs in a UTP cable will behave as an antenna to some degree. The difference is that, as demonstrated by the simplified loop antenna model, the noise that couples onto the screen or shield is actually 100 to 1,000 times smaller in magnitude than the noise that is coupled onto an unshielded twisted-pair in the same environment. This is due to the internal pairs’ well-defined and controlled common mode impedance to the ground plane that is provided by the screen/shield. Following is an analysis of the two types of signal disturbers that can affect the noise immunity performance of balanced twisted-pair cabling: those below 30 MHz and those above 30 MHz. At frequencies below 30 MHz, noise currents from the environment can penetrate the screen/shield and affect the twisted-pairs. However, the simplified loop antenna model shows that the magnitude of these signals is substantially smaller (and mostly attenuated due to the absorption loss of the aluminum foil), meaning that unshielded twisted-pairs in the same environment are actually subjected to much a higher electric field strength. The good news is that the balance performance of the cable itself is sufficient up to 30 MHz to ensure minimum susceptibility to disturbance from these noise sources regardless of the presence of an overall screen/shield
Also WRT to jitter. Everytime some one brings up jitter and asynch Ethernet and the DAC output you can safely place them in the "They have no idea what they are talking about" bin.
|
ethernet interface circuits commonly deviate from that idealized model to an audibly significant degree.
That's a design deviation or poor engineering however. I would say you have an flawed equipment problem if it loses it's composure when two spec passing cables are presented. How many home setups use anything near the max 328 foot run? 150 foot? 75? 50? I'll even go one further: Using the industries most stable, from many aspects and especially impedance stability, Belden 10GX would be the cable to go to for such error ridden playback devices. What does Nordost, WW, AQ, Chord, Supra, know that Belden, a leader in high performance data, analog, broadcast cabling, with an engineering bench that is most likely larger than the entire employee base of some of these other companies on an individual basis, doesn't?
IMO there is a more than sufficient body of anecdotal evidence,
There are a lot of people that believe in the Flying Spagetti Monster, some of whom seem to be credible human beings. But the point still stands now doesn't it?
provided by audiophiles whom I consider to be highly credible
What metrics would make them highly credible for you? I think a highly credible audiophile is one that is able to validate their hearing in an intellectually honest manner. I've tested out 3 botique cables from $27 to $233 a foot. In three systems, didn't do anything more or less than 315 foot of 5e. Also in a forum members system a $13, 100 foot cable, put the bamboozle on that members system when blinded. |
Would you care to list the components involved with these 3 boutique testing? 1: Netgear ReadyNAS <> Cisco SG 200-8<> Cary DMS-500 2: Netgear ReadyNAS <> Cisco SG 200-8<> Cambridge 851N 3: Netgear ReadyNAS <> Cisco SG 200-8<> DIY ASRock 5300 Kabini based motherboard, 8GB Kingston DDR3, 120GB Kingston SSD, Windows 8.1 Pro, JRiver Media Center 22, WASAPI Exclusive Mode, Intel Pro 1000/Pt dual port. Ports 7/8 on the SG 200-8 configured in a active dynamic LACP LAG and on the client computer the Intel NIC in a passive dynamic LACP LAG. This allows for removal of the cabling without interruption of layer 2 CAM table. Either cable can be swapped in real time while the music is playing from the back of the client computer. 18 inch Belkin Gold USB cable to Emotiva Stealth DC-1 balanced out to either a RME Fireface UFX and AKG 701’s for realtime monitoring or Yamaha P2500S to Curt Cambell Statements. But all that doesn’t matter because if you are claiming your N.A. Ford Focus runs an 11 second 1/4 mile it really doesn’t matter what car I own. We aren't testing my claim, we are testing yours. |
IMO, whether Ethernet cable a factor is function to DAC buffer size. Bigger the buffer, less factor.
Less a factor for what?
Problem is most DACs buffer's ~6 seconds and I can easily hear a difference between Ethernet cables. If you in my neck of woods, I would love to demonstrate in my system.
I can set JRiver for 6 seconds. So an Ethernet cable will effect that 6 second buffer but if I set JRiver for 60 minutes of buffer it won't? What about all the other operations going on? CPU caching operations, Interrupts, DMA transfers, Memory Paging, SMPS? This is why I don't buy into Al's argument. With all this going on what ever variation of cable is going to be swamped by the system wide operations going on continuously. What is your neck of the woods and what is your setup? |
Member Bryoncunningham, who IMO is an especially astute and perceptive listener, and is very thorough in his evaluations, described realizing a substantial sonic improvement by changing from a garden variety unshielded ethernet cable to an **inexpensive** shielded type. I would question the quality of the streamer that is so susceptible. Both the $4000 Cary and the $1600 851N were impervious. Broke no sweat with my 315 foot cable. Also the 315 foot generic I made and the 3 foot Nordost both had sub 1ms ping times and both were able to transfer at a sustained 107MB/second. |
Note that I also said as follows in a post dated 4-24-2018:
Member Bryoncunningham, who IMO is an especially astute and perceptive listener, and is very thorough in his evaluations, described realizing a substantial sonic improvement by changing from a garden variety unshielded ethernet cable to an **inexpensive** shielded type.
....It should be noted, though, that Bryon’s experience involved an Ethernet cable that was connected directly to one of his audio components, not to a computer that was in turn connected to the audio system. Regards,
-- Al
Question is, do "Audiophile" Ethernet cables make any difference/ improvement in sound quality?
|
And in fact several of the improvements I and a number of others have
referred to in this thread involved "upgrading" to inexpensive cables.
