Cat 7 Cable

Thad,

From what I've discussed, the integrity of the data transfer isn't the issue - the protocols take care of that.

Yes, you are right. We can regard the network as black box. What goes in one end comes out the other: the workings of the box ensure that --- unless some catastrophic failure takes place. There will always be some circumstances in which the thing may be too stressed to work at all. Fluorescent lights are a common cause of interference, and my idea about somebody wrapping a cable around one wasn't entirely a joke.

However, apparently there is some kind of hash in the audio that gets removed when using a good shielded cable.

asliarun deals with this far better than I can, and has done. What I feel is that us letter mortals (very much including me) should either accept the black box as something that does work, or we should get out the networking textbooks and find out what really happens in the box. I had to do a certain anount inside that box for my work but, somehow, I just don't find it very interesting any longer ;(. Well, ok, interesting, yes, but not very :lol:

I did exchange notes with someone today who confirmed that Cat6 is enough for streaming audio - so I might just do that to start with.
Should be able to get a good Cat 6 cable locally so costs would be reasonable.

You are probably (I'm guessing) going to connect devices with 100Mbs network devices. IIRC, even cat5[e?] goes to 1000Mbs. My Squeezebox Duet wireless device is much less than 100Mbs. It works --- but I cannot remember what the maximum native sample-rate is.

I usually fall into an embarrassing pit if I try to do stuff with numbers. Especially past bedtime ...or I'd calculate the time taken (in round numbers and ignoring overhead) to push a CD's worth of music across a network. But just now my brain is about as keen to do that as it is to work out sub-net boundaries :rolleyes: :lol:

asliarun. yes, I linked to the original BJ article, which I found via the audioholics site.





(4.20am... even owls have to sleep... Goodnight!)
 
Hi Nikhil,

Can you please clarify if you intend to use Cat 7 cable (heard it for the first time too!) to stream the digital signal (containing the music) or are you planning to pass analog audio (use it like speaker wire)?

'''

A small clarification: The hash you described above - it usually works a bit differently. Digital data (packets) being streamed are oblivious of the underlying wire. However, network protocols (at various levels or layers) have software mechanisms built in that detect packet (data) corruption and packet loss. What you can hash is (i think) CRC checksum - i.e. the network devices sending and receiving packets use a checksum algorithm ...


Hi Asliarun,

By "hash" I mean a sort of HF "dirt". Replacing the patch cable between with an S/FTP (Screened Foiled Twisted Pair) cable helps improve audio I am told.

I use a two PC setup running JRiver19 with my Control PC located on the other side of the room. I would like to use this cable to connect my Control PC and Audio PC via the router. At the moment the data is sent over wifi. I have been told that using S/FTP cable is much better.

Here's one vendor I am considering: http://www.vpi.us/cable-flt-cat7.html

Regards.
 
Last edited:
You are probably (I'm guessing) going to connect devices with 100Mbs network devices. IIRC, even cat5[e?] goes to 1000Mbs. My Squeezebox Duet wireless device is much less than 100Mbs. It works --- but I cannot remember what the maximum native sample-rate is.

Thad, the data capacity is trivial - what I am really interested in is the shielding - preferably a Screened Foiled Twisted Pair (S/FTP).
 
Thad, the data capacity is trivial - what I am really interested in is the shielding - preferably a Screened Foiled Twisted Pair (S/FTP).

What will you gain from the shielding? If it is transmission errors you are talking about then for detected errors a re-transmission occurs as per TCP/IP and unless you are using up the entire max bandwidth this should not be an issue. I cannot confirm though but still would like to believe at least for audio that will not make audio jittery what you are only limiting is the external magnetic / induction effect on transmission which the protocol is capable of handling anyway
 
What will you gain from the shielding? ...

Shielding minimizes/eliminates crosstalk and noise interference depending on the type of cable.

Network communications have two layers - logical and physical. TCP/IP protocols are transfer protocols that exist in the logical layer. TCP/IP cannot do anything about the physical layer. The physical layer comprises of the "wire" (broadly consisting of the cable, connectors, etc). Shielding minimizes/eliminates cross talk in the physical layer.

