The Great Ethernet Showdown

This is a debate I really struggle to understand.  Card on the table, I am firmly in the digital (especially Ethernet) cables do not make a difference camp.

What I don't get is this:

People are convinced they hear a difference.  I believe them - how could I do otherwise.  They have conducted their 'test' and they hear the change.  Others explain the engineering and say there can be no difference - look I can prove it.  I believe them too.

Where is gets strange for me is that there are really only two possibilities here. The science is wrong (or incomplete) or there is no difference. Given that there is plenty of research on how our senses (and by extension our brain) can fool us completely and given that in non-hifi applications the question of cable quality (so long as standards are met) is irrelevant then I tend to side with the no difference brigade.

It is understandable (and science can and does explain this too) why people assume that more money = better and are quite willing to spend extra and convince themselves post purchase of the benefit - confirmation bias, it's interesting stuff.

This does not mean that the science should not be challenged - it might be wrong.  There might be something going on that is unexplained that is peculiar to hi-fi data transmission and no other form of data.  I would ask however, what is the likelihood of this?

I have been told before that this is a closed minded approach. Maybe, but is it not equally closed minded to shut out the possibility that it is our perception that shows a difference rather than the reality?

Finally - for those that 'hear a difference' and are happy with their purchases and the money they have spent. I am genuinely happy for you. I make my choices based on my experience and understanding and it is right and proper that you make yours based on your own experiences.

 

 

French Rooster posted:
Beachcomber posted:

Believing your ears is probably less reliable than believing your eyes.  And as many optical illusions have shown, that is not always a good thing to do.  If films, photographs, documents - data of all sorts - can get transmitted over Ethernet without any data loss or corruption, why is audio magically different?  Then consider that the whole effect of HiFi is to fool our ears/brain into thinking something is there when it isn't (I see no orchestra/group or whatever in front of me when I listen to my system, anyway), it is easy to suppose that that same ear/brain construct can imagine all sorts of things, including changes in the sound that don't exist.  Also consider the fact that the same music, in the same room played at the same settings but on different days (or even a different time of day) can sound different, depending on mood/lighting/last meal/argument with SWMBO.  So on the one hand we have knowledge that the data that gets through one cable is exactly the same as through another (you can subtract one from the other and get zero), and on the other we have a very flaky biological system that is subject to all sorts of influences and effects that have nothing to do with the immediate data stream.  I suspect the problem is in the latter.  Placebo effects, if you will.  

And I suspect that 'burn-in' is largely or possibly entirely a similar effect.  After a while you get used to the new sound of the new equipment and that is when it has burned in.  It is possible, I suppose, that the values of some components change a little over time.  Whether that is enough to explain huge differences in the resultant sound I'm not sure.

if hearing improvement in sound quality with better lan cables is placebo effect,  it is not possible to trust anymore our ears?   so the improvement we hear going from one amp or source to better ones is also a placebo?

From cheap lans to audioquest vodka for example, the improvement is not subtle but clear and immediate.    Try by yourself, you will be surprised, with a blind test with a friend.

One amp to another can clearly make a difference particularly if its more powerful or better designed. The difference between a 180 and 250 is obvious, and probably measurable. Just like the difference between a 1 litre and 3 litre car. How/why you are then translating this to mean an ethernet cable makes such a difference is baffling.

 

If you think that an ethernet cable makes a difference then by all means spend £2k on one, this is an age old argument, its only money and once the desire takes man, he will have his shiny shiny.

 There will be much more knowledgeable people in the forum who will be able to explain the difference. However it is relatively simple to do an A-B Test at your local dealer. So why not making that blind test.

if you are convinced of the difference fine, if not fine.