My proposal is that gear that is susceptible to a change in a short length of Ethernet cable is faulty. I'd rather have gear that works with in spec cable reliably, consistently, without variation. |
Here is a windows system that I optimized to only 37 background tasks running. CPU was running at 2-4% I turned on all the cache performance counters. I'll also do you one better: I'll start playback over the network of a 24/192 file and capture it into my ADC. During the capture I'll also transfer a large file of ~ 1 GB in size. I'll do it with a short 10 foot patch cable and a 100 foot $13 cable. Anyone is free to listen and tell me which track is the 10 footer and which is the 100 footer and when the 1GB file is also being transferred over the same cable that is playing the 24/192 track. |
Great, this is coming from a guy who conduct listening tests on a headphone.
WGUtz didn't have any headphones when I was out there. Lying can't help you now. |
To all everyone now sees that your repeated demands for double-blind listening tests and your $25,000 USD listening challenge is a fraud and that you don't see this is a reflection of a problem with you which you have previously acknowledged in this group and which I will not mention because I am not going to make fun of you even if your problem is a result of your own bad choices.
So are moderators allowing monetary inducement now? If so I would like to repost my offer. |
Still waiting for my phone call from Geoff to make my system sound great. All for the low, low price of $60.
Got to figure out how to hardwire my cell phone so the RFI bogymen don't mess it up. |
|
Sigh.... 1. The system used for the test is not sufficiently resolving to distinguish differences that might be audible on a better system. I'll do this on their system 2. There are mistakes in the system that were not found even if there was a search for mistakes or errors. Saying that the test procedures are thorough doesn’t necessarily mean they really are thorough. I'll do this on their system 3. The hearing capability of the testee is not up to the task. The claimant is the testee. If they can make a claim sighted and state it's just their ears then they can do this in the dark. 4. For a blind test involving many trials, say 10 or more, the odds are high that the testee doesn’t have the focus or strength to dinstinguish audible differences for periods of time without tiring. You can’t just say oh, well, that’s the way it goes. Then the differences aren't as 'readily apparent' and 'easy to discern' or other colloquialisms that I've seen bandied about. 5. System issues that go undetected due to naïveté of those involved in the test. Directionality of Fuses, directionality of cables, Polarity check of all connections, etc. I'll do this on their system
6. Using unfamiliar music for the test. I'll do this on their system 7. Using a test system unfamiliar to the testee. I'll do this on their system. In 2015 at an AQ demo I saw a people hear differences in Ethernet cabling in a really crummy room on equipment they had no familiarity with, music that most likely they weren't familiar with. 8. Weather and other “external variables” that affect the sound of any system that could make hearing subtle differences in sound difficult or impossible. I'll try to schedule my visit with you to avoid thunderstorms. What if results are positive? Those get posted too. |
They can only conclude that the drug was not effective in that patient during the test period. And that is why drug trials include more than one patient.
Correct and I'm willing to do this with a few 'patients' (claimants) to see if their drug of choice "sighted evaluation" is merely sugar pill. |
You could do that. But why not swap cables in someone's personal system that he/she's intimately familiar with?
This black box would sit in the persons system. They would provide the USB cable on back. |
It's what I and others have repeatedly stated: long term listening is the correct way to identify sound signatures and clues. You can get the gist of a cable or fuse or what have you but it takes immersion in the music, over time, to correctly assess anything, unless it's something glaringly apparent.
Great. Love it. Then I will setup a server/client/switch where there will be power going into the black box, a USB port on the output side and a Cisco Switch were I can issue a shut/no shut on FA 0/1 or FA 0/2 and turn power off to that port. The client will have a dual port NIC in a LACP Passive LAG (The switch would be port channeled and LACP Dynamic LAG). Use JRiver in WASAPI exclusive mode or ASIO depending on the claimants DAC and we could literally collect data for a 1-6 months EASY. They would never know when the shut/no shut on either FA 01/ or FA 0/2 would have been made therefore they wouldn't know which Ethernet cable was in use. |
What is it that you propose to test?
What ever it is that allows someone to take a $340, 1 meter RJE cable, an declare all sorts of sonic bliss. |
Once again it looks like we’re faced with the likelihood that there is no such thing as a scientifically valid listening test. There are too many variables in the case of long term listening to be able to definitively conclude anything about the device under test. If someone can’t hear it in a short term test there’s no guarantee he can hear it in a long term test.
Interesting so all the people that are saying they hear differences in data cabling aren't hearing anything. |
Michael do you even understand how async, Ethernet based audio works?
When I start playing a song, pull the Ethernet cable, and the song plays on for the duration, like it does with Tidal, what effect is the cable having?
My offer even extends to you at your shop. |
My 2.0 setup is 📻.
My HTPC setup is hardwired. |
There won’t be any music to play if there was no cable between the switch and renderer to begin with....so by pulling Ethernet cable proves nothing
So you can't answer simple questions like what happens to the SQ when the Ethernet cable is pulled and the song still plays? I get it that you have no clue how this stuff works but it at least should be a 'woke' moment for you... |
I get it that you have no clue how this stuff works but it at least should be a ’woke’ moment for you...”
Oh I know exactly how the streaming works, but you can’t seem to grasp a simple fact that once 10, 20 or 30 seconds of downloaded 1’s and 0’s runs out of buffer, you need that Ethernet cable to download the next 10-30 seconds of bits and so on.
Unlike you I enjoy listening to my music uninterrupted, not in 10-20 seconds of voodoo crap you have been trying to peddle. Well that depends on the size of the buffer. Tidal has the ability to buffer the entire track. Williams 11 minute piano piece played entirely after the cable was pulled after the first 10-15 seconds. Buffers are designed to never entirely allow to run out. There are hold down algorithms that make sure the buffer never even comes close to running out. Then we have wire speed to contend with. I get ~36MB/s over my cable broadband. I can cue up entire 24/96, 10 minute tracks, in 6 seconds. Sorry that your playback system is so completely compromised. |
I thought "everything matters"? |
