I know that feeling. For me, it’s not 100%, but there were definitely times in my youth, when I first got into hifi, that the fun trumped the SQ. Now, as good as the SQ is, some of that fun is more ephemeral. I remember the joy of saving enough money for a few new LPs, and listening to them over and over. Now, I can play almost anything anytime anywhere. As good as my systems are, I probably still have the most fun with music in the car. The SQ at home is amazing. The music in the car sounds very good, though not as good, but I listen just for fun.
You know that old joke between musicians about audiophiles: “ordinary people listen to your music with their equipment; audiophiles listen to their equipment with your music.”It’s beguiling.
This! My latest car came standard with a fancy sound system and it sounds damn good. Special bonus - when I play EDM at full volume and bury the throttle it’s like a Fast and Furious movie!
It’s also an old canard rolled out to make fun of audiophiles that has little to do with the truth.
Yes, there are some that do as stated but that’s no more the norm than people with ears who actually believe all CD players sound the same and you should buy on features.
Well, it may be “crude” to you, but along with multi-tone tests and other current measurement techniques, it’s still able to very easily debunk every audiophoolery out there.
Debunk as in prove objectively beyond any doubt - we can measure galaxies 33 billion light years away, and the mass of the Higgs Boson. We can certainly measure the sounds from your HiFi. The only thing we can’t measure is the Audiophile’s vivid imagination.
In any case, we digress. Would be good to see some real new commentary about the topic at hand - comparing Ethernet Switches - rather than useless generalities over and over again.
The thing is, we don’t understand our sense of hearing any more than we understand any of our other sensors.
We can come up with digital sensors to do just about everything except reproduce the signals any of our organs actually send to the brain, so there’s an incredible amount we simply don’t know.
We can only “hear” down to 20 Hz, but why do sounds below 20 Hz affect our perception of space? Is it purely a sensory by feel nature, or is there something else? We don’t know, and no one’s funding the research.
We can predict some aspects of human emotion, but we can’t predict emotion. Despite their being innumerable medications on the market, we don’t understand how depression works, or how to kill cancer cells without also doing a good job of killing the patient by blasting both with radiation.
We can measure the mass of the Higgs Boson, but is light a wave or particle? “It depends on if it’s being measured and how.”
Those for whom measurements mean ALL can’t be accepting of much of quantum mechanics.
The bottom line is we’ve become very skilled at measuring things we know how to measure, but not as skilled at measuring things for which measurements do not tell all, and it’s clear what audio analyzers measure between 20 Hz - 20 KHz are not all that goes into how we hear and what affects our perception of soundstage and our sense of space and reproduction of it.
Measurements mean a lot, but they haven’t explained why some gear that measures extremely well sounds horrible and other gear that doesn’t measure all that well sounds absolutely heavenly. .
For many other sensory pleasures of life, how can they be measured? Can you measure that one vintage wine is better than another, or one painting is better than the rest?
Exactly. Attempting to measure subjective responses is pretty useless IMHO.
In any case, bringing the discussion back to the topic of audible impact of Ethernet Switches… I think fundamentally there are really only 2 questions;
Whether or not noise and/or jitter emanating from Ethernet Switches can somehow make its way to, or otherwise impact, the analog output of a particular Streamer/DAC, such that it’s audible.
Do the results of 1. from one particular Streamer/DAC automatically apply to every other Streamer/DAC?
The tests conducted by AA addresses 1., but without measuring the DAC’s output, it’s really half-done. Other than AA I’ve only seen two other attempts tackle this objectively.
A German audio forum; but similar to AA, they measured only in the digital domain, in this case the Ethernet Switch’s output. While clear differences in noise/jitter were seen, there was no measurement extended to the Streamer/DAC stage, so any audible impact from the measurements was conjecture only. Nevertheless it’s got some interesting conclusions, including debunking external clocking improvements of Ethernet Switches - specifically the Uptone EtherREGEN showing worse results from an external clock. Have a read (it’s in German but works well wth Google Translate): https://www.open-end-music.com/forum/privatforen/thomas-michael-rudolph-tmr/651284-messungen-von-ethernet-infrastruktur-switches-nur-lesen
And the other site was ASR/Amir, where he did a pretty good job of testing two “Audiophile Ethernet Switches” versus normal ones, and all the way to the DAC outputs: https://www.youtube.com/watch?v=BHPwPRLxDWc
As for 2. like practically every piece of equipment, I’m 100% certain YMMV applies
Imperfect comparisons at best. But they illustrate an interesting point. We can both taste the identical wine and reach different (subjective) conclusions. And we can recognize that some people are able to discern (accurately) what grapes and yeast went into its making, while others are clueless but still “know what they like.” We can both see the same painting and reach different (subjective) conclusions. And again, some people can instantly identify the artist, period, and even paint pigments used, while others cannot. And we can both listen to the very same audio system and reach different subjective and objective conclusions about it. All perfectly natural.