 

I am obviously in the camp who believes it makes a difference and that’s why I invested.

garyi posted:
French Rooster posted:
Beachcomber posted:

Believing your ears is probably less reliable than believing your eyes.  And as many optical illusions have shown, that is not always a good thing to do.  If films, photographs, documents - data of all sorts - can get transmitted over Ethernet without any data loss or corruption, why is audio magically different?  Then consider that the whole effect of HiFi is to fool our ears/brain into thinking something is there when it isn't (I see no orchestra/group or whatever in front of me when I listen to my system, anyway), it is easy to suppose that that same ear/brain construct can imagine all sorts of things, including changes in the sound that don't exist.  Also consider the fact that the same music, in the same room played at the same settings but on different days (or even a different time of day) can sound different, depending on mood/lighting/last meal/argument with SWMBO.  So on the one hand we have knowledge that the data that gets through one cable is exactly the same as through another (you can subtract one from the other and get zero), and on the other we have a very flaky biological system that is subject to all sorts of influences and effects that have nothing to do with the immediate data stream.  I suspect the problem is in the latter.  Placebo effects, if you will.  

And I suspect that 'burn-in' is largely or possibly entirely a similar effect.  After a while you get used to the new sound of the new equipment and that is when it has burned in.  It is possible, I suppose, that the values of some components change a little over time.  Whether that is enough to explain huge differences in the resultant sound I'm not sure.

if hearing improvement in sound quality with better lan cables is placebo effect,  it is not possible to trust anymore our ears?   so the improvement we hear going from one amp or source to better ones is also a placebo?

From cheap lans to audioquest vodka for example, the improvement is not subtle but clear and immediate.    Try by yourself, you will be surprised, with a blind test with a friend.

One amp to another can clearly make a difference particularly if its more powerful or better designed. The difference between a 180 and 250 is obvious, and probably measurable. Just like the difference between a 1 litre and 3 litre car. How/why you are then translating this to mean an ethernet cable makes such a difference is baffling.

 

If you think that an ethernet cable makes a difference then by all means spend £2k on one, this is an age old argument, its only money and once the desire takes man, he will have his shiny shiny.

i am like St Thomas, i believe in what i hear .  As for 2k for 1m, it is another story. All depends of how good and resolving a system is.   i would not buy 2k lan for a sonos connect.

Thats fine as long as your evenings arn't spent listening to female vocals to check your hifi sounds great then I think we are all on the same page!

As I say, men must spend their money 'tis the way of things and always has been, there is no doubt a forum somewhere about motor bikes where a man is about to drop 2k on a shiny pedal because he believes it makes his bike faster.

if you take a audiophile lan cable, there is the bare wire, the sleeve which insulates, then the outer sleeve, then the connectors, the plastic which houses them, the metal used in them etc.

the bare wire itself probably has multiple dynamics : the purity of the metal, the effect of mixing it with another metal, the particular ratio involved in that, having the cable in strands of specific thickness, twisting it in particular ways. one particular formula of this probably works better than others. but to get the formula requires interest, time, effort, investment, risk and a lot of experimentation and talent.

And likewise the experimentation involved in the selection of the specific material for insulation, resistance, damping etc in the manufacture of the sleeve. likewise the metal used in the connectors. specific micro measurements to make sure it fits snugly in the socket.

Do you think that the $2 lan cable has such an investment in time, effort, talent and money?

vtpcnk posted:

if you take a audiophile lan cable, there is the bare wire, the sleeve which insulates, then the outer sleeve, then the connectors, the plastic which houses them, the metal used in them etc.

the bare wire itself probably has multiple dynamics : the purity of the metal, the effect of mixing it with another metal, the particular ratio involved in that, having the cable in strands of specific thickness, twisting it in particular ways. one particular formula of this probably works better than others. but to get the formula requires interest, time, effort, investment, risk and a lot of experimentation and talent.

And likewise the experimentation involved in the selection of the specific material for insulation, resistance, damping etc in the manufacture of the sleeve. likewise the metal used in the connectors. specific micro measurements to make sure it fits snugly in the socket.

Do you think that the $2 lan cable has such an investment in time, effort, talent and money?

A certified, standards compliant  $2 LAN cable has much more lab investment behind it, than chord or audioquest could ever spend.  The engineering is backed by AT&T, Bell, Cisco,  et al .. the consortiums have very deep pockets... and much more riding on getting it right! 

