my ethernet cable?


Hi all-

Another beginner question. So I am planning to set up a device (either an Auralic Aries or a Sonos with Empirical Audio's Synchro-Mesh and Dynamo power supply) to stream Tidal. When I was having a TV installed last year, I had the guy pull an ethernet cable to my stereo cabinet with this purpose in mind. I thought I was being smart to ask for him to get and install cat 6 instead of 5e, which was what he usually uses.

Later I read that cat6 can be a lot trickier to install correctly, and sometimes its safer just to go with 5e. (and I note the adapter plate on the wall actually says 5e.) And then I read this article from Blue Jeans cable about how most of the cat6 cable on the market is crap.

http://www.bluejeanscable.com/articles/is-your-cat6-a-dog.htm

So all of this has me wondering if I should buy some cat6 from blue jeans (or even 5e given installation quirks with cat6?) and have this redone? Or is there a way to test the quality of the cable I have and know whether or not it is up to the task?

It seems a bit excessive to replace the cable, but on the other hand, that will cost a fraction of what an Aries streamer does, or a year's subscription to Tidal, so I am wondering if I shouldn't make sure my fundamentals are all solid before proceeding.

Any advice?

Thanks!

Margot
mcanaday
If you are intent on listening to them both, why don't you get a run of 5e and just try it out without having it installed in the wall? If you like it better, then have it strung properly.

I have used neither for audio purposes, but did just have 15 miles of cat6 installed in our new corporate HQ with some runs well over 300 feet, and not a single cable has had to be replaced or reterminated. The majority of the drops are used for business class workstations, but we do have some server class machines with persistent DB connections where packet loss has to be minimized.

If I was being asked this question specifically for a data network build I would recommend cat6. For audio, your mileage may vary.
Hi Margot,

Wow! The Blue Jeans article is certainly eye-opening, and everything it says strikes me as technically persuasive. But on the other hand keep in mind the following:

1)As long as the ethernet link conveys the bits reliably and consistently, the fact that the cable may not meet cat6 or even cat5e specs is irrelevant. By its nature data transmission via ethernet cannot affect sonics as a result of jitter, waveform degradation, or other such issues that can certainly have relevance in the cases of S/PDIF, AES/EBU, USB, etc., unless the problems with the ethernet link are so severe that dropouts occur. See the paragraph entitled "Jitter and Networked Audio" in this paper by Steve Nugent of Empirical Audio. (Scroll down to just past the middle of the page).

2)Regarding testing the quality of the cable, the one suggestion that occurs to me, if practical, would be to network two computers together such that the cable is included in the path between them, and transferring a number of large files back and forth from one to the other.

3)My perception has been that anecdotal reports of dropouts or other performance problems with wired ethernet connections in home environments seem to be exceedingly rare.

4)One or two members here whom I consider to be particularly credible have reported in past threads that in their systems a perceivable sonic benefit was realized by upgrading their ethernet cables from unshielded cat5 or 5e to shielded cat6. Presumably that resulted from the shielding providing a reduction in RFI or digital noise being radiated or coupled from the cable to unrelated circuit points or grounds within the system. That kind of effect, however, if it occurs at all, figures to be highly system dependent and placement dependent. If you do end up replacing the cable I would suggest asking Blue Jeans if they can supply a shielded variety, but my instinct would be to not go to the trouble of replacing the cable based on that consideration alone.

Best regards,
-- Al
Margot,

If you are worried about the installation of the cable you can hire a certified cable company to test the cable with a cable scanner. There are cable installers and then there are trained certified cable installers. A trained certified cable installer would have tested the cable and supplied you with a copy of the test results.

If the installer installed CAT6 cable and terminated it on CAT5e RG45 jacks then you basically have a CAT5e Ethernet cable.

Do's and don'ts of installing CAT5e and CAT6, CAT6a cables.

Here is a video on testing cables. Note there is a lot more to testing a cable than just checking for continuity.

With a good tester the test scan will fail, if during the installation of the cable the cable was kinked. Even if the installer caught the kink and straightened it out the cable can fail the test or pass but only marginally.

If the termination of the cable to the RG45 jack was not done properly, that will show up on the test. The wire twist of each pair must be maintained right up to the point of connection to the punch down connection on the jack. No more than a 1/2" of the twisted pairs of the finished termination should extend beyond the outer jacket. Improper termination of the cable to the jack can make a big difference on how the cable will test and preform.

So you can hire somebody to test the cable for you or first just try the cable and see how it performs. It may work just fine for your application.
Jim
I replaced a 75' run of cheap Monoprice Cat6 that goes from my office router, down into the basement, and up into the living room stereo system with a 50' run of Blue Jeans 6a for the purpose of getting the most out of my TIDAL subscription.

