Digitial Coxial or Toslink (1 Viewer)

[RoyGBiv]

Balanced XLR are the best type of connection because there are two shields that can isolate a cable, but they really shouldn't be needed for this use because, again, these are digital transmissions not analog and not subject to the same kinds of interference.

SMK

 

[Blaine_M]

My local store salesman (marantz/paradigm dealer) says you get a larger bandwidth using coax instead of optical. Is that BS?

 

[MikeSHU]

Yes its BS and your salesman probably doesn't even know what bandwith is or why on earth it would ever affect audio quality, LOL

1. Okay, actual 1's and 0's are not transimed but electrical pulses representing them are. Think morse code -thats a digtal signal. Think of a DVD player sending video/audio information to your reciever using morse code. Thats what a digital cable does. Your DVD player says to the reciever _._..._._.__._._ (01011101010010) and the reciever's DAC says to it'self (okay i'll make a 76.34234hz electrical wave for the speakers). If you were to look at a digital signal on an oscilloscope you would see it looks like a square wave _|-|_|---|__|-|_|-|_|-|_.

2. In order for a a optical cable to pick up "jitter" that would mean something would have to alter the light pulses going through the optical cable the same way an electro magnetic field alters an electrical current in a analog cable. I know of nothing that could do that to a light pulse. I think optical and digital differnces are really all placebo. As I said before, assuming the digital coax cable is well shielded, there is no reason why it should sond any better or worse than a optical cable.

 

[John Garcia]

Something as simple as bending the cable to the point that it affects the light will do this. A bad or scratched lens on either end could do this. Reflections within the cable itself can be an issue.

If digital signals are always perfect, then why does a $100 transport not perform as well as a $10,000 one? Digital is still suceptible to variances, though on a much lower level than analog would be to EMI/RFI.

 

[Phil A]

Yes Mike that is correct about the transmission. My explanation was brief and I just wanted to clarify that 1s and 0s are not going over the cable. This is one explanation of jitter issues:

http://www.audiovideo101.com/learn/a...nid=101&List=7

"Coaxial cables are subject to jitter problems, it is not an issue for TOSLINK cables only, however, coax cables tend to suffer from less jitter problems. Thus is the signal is the same but the coaxial cable introduces less distortion in the form of jitter timing errors the coax connection can produce a more perfect signal. Is there a difference? Yes. Is it audible? It can be. Should you be concerned? Not overly. If you have the option between coaxial and TOSLINK digital connections, go the coaxial route. If you have no option, the differences are subtle and TOSLINK works fine as a transmission format for digital bits."

As I noted for DD I would doubt that anyone would hear a difference since there is packet transmission. For those who listen to 2-channel music critcially on a good system, it could be audible. Lots depends on the quality too. You can get cheap optical cables or cheap coaxial cables with less than good quality connectors.

 

[MikeSHU]

John, you are correct. But a digital signal does not get distorted like an analog signal. It is either the correct signal or the corrupt signal. Not a kind-of correct signal. AFAIK digital audio signals have error correction and do use error-correcting algorithms (although I could be wrong). Here's how computers use it (I’m sure digital audio signals have something similar) Lets say you have a byte which is 8 bits, 10110100. Now while being copied it becomes corrupt and the 5th bit becomes unknown 1011X100. Without error correction you are screwed. But let me give you the simplest example of error correction.

An error corrected byte would have 9 bits, 8 bits representing the actual data and an extra bit known as the parity bit. The parity bit tells the computer whether the sum of the bits in a byte is odd (1) or even (0). In the case of 10110100 the parity bit would be 0 (1+0+1+1+0+1+0+0 = 4 [even]) so the over all piece of data would look like this 101101000.

Now let’s picture that corruption scenario again with a parity bit. 101101000 is corrupted into 1011X1000. The computer says to itself:

"Shit! I don't know what that 5th bit is"

"Hold on a sec, let me check that parity bit"

"The parity bit is 0 so that means the sum of the bits of the byte must be an even number"

"1+0+1+1+1+0+0 = 4 [notice I excluded the 5th corrupt bit and the parity bit which has no "data value"]"

"4 is an even number! That means that the 5th corrupt bit must be a 0! (because if it was a 1 then the sum would be 5 and the parity bit would be a 1).

This is an example of the simplest way of how a computer can correct a flawed digital signal. There are far more complex algorithms. So assuming that the digital audio signal has error correction code built into it (which I'm pretty sure it does), bending the optical cable (unless you brake it or bend it to the point where no light goes through) would not make a difference.

"If digital signals are always perfect, then why does a $100 transport not perform as well as a $10,000 one? Digital is still susceptible to variances, though on a much lower level than analog would be to EMI/RFI."

There are 2 explanations. The 1st obvious one is marketing. The second is that a $10,000 cable uses a much better quality optical material and can be stretched a much longer distance without the use of a repeater.

That is another advantage of optical over digital coaxial, BTW. Optical cables can be used over very long distances whereas a signal sent over copper is subject to electrical resistance. However, unless the distance between your receiver and your DVD player is that of a tennis court this wouldn’t matter to you.

 

[MikeSHU]

Makes sense now. A TOSLINK cable is more susceptible to corruption and therefore can overload the error correction circuitry buffers. This can create a bottle neck from the correction circuitry to the DAC which can result in frames being dropped. This seems possible. If it is then it would not be only the quality of the TOSLINK cable but also the quality of the hardware (specifically the capacity of its buffers) that determines the audio quality of a digital signal. But once again this is unlikely. If I was to make a choice between TOSLINK and Digital Coaxial I would go with the TOSLINK. I feel that EMI emitted from a digital coaxial cable has a greater chance of distorting an audio signal than jitter from a TOSLINK.

