Ethernet Cables in for evaluation

Discussion in 'Audio Hardware' started by Puma Cat, May 17, 2019.

Thread Status:
Not open for further replies.
  1. Brad2021hk

    Brad2021hk Forum Resident

    Location:
    Sacramento, CA
    I'm not usually the "stay in your lane" type of debater. It might apply here.

    Once the signal quality crosses the threshold to good enough that the receiver can recover the data with a sufficiently low bit error rate, it doesn't matter anymore. On your way to getting good enough signal quality, there are analog transmission line principles used. You can't have better than a bit-perfect packet of data. The aesthetics of a "more perfect" waveform will not matter. I've typed this on this message board some other time. Jitter is a real thing. Jitter does reduce timing margin in digital circuits. Once you reduce jitter (and other things that hurt margin) sufficiently that you close timing, it doesn't help to reduce it more. Seriously. I'm a computer engineer. I'm not pointing that out to brag. I'm point that out to make the point that I understand how these things actually work.

    Jitter on DAC clock can also cause distortion. That jitter is disconnected from the jitter on a wire transport that brings data into the DAC. Making the input data less jittery does not make the DAC's clock less jittery. The same word. Similar phenomenon. Not actually connected.

    If DAC chip supply noise could make the internal clock jitter so much that it is audible, that would be really insanely bad design. This is what is argued in that silly Ethergen white paper. If it is not insanely bad design, computers don't work. Like I said, jitter in a digital domain reduces margin. If power supply noise at a reasonable level produced more than 10's or at most 100's of picoseconds of jitter, that jitter would make it impossible to close timing at any reasonable frequency. Computer chips wouldn't work above a couple hundred megahertz. Computer chips do work. Jitter induced by the supply is really small. The amount of impact that jitter can have on the analog waveform is therefore extremely limited.
     
    Jinjuku, jimbutsu and jeff kleinberg like this.
  2. jeff kleinberg

    jeff kleinberg Senior Member

    Location:
    Ct
    Brad to be clear, I really do defer to your expertise in this field. I appreciate you taking the time to provide technical responses, and not talking down or name calling. My understanding of jitter is it can be induced by noise leading to clock errors and functionally can result in all of the frequencies associated with a distinct musical event not arriving at the listener in a manner to which the ear is accustomed. Not necessarily just the initial sound but the delivery of the reflections and reverberations as well. This can lead to a less realistic presentation.

    I am not dead set on jitter as an answer. I am just asking a question, why does this cable change my systems performance in such a reliable and positive way? I have read white papers for a component and 20 pages of a negative review of same component to try to understand what I am experiencing. I've taken a lot of **** in this thread as well for the same reason. I'm am asking questions and seeking explanations. I like to understand why I am experiencing something counterintuitive.
     
  3. jeff kleinberg

    jeff kleinberg Senior Member

    Location:
    Ct
    I am in complete agreement this is not a bits issue, but possibly a jitter in the clock phenomenon due to electrical noise or increased heat due secondary to the processor doing a lot of error correction or just poor design/location. There are a lot of companies going whole hog on clocks now, from satellites to super ovens:), so someone thinks there is serious room for improvement. Again appreciate the knowledge.

    I just wanted to keep the facts straight about digital signals BEING analog square waves.
     
  4. Brad2021hk

    Brad2021hk Forum Resident

    Location:
    Sacramento, CA
    What you're describing with timing is phase relationship. The difference in timing from jitter on a digital clock would be on the order of nanoseconds. A 10Khz sine wave is going to have a period of 100us. A phase shift would be changing timing like 10's of microseconds. It would be really hard to figure out how jitter in the digital domain is causing phase shift to this degree.

    To put this into perspective, you can get 10us of delay by moving your head about 3cm farther from a speaker. That's not very much. This explains why speaker position and listener position is generally important to sound quality. Sound moves relatively slow through air. Some people are really sensitive to phase. It doesn't take a lot of positional change to impact phase.