The difference was very easy to detect. I'd describe the sound as cleaner, purer, and better-defined.

I know it shouldn't matter, yada yada, a packet is a packet, there are no errors with ethernet, yada yada. But it does.

Jim and Al,

You guys are the best. Thanks so much. So it sounds like my first step is just to go ahead and as long as I don't hear drop outs, don't worry about it. If I do hear drop outs, then maybe I want to get some shielded cat6 from blue jeans cable, if possible. If I end up there, Jim, will most electricians know about how to terminate correctly etc, or should I try to track down a certified cable installer? with enormous thanks to you both,

Margot
Oh, one other follow up:

If I do end up having drop outs and needing to replace the cable, I assume I only have to replace the line that goes to my stereo? (ie: I can leave the line that runs to the computer and the line that runs to the TV as is.) Or will this affect the performance of the new cable I have installed as well?
Thanks! M.
Thanks for the nice words, Margot. Regarding your last question, correct me if I'm wrong but I'm envisioning that the cable in question goes between the stereo and a router or network switch or something comparable. If so, the other lines will not affect the performance of the new cable. Also, if I understand correctly those other lines can be presumed to be functioning in a reliable manner, or you would have already noticed problems with them. That being relevant, for example, if you were to play music from the computer to the stereo via the ethernet connections.

Best regards,
-- Al
Margot,

IF, you have problems and the problem turns out to be the Ethernet data cable, you do not need to hire an electrician. You need to find a good low voltage cable company in your area that has trained certified data cable installers. You also need to make sure the company has/uses the proper test equipment to certify the cable after the data cable has been installed and terminated with RJ45 jacks and or RJ45 plug ends. The company should be able to furnish you with a copy of the test certification.

LOL, I noticed in my earlier response to your posted message I typed RG45 jacks instead of RJ45 jacks. It should read RJ45, So much for proofreading, my bad.
Jim
Thanks, everybody, again!

Cymbop, I'm curious, was your Cat 6a shielded?

and Jim, is RJ45 the name of the jack for 6a as well as 5e?

Many thanks!

Margot
Just speculating, but it seems to me to be possible that even if both of the cables Cymbop compared were unshielded, differences in their characteristics may have resulted in differences in the coupling of RFI or digital noise from the cable into his system, resulting in the sonic differences he perceived. If so, I would not expect a great deal of consistency between the results he experienced and the results of a comparison between the same two cables as performed in different systems, or even in a system comprised of the same components but having a significantly different physical arrangement.

And I would expect that any relation between the improvement he experienced and the fact that the Blue Jeans cat6a cable is spec compliant, while the Monoprice cat6 cable may not have been, would be an indirect one at best. Spec compliance helps to assure that the cable will perform its function reliably, but does not have a direct relation to minimizing possible effects of the signals and digital noise being conducted by that cable on other cables or power cords or component circuitry which may be nearby.

Best regards,
-- Al
Margot, it doesn't look like the Bluejeans 6a is shielded, but my perception of the sonic improvements agrees with Al's hunch that they are somehow related to noise rejection. Also the shorter run means that it's not a scientific comparison.

As I've said before: Almarg for President 2016.
I agree about Al. Where do I send my campaign contribution? :)

Thanks so much, everyone. Much appreciated!
Margot & Cymbop, thank you most kindly. If truth be known, though, I would much prefer listening to my stereo system to dealing with the tribulations of serving in that office :-)

Best,
-- Al
I would also suggest taking a look at your network switch/router. Many sold are still 10/100 and if you want maximum network speed with the Auralic you would need a gigabit switch. The sonos only has a 10/100 port on it, so no big deal.

Cat5e and Cat6 can both handle gigabit if installed correctly. Cat6 installed incorrectly (or patch cable manufactured incorrectly) can be worse than a good job with cat5e, so who installs it is almost more important than which cable you choose.

Al's proposal for testing is exactly what I do for my home installs. When I paid professionals to do it for work, part of their deliverables were the test reports.

Also, a last bit of advice for anyone else doing this is to run an extra cable or two. You never know when you may need it. I only ran two to my living room and am kicking myself. You could imagine a tv, home theater, xbox, dvd player, and sonos all needing an Ethernet connection. You could always put another network switch in the living room, but then you are splitting your gigabit home run amongst all of those items, not to mention eating up one more electrical outlet. Best case scenario is that only one is actually doing any serious transfer of data at once, but you've still added one more item in between your source and destination.
Thanks, Kriskdf.

Will the router say 10/100 on it? And what is the designation for a router that is adequate for gigabit?

If I did this --keeping in mind Al's reservations!--how do I go about finding a good cable installer? Obviously the guy who did the TV didn't exactly know what he was doing (because he connected Cat 6 cable to Cat 5e jack. Sigh.)