Recommended 'curated' reading  This took me a while, so, well.. you know.. 

1. A file on your hard disk - Wikipedia: Digital recording (see 2,3,4.1,and 4.2)

2. A file from a wave - Wikipedia: Analog-to-digital converter (intro, 1.1, 4.1 and 5)

3. How many slices? - Wikipedia: Pulse-code modulation (see intro, 2,3,4,5, and 8)

4. Significant digits - Wikipedia: Audio bit depth (see most of it: 1,2,2.1,5,6)

4. If You build it, they will come - the OSI model - Wikipedia: OSI model (see intro, 2, 2.1, and 2.2)

5 . Bits in a line - Wikipedia: Frame

6. Frames in an envelope - Wikipedia: Network Packet (see intro, 2, 2.2)

7. No speeding! -  Wikipedia: Symbol Rate (see "Line codes for baseband transmission" 1.3)

8. Failure is not an option - Wikipedia: Error detection and correction (see 3,4,5 (all), 6, 6.1 and 7.1)

9. Pragmatic metallurgy - The Physical Layer 100BaseT standard - Wikipedia: Fast Ethernet (see 100BaseTX 3.1)

10. In through the Out door -  Wikipedia: Network interface controller (see 1,2)

11. Less Work = Less Pay - Wikipedia: Sampling (signal processing) - (see Intro, 3.1, and 3.1.1)

Good Luck!

 

So I started this thread and I don't have the technical knowledge to argue with some of the points above.  In fact from what is being said there my not be a rational technical argument to be made.

I would just urge doubters and believers alike to listen to Chord Sarum T against standard CAT5 in the context of a half decent system and for the doubters to maintain that the differences I heard were a placebo effect.  I don't really believe for a second that the differences I heard between the various other cables were imagined either but I suppose it's slightly less improbable.

If it was a case of hearing what I wanted to hear it would have been Music that was the 'wow moment' for me as that's what I've gone for with other cables.  Music was better again than Sarum T but it wasn't the huge improvement I heard between Signature and Sarum T.  Plus I'm not even about to buy an Ethernet cable although am lucky enough that I could afford to so I don't have any agenda and approached last Saturday with a genuinely open mind.

HungryHalibut was sceptical about my 'two black box upgrade' claim which was not meant absolutely literally.  Probably a very good single box upgrade might have been a more accurate claim.  My previous Naim upgrades include NAC82-552 (non DR), NAP180-300 (non DR) and CDX2 (admittedly with PS555 [non DR]) to CD555 with the same power supply, basically bypassing some very highly regarded equipment and I thought what I heard at the weekend compared favourably.

So will some other forum members please have a listen and report back.

Bryce, I don’t intend to try the Chord, but I’m totally convinced that these cables can make a difference. I’ve heard it myself and have bought the cables myself, albeit the cheaper Vodka. I’ve no idea why the doubters feel the need to wheel out technical references or to quote physicists and electronic engineers. All you need to go is listen. The doubters say that the believers are succumbing to autosuggestion and that there is no difference. One could also say that they are determined not to hear a difference and thereby convince themselves that there isn’t one. 

I’m not that bothered about why what makes a difference, though I read once that it’s something about how the cables help to reject analogue components that can lead to an impact on sound quality. Few - other than Harbeth’s designers - insist that amplifiers make no difference, few insist that all speaker wires sound the same, and I wonder why people take such an entrenched position with ethernet cables. Just because one doesn’t understand things doesn’t mean they are not there. One doesn’t have to understand everything. 

Ive also tried various ethernet cables from nas to switch and from switch to nds. They do make a difference which is immediately identifiable, even with blind testing. I don’t have any technical idea why they sound different but they most definitely do.

I now use an audioquest cinnamon from nas to switch and a vodka from switch to nds. In my system using two vodkas made the sound very “overblown” and airless. A friend however uses two vodkas in his system and they work better than the cinnamon/vodka combination, which shows how system dependant the type/combination of ethernet cables are.