You might like reading this: http://www.siemon.com/us/standards/Screened_and_Shielded_Guide_1_Overview_and_History.asp
 
Last edited:
Shielding minimizes crosstalk.

Network communications have two layers - logical and physical. TCP/IP protocols exist in the logical layer. The physical layer comprises of the "wire" (broadly consisting of the cable, connectors, etc). Shielding minimizes/eliminates cross talk in the physical layer.

And every layer is coupled and decoupled to each other in such a way that it guarantees delivery of packets with no loss in integrity.

So then, what does one achieve by minimizing the cross-talk? When the protocol as a whole is designed to withstand cross-talk and interference?

Have you ever been to a computing data center? Please try and pay a visit if possible. There are literally 100s of CAT5/CAT5e cables of utilitarian quality tied together with colorful cable ties. By audiophile standard, there is an unthinkable amount of interference and cross-talk there. Yet, things just work beautifully. It is guaranteed to work.

I'll tell you the bottom-line. Audio equipment manufacturers attempt to make audio equipment buyers superstitious the same way a self styled baba attempts to sell his all-in-one black-magic formula to naive people. The only thing people get for believing such babas is - separation from money that would otherwise belong to their wallet.

This shielding an all matters when you are going to push the envelope to the extreme. Like you were going to need 5000mbps of sustained performance for streaming data continuously. There is no audio application as of now known to mankind that needs that kind of performance.

I have all kind of network cables. Self made ones, stock ones that you buy over the counter, from the reputed consumer electronics manufacturers. And they are from all categories. CAT 5-5e-6-7. I have tested them all brutally. Running them on high performance computers all running gigabit capable devices. The performance difference within a class was 10% between the best and the worst. And this performance difference could only be observed under stress testing conditions, not under normal use. And any audio application can only qualify as normal use because their data transfer need is a fraction of the potential of the interface.

If an ethernet cable A works better than another one B for audio application, it doesn't mean the cable A is better. It just means there is something wrong with the cable B. At least it won't have anything to do with the shielding. Because TCP/IP was designed to be immune to interference and cross-talk.
 
Ranjeet, nobody says it doesn't work. But there are issues.
That's why the gradual progression in Cat-n specifications.

I've said it before - Transmission of data is trivial. What's the point of discussing that?
People I know and respect have used and found that using a better shielded patch cable has made an audible difference in their setups. My assumption was that Cat7 having the highest spec should easily meet this criteria. I'd like to use Cat7 or Cat6.

If it helps improve my setup - whats wrong with that?
 

I read and look who is saying it, a company which provides backbone equipments of cellular networks where bandwidth is many a time maxed out due to increasing customer base and heavy usage during peak times which is not going to be your case with audio.

And every layer is coupled and decoupled to each other in such a way that it guarantees delivery of packets with no loss in integrity.

.
.
.

So then, what does one achieve by minimizing the cross-talk? When the protocol as a whole is designed to withstand cross-talk and interference?

It just means there is something wrong with the cable B. At least it won't have anything to do with the shielding. Because TCP/IP was designed to be immune to interference and cross-talk.

Very well explained ranjeet and that was my point too. Hurts being referred to articles over the net when in reality we had to spend days studying these in college and later on in work.
 
If it helps improve my setup - whats wrong with that?

The answer to this lies in the last point of ranjeet. Nevertheless if this is just for experimenting I believe nobody will stop you, probably we can all benefit from the results (positive or negative) in days to come.
 
I keep on ["+]-ing posts for reply, and then finding that the points have been dealt with by others anyway :). So If I have not answered any answers to my answers, it's just because others have done so, and expertly too.

Could you guys clarify the crosstalk thing? I am under the impression that it is about interference between pairs in one cable, and not about external interference from other cables or sources.

Years ago, and when I was first encountering this stuff as I moved from co-ax to TP, we used to call it unshielded twisted pair. This was part of the whole package: the flexibility, the ease of use coupled with the fact that, excepting the exceptional glitch causers, it worked exceptionally well. Just as well as the previous coax stuff had, and with the added advantage of being capable of ten, then a hundred, times the speed, without shielding.