As for the extra cable, so just having something there unconnected does not affect the functioning of the lines to stereo, tv etc?

Thanks!

Margot

I replaced a 75' run of cheap Monoprice Cat6 that goes from my office router, down into the basement, and up into the living room stereo system with a 50' run of Blue Jeans 6a for the purpose of getting the most out of my TIDAL subscription.

The difference was very easy to detect. I'd describe the sound as cleaner, purer, and better-defined.
01-14-15: Cymbop

I believe you when you say you could hear the difference between the two cables. Blue Jeans' uses Belden cable a highly respected data cable manufacture, Monoprice doesn't say whose cable they use. They just say it is pure stranded copper. The construction of the raw cable can make a difference how the cable will perform.

I would discount the differences in length of the two had anything to do with the reason why you heard a difference, but I would like to know what you did with the additional 25' of the 75' Monoprice cable. Just guessing you coiled up the excess 25' of cable at the switch end. If that is the case, the next question would be how tight, rough physical diameter of the coil of the wire? I have found, and don't know why, that can cause errors on a PC connected to a network in the case of horizontal installed LAN cable. Testing the cable that has the excess cable coiled up in a small diameter coil with a scanner might only pass marginally. Increasing the diameter of the coil would increase the test results of the cable.

Best practices say two data cables should be installed at every work station. If you were to compare the certification test reports of a work station’s two data cables that were installed at the same time, same cable manufacture, terminated on the jacks on both ends by the same person, the cables will not test out the same. One will test better than the other. And sometimes the differences can be quite a lot with both cables still passing the scanner test.
.
Margot,

IF, you have problems and it turns out to be the cable you had installed, you could always just buy a readymade to order cable from Blue Jeans Cable and you or someone help you install it, providing the majority of the pathway is not concealed making it difficult for you to install the cable. If you do install the cable yourself be careful and protect the RJ45 plug on the end that is being installed, pulled, to the desired location. Be careful do not kink the cable. Do not step or walk on the cable as this can damage the lay of the 4 twisted pairs inside the jacket of the cable.

Pros of the premade Blue Jeans' cable?
You can save a lot of money.
You won't need additional patch cords, also eliminating two more connections.

Down side?

Physical damage to the cable at either end and or the RJ45 plug. When jacks are used on each end of a physically protected installed cable, patch cords that plug into the jacks can easily be replaced.
01-17-15: Mcanaday
Will the router say 10/100 on it? And what is the designation for a router that is adequate for gigabit?
10/100/1000 is a common designation for a router capable of gigabit speeds (i.e., 1000 megabits per second). If there is no indication on the router as to whether it is a 10/100 or 10/100/1000 type, Googling its make and model number plus the word "specifications" should lead you to the answer.

100 mbps is plenty for audio purposes, though. And if the router is 10/100 you can always upgrade it at a later time if gigabit speeds become desirable for other reasons.

BTW, I would not rule out the possibility that there could be at least slight sonic differences between interfacing to the Auralic at 100 mbps and at 1000 mbps, due to differences in coupling of digital noise to points in the system that are downstream of the ethernet interface. But if so, I doubt that which speed would be better in that respect can be predicted.
As for the extra cable, so just having something there unconnected does not affect the functioning of the lines to stereo, tv etc?
I'd be very surprised if having an unused cable connected to its own port on the router or network switch would have any effect on the others. Even if it did, though, perhaps as a consequence of poor design of the router or network switch, you could of course simply disconnect it.

Best,
-- Al
Thanks again, everybody! A thread about like this one here on Audiogon walked me through getting dedicated circuits set up last summer. Can't imagine taking a step without you guys.

Margot
I noticed that there are now cay 7 shielded cable available. Would that be an upgrade?
I noticed that there are now cat 7 shielded cable available. Would that be an upgrade?
Good question, Ozzy. I suppose that cat7 could sometimes provide a sonic benefit as a consequence of the double shielding it usually incorporates, which may reduce coupling of RFI and digital noise from the cable into the system. But I would expect any such benefit to be highly system dependent, and probably also placement dependent, with little or no predictability.

Regards,
-- Al
This may be of interest.
Ethernet LAN cable certification testing.
.
Oh, c'mon, Jea! It's just zeroes and ones -- why overthink it like that? ;-)

To answer your question, the extra 25' of my Monoprice run was in a jumbled mess in the basement. Perhaps not as detrimental as a coil, but the music is better now for sure with the neater, shorter, Blue-Jeans-ier run.

Oh, c'mon, Jea! It's just zeroes and ones -- why overthink it like that? ;-)
01-19-15: Cymbop
LOL, but unrestricted 1s and 0s are happy 1s and 0s, and therefore more productive.

Copper Cabling Troubleshooting Handbook - Fluke Networks.
.