 

 

This topic has been done to death many times and the (engineering) reasons why different LAN cables and setups can make a difference on a device where digital to analogue conversion is taking place are understood - It's not magic or some strange science we don't yet understand or even human perception.

I'd still question the monetary value on some of the streaming cable pricing, but as always if it works for you then that's all that counts. I'd try experimenting with a decent Cat5 cable and add a few ferrites if you want a good bang for your buck.

Hungryhalibut posted:

Bryce, I don’t intend to try the Chord, but I’m totally convinced that these cables can make a difference. I’ve heard it myself and have bought the cables myself, albeit the cheaper Vodka. I’ve no idea why the doubters feel the need to wheel out technical references or to quote physicists and electronic engineers. All you need to go is listen. The doubters say that the believers are succumbing to autosuggestion and that there is no difference. One could also say that they are determined not to hear a difference and thereby convince themselves that there isn’t one. 

I’m not that bothered about why what makes a difference, though I read once that it’s something about how the cables help to reject analogue components that can lead to an impact on sound quality. Few - other than Harbeth’s designers - insist that amplifiers make no difference, few insist that all speaker wires sound the same, and I wonder why people take such an entrenched position with ethernet cables. Just because one doesn’t understand things doesn’t mean they are not there. One doesn’t have to understand everything. 

I don't use ethernet cabling in my main system (well, not to send music anywhere, just to control my Melco) but it does seem to me that even standard cables must pass the data without error. However, I have no doubt, given the credibility of the folks that notice a difference, that there are improvements to be gained, and therefore it must be something to do with interference being injected into sensitive analogue circuits.

Us old fogeys who've been hi-fi enthusiasts since before they started the roadworks on the M6 are well aware of the placebo effects that can apply in these situations, but when you're as tight-fisted as me you're not inclined to drop large lumps of wonga on something unless you've given it an extensive home trial and you're genuinely happy it's a worthwhile improvement. I'm always interested to read about the technical side of kit, but too many times in the past we've all been lured into believing "facts" about why something should work or not, and later discovered this just wasn't the case. "Perfect Sound Forever".

>A certified, standards compliant  $2 LAN cable has much more lab investment behind it, than chord or >audioquest could ever spend.  The engineering is backed by AT&T, Bell, Cisco,  et al .. the consortiums have very ?>deep pockets... and much more riding on getting it right! 

"certified, standards complaint"" yes - but not necessarily to audiophile standards. their emphasis in engineering is not towards the goals of audiophiles.

yea the consortiums have deep pockets - but they are not spending millions to sell lan cables for 2 dollars. they are not in business so that u and i can have a great music listening experience. can these consortiums afford to sell chord c-stream for 2 bucks given the cost of manufacture of such a product?

I don't consider myself an audiophile btw. but i do respect good quality cables. for usb and lan, i usually buy the baseline models from chord for like $100 or so. and i do think those cables are qualitatively better - in technology, in materials used, in manufacture - than their 2 dollar counterparts and do have a positive effect on the SQ according to my own ears. but i cannot comment on exotic highly priced cables - as i don't own any of those.

You can pick up superb ethernet cables for like £10. It would be interesting to understand what is so expensive or special with a cable you are being charged say £2k for. I think the whole premise of boutique cables is fine as long as you have sorted everything else in the chain first.

There wont be a single 1m length of ethernet cable that cost more than 50 quid to make, I'm sure, but then that can be said for so many things in so many hobbies, some fishing gear is eye watering.

garyi posted:

You can pick up superb ethernet cables for like £10. It would be interesting to understand what is so expensive or special with a cable you are being charged say £2k for. I think the whole premise of boutique cables is fine as long as you have sorted everything else in the chain first.

There wont be a single 1m length of ethernet cable that cost more than 50 quid to make, I'm sure, but then that can be said for so many things in so many hobbies, some fishing gear is eye watering.

as long as you have sorted everything else in the chain:  can you describe your system precisely, i am curious to know what you have sorted in your chain?

Excuse me where did I state all cables are the same?

