More than once we experienced CAT 6/7 ethernet cables to be of non copper material, we even found iron cables that were magnetic. This happens even in professional environments where high quality and high reliability is demanded. Fortunately there normally is a Fluke ethernet cable testing device available which quickly shows the cable to be not upto specs. When a supplier delivers such crap there will be a meeting where it will be stipulated that such quality will not be accepted in the future and that we can choose another supplier. This keep the supplier and its competitors sharp.
Believing is done in church
There is no way self crimped cables (even with the best connectors money can buy) will be able to compete with robot produced and professionally tested/certified cables. Of course this is audio where many opinions exist that contradict science. In the audio area clever people invent variations on a theme (that no industrial oriented person would even consider) and sell them in numbers with a large markup which is the best measuring parameter.
The issue with ethernet is that the absolutely worst but cheapest connectors were chosen by design. There is no easy way to overcome that, we have to deal with that. Quality cables by reputable manufacturers have good connectors. Self crimped cables with Telegärtner connectors will measure worse in general when done by DIYers.
Believing is done in church
There is no way self crimped cables (even with the best connectors money can buy) will be able to compete with robot produced and professionally tested/certified cables. Of course this is audio where many opinions exist that contradict science. In the audio area clever people invent variations on a theme (that no industrial oriented person would even consider) and sell them in numbers with a large markup which is the best measuring parameter.The issue with ethernet is that the absolutely worst but cheapest connectors were chosen by design. There is no easy way to overcome that, we have to deal with that. Quality cables by reputable manufacturers have good connectors. Self crimped cables with Telegärtner connectors will measure worse in general when done by DIYers.
Last edited:
My two cents as a sound engineer:
Ethernet cables won't really affect your sound unless the devices connected to them are vulnerable to it. That excludes running cables so long that you exceed the ethernet standards.
Just like with USB dacs, dirty power will do a number. Inductive coupling sucks.
Just like with power adapters and sockets, ground loops can do one too.
Just like with preamps, DC offset outside the limits of the transceiver can cause it to misbehave (and have service issues.)
None of this is truly the fault of a bad cable (in most cases) but more so user error. Good equipment design and mindful use of it will eliminate most of your issues.
Again: an ethernet cable will not affect your sound quality beyond digital signal integrity loss. The latter will only result in interrupted service or very audible quantization (as the device tries to work with the tools it's got.)
You do not lose or gain harmonic distortion/channel separation (the "warmth", "depth", etc) with different cables; that's exactly the point of digital transfers and is why we want as much of our stuff to be digital in this modern world.
Ethernet cables won't really affect your sound unless the devices connected to them are vulnerable to it. That excludes running cables so long that you exceed the ethernet standards.
Just like with USB dacs, dirty power will do a number. Inductive coupling sucks.
Just like with power adapters and sockets, ground loops can do one too.
Just like with preamps, DC offset outside the limits of the transceiver can cause it to misbehave (and have service issues.)
None of this is truly the fault of a bad cable (in most cases) but more so user error. Good equipment design and mindful use of it will eliminate most of your issues.
Again: an ethernet cable will not affect your sound quality beyond digital signal integrity loss. The latter will only result in interrupted service or very audible quantization (as the device tries to work with the tools it's got.)
You do not lose or gain harmonic distortion/channel separation (the "warmth", "depth", etc) with different cables; that's exactly the point of digital transfers and is why we want as much of our stuff to be digital in this modern world.
Instead of invoking "I believe there are differences..." get some data to support your claim. Here's an idea on that:
Take your "best" and "worse" ethernet cables and connect two computers however you want (e.g. via router or PTP). Now get yourself a large audio file and transfer it from computer A to computer B. Repeat with each cable, renaming the file each time. Now repeat this process, so each renamed file is transferred from computer B back to computer A using the corresponding cable that was used for the A to B transfer. Now the original and transferred/corrupted files are on the computer where the original file started and they will have been sent over each cable twice. Next, perform a bitwise comparison between the original and the transferred copies to see how many, if any, differences you get and post the results back here.
Without performing some actual tests like these that support your claims, your "I believe" is just not believable!