Nikhil, you asked about why the progression? Is each iteration not better than the previous? It is better, but essentially only in two ways: speed and distance. If we fix the speed and do not increase the distance, then no, it is actually not better. I suggest that the data integrity of 10Mbs equipment connected by Cat 5, 6, or 7 cable would be no better than thick/thin ethernet of twenty years ago. There are no real issues with the older systems: the increments have been driven by ever greater demands.

It's a bit like saying that a Ferrari must be better for city shopping. Of course, it isn't: leaving aside the disadvantages of running a car like that in stop-start traffic, the point is that it doesn't do 10kph any better than any other car.

I'll tell you the bottom-line. Audio equipment manufacturers attempt to make audio equipment buyers superstitious the same way a self styled baba attempts to sell his all-in-one black-magic formula to naive people. The only thing people get for believing such babas is - separation from money that would otherwise belong to their wallet.

This is indeed the crux of the matter. Networking is not an audio protocol. It is not a "digital cable" like S/PDIF or USB-Audio that we could argue about. None of the "audiophile" considerations, sensitivities, beliefs, etc apply. Not even the reasonable ones.

There is an entirely academic (because LAN speeds give us room for a lot of error, but in fact, there will be little or none) reason why tcp/ip is not theoretically ideal for audio or any other real-time application: It guarantees getting the packets from A-B (or at least trying its damndest before failure) and assembling them in the right order, but it does not guarantee how long that will take. This leaves us with just one question: is it fast enough to transmit our audio, end to end, without breaks or stutters. Obviously, as we watch Youtube, we know that the internet, sometimes isn't --- but our LAN, even at the most basic level, is.

Nevertheless if this is just for experimenting I believe nobody will stop you

True, and despite my avowed lack of interest, I'm serious when I say that it would be fun to be able to show people a high-speed fibre/other network in the home. What is not fun is when people say that it is needed or sounds better.

It is not an audio protocol; audio thinking does not apply. If it actually does sound better, then either something was very wrong before (as per Ranjeet's Cable-A/Cable-B), or we need to look to those fascinating cables inside our heads for the explanation. I really think we should be paying those cables a lot more attention, but that, of course, is another story.
 
I may well end up going the Cat6 route but really what appeals to me about Cat7 is the shielding. People I have been in contact with have confirmed that it makes a difference especially over the 25m length I am looking at.

When I tried to extend my PC display to another room via CAT6 casble as display cable, there was too much distortions. I suppose Cat7 should overcome this with ease.
 
So then, what does one achieve by minimizing the cross-talk? When the protocol as a whole is designed to withstand cross-talk and interference?

It would be nice to get a reference to this.
I have a very basic understanding of the TCP IP protocol/stack and nothing I know of says of this specifically.

It is guaranteed to work.

Especially this part.
 
Guys, I have been following this post with Both interest and .... To A Greater Extent .... Despair.

I believe that a "Time Out" is called for here.

Nikhil started this tread ( and therefore its his thread) with this post:

I was looking around for Cat7 cable and was surprised to find that my local computer bazaar is clueless about CAT7 cable. PLenty of Cat5 and Cat6 cables but very little info on Cat7 cables. Even while looking up Cat7 cable online, all results are overseas suppliers. Would like some help from FMs on sources of CAT7 cable for approx 25m length.

The best offer I can get on a "bulk" cable is for Cables Unlimited RJ45 terminated 50 ft bundle for approx Rs 6,300 on Amazon.in

Any help would be appreciated.

Right from the next post, this thread went South, with Covert & Overt suggestions that CAT-7 was not necessary.

Please read the 1st Post by Nikhil. He has not asked for wisdom on whether CAT-7 was suitable for his application .... In fact he did not even mention his application!

Against a post that called for help to source it .... Nikhil seems to have been called to defend his decision on using CAT-7 cable ! :eek:

Even when he has given reasons and his point of view, the argument seems to have taken a life of its own .... & irrelevant to his sourcing request. :p

Discussions ( make that Arguments) on Cables and their audible differences are the bane of HiFi forums.