    At the same time, if you want to know the equivalent position change for 1ns of clock jitter, that would be like moving your head 300 nanometers. The timing impact of jitter in the digital clock domain is imperceivably small. I explained in other places why the impact to distortion is also really small.

    I suggest you do a true double-blind test with your ethernet cable. If you can't differentiate them reliably blind to what is installed, it's probably another factor.
     
  5. Rolltide

    Rolltide Forum Resident

    Location:
    Vallejo, CA
    Remember, double blind tests are unreliable because it takes several weeks to detect the obvious differences. I mean, you can’t expect somebody to detect a few picoseconds of jitter in one session, right? Be reasonable.
     
    jimbutsu and Agitater like this.
  6. thebiglebowski

    thebiglebowski Forum Resident

    Location:
    Southampton, UK
    Jitter is 99% non existent in ethernet (TCP)
     
  7. jimbutsu

    jimbutsu WATCH YÖUR STEPPE

    Who ever said they weren't? I think you've gone so far around the loop trying to show us all how smart you are that you've shot clean around the other end and completely misread a lot of things from a lot of people at this point. Slow down...
     
  8. jeff kleinberg

    jeff kleinberg Senior Member

    Location:
    Ct
    Talking clock jitter, not data
     
  9. jeff kleinberg

    jeff kleinberg Senior Member

    Location:
    Ct
    You did!
     
    MGW likes this.
  10. jimbutsu

    jimbutsu WATCH YÖUR STEPPE

    Where? Quote it.
     
  11. Uglyversal

    Uglyversal Forum Resident

    Location:
    Sydney
    I believe there was a prior exchange to that too and Digital media was not how Jeff worded his statement.
     
    jeff kleinberg likes this.
  12. jeff kleinberg

    jeff kleinberg Senior Member

    Location:
    Ct
    I hear what I hear, I have zero doubts about the improvement, lets please keep it to a technical discussion. My increased enjoyment with listening to digital is self evident. I have sat and listened to $2500 Nordost Tyr speaker cables(biggest cable purchase at that point) and quickly sold them as the way they presented music was not enjoyable. Same with the Cardas Clears, I have no qualms with returning stuff I don't need. If anything I'm more likely get buyers remorse and feel guilty about dropping the cash. You are welcome to come listen for yourself anytime its safe, I have nothing to hide. Could you please answer the question about the soldering iron noise getting into the dac, interested in your thoughts? Also when listening to music if you were constantly moving your head an inch and a quarter in random directions, it might decrease the musical enjoyment. If moving a speaker by that much can change things, why cant the image shifting by the same distance?
     
  13. jeff kleinberg

    jeff kleinberg Senior Member

    Location:
    Ct
    I have 19 years of history here, got any evidence for your diss? Plenty of bits to peruse.
     
  14. elvisizer

    elvisizer Forum Resident

    Location:
    San Jose
    ethernet doesn't function at all like USB. USB cables can definitely make a difference, though that's usually more to do with the usb interfaces (and how USB data transmission was implemented in those devices) than the cable itself.
    other cables that can make a difference: anything analog, spdif.
    i haven't actually heard a difference yet from power cables but I could see them making a difference under the right circumstances.
     
  15. jimbutsu

    jimbutsu WATCH YÖUR STEPPE

    That's the point. I said digital media. I was talking about how digital media is stored and comparing the retrieval of it to retrieval of analog media (specifically a needle in a record groove). Never made any statements about whether a square wave manifested electrically in any way, shape, or form. I did say "It's debatable whether or not a square waveform is a 'physical waveform' vs. a 'digital waveform,'" but again in the context of media retrieval because the discussion was about the reasonability of comparing swapping Ethernet cables to changing a stylus. I'm sure we still disagree on that point, but it doesn't change what I wrote, or that it's been misrepresented to advance an imagined narrative.
     
  16. jimbutsu

    jimbutsu WATCH YÖUR STEPPE

    Can you cite where I said in any way that a digital signal was not an analog square wave? That'd be one...
     