 

[John Garcia]

The reality is, the difference is going to be minimal between the two. Going from a cheap plastic optical to a good quality glass optical will probably make a bigger difference than between similar quality coax and optical cables.

 

[Steven_Lazarus]

Ahh the good'ol Coax vs Optical strikes again !, I say just get some headphones and call it good! gee-sh saves money by bypassing the whole speaker thing to! and of course "cables not included"
Hmmm, but wait, that raises the question, do I go with coax or optical headphones???
Just use both and switch the out on a monthly basis, that way your only missing out either way half the year and benefiting from the superior sound quality the other half

 

[Phil A]

Actually when I was at a (hi-fi) show one time, a small cable manuf. had a set of Grados that were modified and instead of the std. headphone cable had RCA input jacks for each channel so that one can listen to different (analog) cables. Would be interesting to try with a DAC (output to the headphones) using different digital cables from a transport to the DAC too.

 

[John Wes]

I tried doing this....

I connected both a high quality coax and a "glass" TOSLINK from my Pioneer 45 to my Pioneer 47tx Now I realize the DVD player isn't exactly high end but it's only being used as a transport in this fashion, right?

I hooked one up to the DVD digital in and the other to the CD digital in.

Then I grab several CD's and invited my daughter to listen with me, switching between CD and DVD input.

Neither of us could tell a lick of difference. She can tell which side right or left, I have my home made Van Haus cat 5 speaker cable hooked to vs high end monster for speakers..100 percent of tests done. I can only tell about 60 percent correct...

Sooo..imho, if she can't tell the difference, then chances are, there isn't any.

 

[Steven_Lazarus]

Sooo..imho, if she can't tell the difference, then chances are, there isn't any

That's kinda risky
My wife tells me things like that all the time just so I leave her alone

But!, yup, unless the clouds parted and sound waves filled your ears and warmed your cocclia (SP?) I don't think your going to be able to tell the difference. But this argument will be continued in the next TOS vs Coax post so stay tuned !!!

 

[Lewis Besze]

It's not a big issue but a "blown out of proportion by some audiophiles" issue,which won't matter to most.

I use both connections,for equally good results.
I agree that a good coax cable can be cheaper[I.E. composite video cables]

 

[Jan H]

Generally speaking, Coax is preferable to Toslink for CD's, IMO, and I've always preferred the sound of my Kimber Illuminati D-60 over any optical cable out there. But if you're going to go Toslink, definitely go 'glass.'

 

[Phil A]

Jan, I agree. Most here are likely into HT vs. a really good music system. I had a friend who worked at a high end shop (where they had some HT but not the focus of the store) at one point and for 5 yrs. I did basically every install/set-up with him and we would go out for beers afterwards. Before I did this with him, I thought my HT was great music playback and it was for a HT. I can't tell you how many times I had to sit thru (and be polite) someone playing something on a HT system that absolutely sucked and they thought it was great music playback. I would bet that a majority of people here would hear minimal or no differnce based on the way they set-up and use their systems. I've done lots of listening with top of the line stuff and even still have an optical cable that I got from a European manuf. that at the time had no US distributor and would likely be something that sold at $125-150 at the retail level. I would not say it can't beat any coax on music since I have not listened to every coax, but certainly not something I would use listening to music.

 

[Jan H]

Agreed. Most people on the HTF would mention phrases like 'placebo effect' and that would be that. In any case, I use a glass optical cable from my DirectTV HD box and am extremely impressed. Originally tried the cable that Hughes included and it was dropping out like an illiterate 9th-grader. Picked up a glass optical and it has never dropped out once in 18 months. So, I'm definitely a fan of glass optical cables for certain applications. But nothing beats a top notch coax for music reproduction.

 

[matt-f]

Mike,

I was told that digital outputs don't have like a CRC/ECC checking. This will be different comparing fiber vs eithernet on a computer which does use CRC/ECCs.

 

[Lewis Besze]

Rightly so, as most of the so called "claims" can't be validated by any objective method[maybe DBT],so we are left with the anecdotes only.

 

[MikeSHU]

matt-f,

It would not make any sense why ECC would not be implamented. It is simple and very effective

 

[Phil A]

Lewis, I've done DBTs with analog and digital cables before with an old set-up that made it easy. There's probably a thread in the tweaks area perhaps a year old or so. All the tests were done with my old set up from a few years back which included a Proceed PMDT transport for digital (it had no analog outs and several digital) into a Proceed AVP pre/pro, Thiel 7.2 speakers, etc. It was done so the person inserting the cables knew nothing about audio basically and did not participate in the test and I did the A/B switching via assigning the inputs but did not participate (only kept a tab of votes) and did not know what was in A/B. In all I had about 20 people tested incl. a little less than half of those engineering types who insisted there could be no difference.

Bottom line, in every single test every person was able to tell a difference. Differences were bigger on analog cables (used with something other than the PMDT). That's 20 out of 20. The usual response from there's got to be a placebo crowd was that there must be a difference in the preamp inputs or something in the digital outs and they wanted to take apart my equipment. Since the test was repeated with different music and over and over the votes tallied confirmed there possible explanation was not possible by the combinations done. It was kind of amusing to watch them squirm and try to explain that. Theory is one thing. Real world listening is another.

My current set-up does not allow convenient testing and to be frank I don't care if someone uses coat hangers to connect components. About a week ago I made cables for an audiophile friend and another friend who just believes in something adequate. It's not a problem for me, I was happy to help both of them. Discussions really don't resolve anything except some name calling or implications of placebo or sucker or perhaps a little on the crazy side. Anyone who wants to send me money via PayPal to help fund my straight jacket are welcome to.