Can we just not agree to disagree, instead of trying to convert Nikhil to the other point of view that CAT-7 is irrelevant !

Do think it over. :rolleyes:

Good Night.
 
Nikhil,

You can research if long VGA cables cane used as LAN cable. VGA has more wires running + all are shielded as you expected (most of the VGA cables I spliced) Also available with ferrite cores at the ends.

This will work out cheaper for you.
 
Could you guys clarify the crosstalk thing? I am under the impression that it is about interference between pairs in one cable, and not about external interference from other cables or sources.

In my context at least - I am also concerned about external interference. I have concerns about noise that can be picked up by the cables from the electrical environment around my place. If possible, I would love to send the cable right through the existing electrical duct if the shielding is robust enough!!
 
Last edited:
Can we just not agree to disagree, instead of trying to convert Nikhil to the other point of view that CAT-7 is irrelevant !

Do think it over. :rolleyes:

Good Night.

+1, I agree that sticking to the topic is imperative in such threads.
Cheers,
Sid
 
When I tried to extend my PC display to another room via CAT6 casble as display cable, there was too much distortions. I suppose Cat7 should overcome this with ease.

There's a crucial difference. You were trying to move an analog signal instead of a digital signal. Analog signals are highly susceptible to noise, EM, crosstalk, etc. For your usage case, typically, a balun is used.

Digital network is (nearly) immune to most of these problems. The issue with digital networks usually comes from targeted bandwidth vs required bandwidth. Yes, noise reduces bandwidth, but doesn't damage data like it does in analog transmission.

But to Arunlouie's point, OP's question was different. And if one can get a pipe/wire with extra bandwidth and extra shielding, why not? I actually learnt something new from this thread, so thank you, Nikhil! I actually quite like the "flat wire" option you shared - it would be especially unobtrusive if running under carpets etc. Cost is quite reasonable as well.

Sorry, couldn't contribute anything of benefit to this thread.
 
If possible, I would love to send the cable right through the same electrical duct if the shielding is robust enough!!

I sort-of wish you'd put that in the first post! it is not wise. I don't know if best-practice recommendations have changed, but I suspect (Ranjeet and Arun will tell us) that they would still be to separate electricity and data runs, avoid fittings such as fluorescent lights, and when electricity and data must cross, do so at right angles.

Of course, best practices are one thing in brand new buildings, and quite another some years later.

If one must run a network cable for some distance in a mains cable duct, would a shielded design work better than an unshielded design? Specifically, would CAT-7 work better than earlier cables? I don't know.
IndianEars said:
Right from the next post, this thread went South, with Covert & Overt suggestions that CAT-7 was not necessary.
It has been a good and useful conversation, and the only useless thing so far is any suggestion as to what we should or should not be saying. There has been no cable religion here, just a good give and take about what might, or might not, work, and why. I didn't see any complaints from Nikhil, who has taken all in his stride, So please do not take it "South" now.

arunlouie said:
You can research if long VGA cables cane used as LAN cable. VGA has more wires running + all are shielded as you expected (most of the VGA cables I spliced) Also available with ferrite cores at the ends.

No. Network cable is very specifically designed and optimised for its purpose.

There are methods of networking over other kinds of cable, even combining it with house electrical wiring is possible! This takes special equipment
 
Last edited:
It has been a good and useful conversation, and the only useless thing so far is any suggestion as to what we should or should not be saying. There has been no cable religion here, just a good give and take about what might, or might not, work, and why. I didn't see any complaints from Nikhil, who has taken all in his stride, So please do not take it "South" now.

Thad, in IndianEars defense he is spot on. It was going South as far as I was concerned.
I don't mind a good discussion but some of the questions asked (in the context of my original post) were to put it mildly .... condescending.

Have I been to a Data Center?
Have I driven a Ferrari at 10 kmph?
Am I running a Gigabit network?

No. But I would like to use a Cat 6/7 cable anyway! Preferably shielded! :cool:
 
Wharfedale Linton Heritage Speakers in Walnut finish at a Special Offer Price. BUY now before the price increase.
Back
Top