I contend that if an ethernet cable is built to standard then it will do as intended. Like anything, dirt cheap ones are available that will fall to bits, however thats a problem of construction and in my opinion nothing to do with sound quality (until such time as it stops working of course). £10 ethernet cables from reputable companies (excel etc) do not fall to bits and work perfectly in their intended environment.

Rooster, I am cynical as I believe you can tell, however I do take an interest in my network topology because I find the subject interesting and I like to tinker. Therefore I have a custom router using a core i5 dell PC, that lives in the loft, running Untangle, and is connected to a 48 port switch which then feeds all rooms in the house. Four ports of this are in LAG to another switch down in my office, the rest just go to wall outlets, a mixture of cat5e and cat 6 is used in the walls.

8 ports goto the living room where I have various kit needing internet, including another core i5 pc (6th gen) with 8 gigs ram. This runs Jriver outputting via USB into DAC. There is also a 24in touch screen on the wall so that family members have easy access to music in a jukebox style.

Another PC I built houses a bunch of drives and runs OpenMediaVault, this is the primary server for music, videos etc to the hifi and via plex to any connected devices, this lives in the loft too.

3 wifi access points working in mesh cover the house, each is connected to the network with ethernet.

If money was no object I would probably build a custom PC for the streamer, perhaps with a passive cooler for the CPU, not that you can hear it now, but just because that would be cool to do.

I have heard hifi systems much much better than mine. Better amps, better sources and better speakers, but I am happy with mine as is.

One thing I can confirm, of all the various ethernet cables I have, never once have I discovered one sounding 'better' than the other, a cheap one claiming to be cat5e would not do gig though, so always good to spend a bit more than rock bottom.

garyi posted:

Excuse me where did I state all cables are the same?

I contend that if an ethernet cable is built to standard then it will do as intended. Like anything, dirt cheap ones are available that will fall to bits, however thats a problem of construction and in my opinion nothing to do with sound quality (until such time as it stops working of course). £10 ethernet cables from reputable companies (excel etc) do not fall to bits and work perfectly in their intended environment.

Rooster, I am cynical as I believe you can tell, however I do take an interest in my network topology because I find the subject interesting and I like to tinker. Therefore I have a custom router using a core i5 dell PC, that lives in the loft, running Untangle, and is connected to a 48 port switch which then feeds all rooms in the house. Four ports of this are in LAG to another switch down in my office, the rest just go to wall outlets, a mixture of cat5e and cat 6 is used in the walls.

8 ports goto the living room where I have various kit needing internet, including another core i5 pc (6th gen) with 8 gigs ram. This runs Jriver outputting via USB into DAC. There is also a 24in touch screen on the wall so that family members have easy access to music in a jukebox style.

Another PC I built houses a bunch of drives and runs OpenMediaVault, this is the primary server for music, videos etc to the hifi and via plex to any connected devices, this lives in the loft too.

3 wifi access points working in mesh cover the house, each is connected to the network with ethernet.

If money was no object I would probably build a custom PC for the streamer, perhaps with a passive cooler for the CPU, not that you can hear it now, but just because that would be cool to do.

I have heard hifi systems much much better than mine. Better amps, better sources and better speakers, but I am happy with mine as is.

One thing I can confirm, of all the various ethernet cables I have, never once have I discovered one sounding 'better' than the other, a cheap one claiming to be cat5e would not do gig though, so always good to spend a bit more than rock bottom.

i think that in your case you have not the choice: you have a big house and you need long distance lan. Audiophile lans would cost you a lot of money.  I can easily understand that you don’t want to invest in that.

For me i have a 8 port cisco 2960 switch and use only 3 ports. I need only 3X 0,75 m lan for streaming.  It is not 10 or 20 meters as you probably.

Gary, just a point, a true Cat5e cable irrespective of price will support 1000BaseT (Gig) at least to 100 metres..... clearly counterfeit goods labelled as Cat ‘x’ that are not then all bets are off. But as you say buy Ethernet cables, and any cables for that matter, from a reputable trader, rather than some dodgy back street lockup..  

Add Reply

Likes (0)
×
×
×
×