  17. Uglyversal

    Uglyversal Forum Resident

    Location:
    Sydney
    As far as I remember the conversation I remember wasn't quite like that. You've said your post is misrepresented, I hope you are not insinuating I have because the full thing is there without editions.

    At the moment I am sufficiently busy in more productive tasks not to pay too much attention to this pointless thread but I will re-read it in the next few days and will comment if I feel there is any need.

    I hope you and Jeff have an enjoyable Christmas and manage to stay away from these Toxic people since you are part of only a handful of posters that have behaved in an acceptable manner. Cheers!
     
  18. blakep

    blakep Senior Member

    1) I appreciate that you have brought up the differences in power in the UK and North America and
    how that might relate to cable performance (particularly power cord performance in my view but that
    is another topic). I have made the same point in this forum before.

    2) I believe that the vast majority of the skeptics in not only this thread, but almost all cable threads, haven't experimented meaningfully with cables because a) they rarely talk about any cable comparisons they have done and b) many/most clearly state in these threads that they choose not to experiment because they know that "technically" the cables cannot make a difference. If they do, as I mentioned previously, I certainly appreciate their opinion and give credence to it if they seem genuine.

    3) I appreciate that you've gone through a dozen interconnects and settled on something which appears to be reasonably cost effective and that performs well for you.

    Don't mistake me for someone advocating the use of uber expensive cabling; we are actually not that much different in that I currently use speaker cables that retail at about 1/8th of the price of the retail price of the previous speaker cabling I was using.

    And DIY power cords that cost about $125 to put together that outperform a number of commercial power cords that I experimented with that retailed up to around the $500 price point. I recently loaned out a DIY power cable that consisted of the previous raw cable I was using with a less expensive version of the power connectors I'm currently using to a friend with a better system than mine and he felt that cable (which I feel is pretty significantly inferior to what I'm using now) equaled the performance of a $1200 commercial power cord in his system.

    So although I am clearly suggesting there are differences in performance in cables, I am certainly not suggesting that performance is strictly related to price. There is a lot of craziness in the audiophile cable market; I'm fully prepared to admit that. But I'm not prepared to admit that high quality cabling makes no difference and that there are no audible differences in cables products (let alone ones that I have never heard!) as that has clearly not been the case in my experience and in my system.

    4) I have a number of friends with much better systems than mine. I don't for a minute think that how a cable performs in my system is an indication of how that cable might perform in theirs. It might perform better or worse (for them) depending on their subjective listening preferences and system synergy.

    5) I fail to see how comparing stock Cat 5 to stock Cat 7 is really meaningful to this particular discussion but perhaps I'm on the wrong page on that one. It would seem to me to be like comparing two different stock computer USB cables from different manufacturers and saying that would give me an idea of how a USB cable designed specifically for audio use might compare.

    I see elvisizer has pointed out to me that USB functions very differently and can make a difference which technically may be so. That did not prevent dozens of posters in the USB cable thread, however, from being very insistent that they could not.

    So, once again, if I'm curious, all technical arguments/discussion aside, I will experiment myself. And I'll do that by listening as that has usually put me in the right direction. I don't have anything else to add or respond to really at this point so I'll simply bow out and enjoy the music.
     
    jeff kleinberg, MGW and elvisizer like this.
  19. jimbutsu

    jimbutsu WATCH YÖUR STEPPE

    1) This thread has taken so many twists and turns that I'd not be surprised to hear that the discussion is tough to follow. I wasn't saying that *you* misrepresented my words.
    2) It's summer where you are, here we have winter storm warnings. Go enjoy the weather and don't waste your life on this!
    3) Thanks, you too.
     