But when one changes an element of the evaluation, e.g., switches from Riedel glassware to another brand, and purports to taste the difference in the wine, then a very different perception process has happened. And that process is known to be subject to human cognitive bias. If one doesn’t take steps to remove or blunt that bias, one cannot know with any degree of certainty if one is actually tasting a difference or not.
Again, this is a hobby, and I fully support the tweaky nature of our hobby(itself a source of pleasure for many independent of the music), and every person’s right to spend their money how they see fit. More importantly, I believe each of us hears what we say we hear. My listening auditions of dCS equipment had no ability nor made any effort to remove cognitive bias. My addition of Vivaldi Clock, and then the Kronos1 Reference Clock didn’t even involve comparative listening until after the fact. My lengthy evaluation of Black Cat clock cables could not completely remove cognitive bias, but I made an effort to blunt it. But in each case, I know what I heard—or didn’t.
This caught my eye. I read through the evaluation and also followed the links to the Neutron Star 3 Reference Clock. I suspect there might have been howls of protest from the EtherREGEN fan club about the “quality” of the NS3 clock, and perhaps justifiably so. I’m not here to disparage anyone’s efforts, but the NS3 product page did not inspire great confidence in the product. In either case, the Open End Music tests don’t surprise me. Although I did connect my EtherREGEN to my Kronos reference clock at one point, I did it for fun, and could discern absolutely no change in SQ of my system with or without the connection (or with or without the ER, for that matter).
And to your point #2 Anup, we’ve learned from James’ detailed explanations that the type of reference clock will have an impact on whether it is effective as a reference for dCS master clocks; I suppose the same could be possible for something like an EtherREGEN or other switch. Perhaps the NS3 was the wrong type of reference clock? But wrong or right type, my interpretation of this data is that it underscores the absence of any evidence that signal or SQ at the DAC output is affected by these upstream devices.
Actually, it makes sense; I posted about the EtherREGEN’s External Clocking circuitry awhile back. Here’s the key paragraph;
Well, the EtherREGEN uses the Microchip Gigabit Ethernet Switch chip (KS9897S), which requires a 25MHz clock, not 10MHz. So, that ultra precise external 10MHz clock input (that some folks connect to a multi-$1,000 Rubidium Atomic Clock), has to get multiplied/divided by a $20 Silicon Labs clock generator chip (Si5340B), dramatically negating any supposed “phase-noise reduction” properties of the EtherREGEN
In other words, regardless of whether it’s the NS3, or even a PERF10, any external 10MHz clock will be “bottlenecked” by [that] Silicon Labs clock divider inside the EtherREGEN, possibly resulting in a worse performance.
Not to mention this whole notion of re-clocking when the EtherREGEN already has a perfectly competent ultra-low phase-noise Crystek CCHD-5575-25 (the internal 25MHz Crystal) is totally nuts.
You are commenting on an article you haven’t fully read. You haven’t even seen any measurements we did. Thats strikes me to be honest. Also: you don’t know me or my background. So how can you say anything about credibility? I have Julian Dunn’s book about jitter in the office. And yes: I read it. I just reference Hans because it’s easy to understand for a lot of people.
And about understanding TCP/IP and other network protocols…well I have the CCNA-books in the office. And yes: I read them fully. Not sure why I am defending myself actually.
I have done months of research on this topic and discussed it with several experts. There are still a lot of questions. But fact is: I can measure differences in clock jitter when swapping out switches. It’s constant and repeatable. And no: I didn’t use a simple scope to measure this. I have actually bought a Wavecrest SIA-3000 to do this.
Anyway. Believe what you want… I don’t care. But please: don’t judge people you don’t know. Not everything is what it seems.
I’m not sure that actually is a copyright violation, though I suppose it could depend on specific jurisdiction. But no matter; In the interest of propriety, here is a link to the setup guide.
I already deleted the post about copyright, because he included the link to the website. So that’s fine… He also just could have posted a link. It’s easier and safer.
But to be clear: creating a PDF of an article and posting it on a forum is actualy copyright violation, because he didn’t ask. That’s how the law in the Netherlands works.
As I said: jurisdiction specific. Here, it does not look like he created the PDF. He linked it. Can I respectfully suggest something? Instead of playing forum cop, why not privately message Erno and the admins and inquire first? Have a great day!
and what fun that was…
.