Take your "best" and "worse" ethernet cables and connect two computers however you want (e.g. via router or PTP). Now get yourself a large audio file and transfer it from computer A to computer B. Repeat with each cable, renaming the file each time. Now repeat this process, so each renamed file is transferred from computer B back to computer A using the corresponding cable that was used for the A to B transfer. Now the original and transferred/corrupted files are on the computer where the original file started and they will have been sent over each cable twice. Next, perform a bitwise comparison between the original and the transferred copies to see how many, if any, differences you get and post the results back here.
Without performing some actual tests like these that support your claims, your "I believe" is just not believable!
Last edited:
The best possible test would be invoke multiple mixed file transfers in each direction simultaneously until you break the connection, and then test each cable ten times to see if they actually make a difference to when they break.
I've not been able to observe manipulation of even a single bit while transferring hundreds of gigabytes of data over ethernet. Often I will format a drive on a Linux box and fill it from a Windows share, come back 10 hours later while the drive is still filling, no dropoff in speed and not a single byte lost.
However, and I say this with a caveat, one specific cable I use is sixty feet long and ultra-slim format to fit in doorjambs, and it does show a slight reduction in speed due to dropped packets. Because of the way ethernet transmission works, the final data is still 100% accurate, just takes a small amount longer.
I've not been able to observe manipulation of even a single bit while transferring hundreds of gigabytes of data over ethernet. Often I will format a drive on a Linux box and fill it from a Windows share, come back 10 hours later while the drive is still filling, no dropoff in speed and not a single byte lost.
However, and I say this with a caveat, one specific cable I use is sixty feet long and ultra-slim format to fit in doorjambs, and it does show a slight reduction in speed due to dropped packets. Because of the way ethernet transmission works, the final data is still 100% accurate, just takes a small amount longer.
Unless your computer is cheating, a file transfer should not change the bits or correct them if they are. dd with cs has never ever yielded me a file that is not the same. That is not how the tech works.
There are differences in network cards and the ones built in on Asus boards from the latter years have been known to be issue makers with odd errors and timeouts and not cooperating.
And this is in the end the only real world scenario you will find: Dropped packets. There will never be wrong saved bytes or bits. If it cannot get a transfer of the bits it wants then it will just not complete. If there was any way that it could then I know we would not be using this technology anymore. Datacenters would go boobs up all the time.
There are differences in network cards and the ones built in on Asus boards from the latter years have been known to be issue makers with odd errors and timeouts and not cooperating.
And this is in the end the only real world scenario you will find: Dropped packets. There will never be wrong saved bytes or bits. If it cannot get a transfer of the bits it wants then it will just not complete. If there was any way that it could then I know we would not be using this technology anymore. Datacenters would go boobs up all the time.
Last edited:
There is no mechanism that can lead to sound quality changes in an ethernet transmission. You can have dropouts, yes, but not 'signal changes' in the analog audio reproduction of a data file being retrieved.
Jitter and noise are very well-understood concepts in this domain, are not bandied about with snake oil liniments, and are handled professionally and properly.
If you try to send an actual analog audio signal down multiple twisted pair lines (as was the fashion with DIY speaker cables a decade ago and yes, I admit, I've been guilty of trying them) then all bets are off the table.
However, that's not what we're talking about.
Jitter and noise are very well-understood concepts in this domain, are not bandied about with snake oil liniments, and are handled professionally and properly.
If you try to send an actual analog audio signal down multiple twisted pair lines (as was the fashion with DIY speaker cables a decade ago and yes, I admit, I've been guilty of trying them) then all bets are off the table.
However, that's not what we're talking about.
And this is also exactly why I said most issues would be encountered because of equipment not being up to the task, like insufficient resistance to noise or an oversight causing a ground loop. Stuff that is absolutely not the fault of the cable but rather the layout of currents
My limited experience with ethernet cables is that it's all or nothing. And the nothing is probably due to a wire coming loose in the connector.
On a side note, here's a fun way to waste an hour - try to troubleshoot your non-working router, switch, AP's configuration only to find out it's a bad cable.
On a side note, here's a fun way to waste an hour - try to troubleshoot your non-working router, switch, AP's configuration only to find out it's a bad cable.