    BeSteVenn and Uglyversal like this.
  20. elvisizer

    elvisizer Forum Resident

    Location:
    San Jose
    People are often wrong online shoot it’s even happened to me
     
    Brother_Rael and blakep like this.
  21. elvisizer

    elvisizer Forum Resident

    Location:
    San Jose
    I think there are a couple of key concepts to keep in mind when trying to understand why techy people freak out about the claims for audiophile Ethernet cables.
    • Ethernet is the basis for most modern networks which means that it is a very well understood tech that is widely used successfully
    • Successful use of Ethernet means being able to reliably and consistently confirm that the network is transferring data unchanged, end to end. This involves all the steps to put that data from software onto the wire and back again. You can capture the packets and compute checksums, not guess. Tcp will not work if data is consistently being changed on the wire, the nics will see that packet checksums don’t match and will request retransmits until they do. If this happens to your audio steam the audio will stop playing once the data in the receiving buffer on the playback device has been exhausted.
    • If tone and timbre are changing on the wire then data is changing. Since the network works and it can’t if data is changing on the wire, we know this isn’t true.
    • If the change to tone and timbre is happening outside of tcp data transmission then we have to envision a problem being solved or prevented from affecting the audio data by the audiophile cable that is capable of changing only audio data to the point that it is audible to the human ear while somehow not affecting any other data transmitted over non audiophile Ethernet cables in the same networks. Because don’t forget all the other non-audio data on the network is being transferred intact and we know this because no one claims their computers didn’t work without an audiophile Ethernet cable.
    So that’s basically it- I don’t see how that can ever be possible. If it is I’ll apologize to everyone but ugleversal, he hurt my fee fees with the quote edit
     
    Last edited: Dec 23, 2020
  22. Claude Benshaul

    Claude Benshaul Forum Resident

    Just opening a window in my room will change the sound perception in a way that no cable, switch or router can ever hope to achieve. Therefore I'm thinking of starting a company that will manufacture high end networking audio components with each one fitted with a tiny glass window (quartz or sapphire will also be available as a paid upgrade) that can be opened on demand.

    The only issue I have that prevents me from going full head on commercial and take over the market in a storm are the cables. I'm still trying to figure out how to fit tiny portholes along the jacket. I believe it can be resolved but right now it's way above what I can design or test.
     
  23. jeff kleinberg

    jeff kleinberg Senior Member

    Location:
    Ct
    • If the change to tone and timbre is happening outside of tcp data transmission then we have to envision a problem being solved or prevented from affecting the audio data by the audiophile cable that is capable of changing only audio data to the point that it is audible to the human ear while somehow not affecting any other data transmitted over non audiophile Ethernet cables in the same networks. Because don’t forget all the other non-audio data on the network is being transferred intact and we know this because no one claims their computers didn’t work without an audiophile Ethernet cable.
    So that’s basically it- I don’t see how that can ever be possible. If it is I’ll apologize to everyone but ugleversal, he hurt my fee fees with the quote edit[/QUOTE]

    Experiments have shown noise can enter a dacthrough the ethernet cable, this is likely the source of sound improvement, not a bits issue. Hoping for more data and information to further delineate this. Also I am not sure it is really changes in tone or timbre I am hearing, more a change in spacial presentation. Each musical note seems to have a more defined beginning, middle and decay.
     
  24. MattHooper

    MattHooper Forum Resident

    Location:
    Canada
    Like a breath of fresh air :)

    ( I've used many different Ethernet cables for my sound-work and not once has the sound changed noticeably).
     
    jimbutsu likes this.
  25. elvisizer

    elvisizer Forum Resident

    Location:
    San Jose
    Well ANY change is a change to data or has to fit the 2nd possibility that I mentioned. I figured from your posts that you would think that the non-data transfer effect was responsible, and that's not a super crazy stance to take . . . I just don't agree, I don't think it's possible for that mechanism to affect audio the way it's claimed without affecting any other data. . . and I haven't seen any experiments that show this could be happening either. I'll keep an eye out for that though! always have to be open to the possibility of weirdness but I'm very very skeptical.
     
Thread Status:
Not open for further replies.

Share This Page

molar-endocrine