Coaxial vs Optical S/PDIF bandwidth - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 14 Old 01-21-2007, 09:38 PM - Thread Starter
Senior Member
 
trueimage's Avatar
 
Join Date: Sep 2005
Location: Calgary, AB, Canada
Posts: 258
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I have read many posts here and searched google, but can't find an authoritative, definitive answer on this one.

What is the bandwidth limit of these cables, and are they the same?

Here is why: On my HTPC both optical and coaxial output worth fine for everything. Excpet a DTS track. Only the coaxial gives me sound. Optical gives no sound at all.

I thought I read somewhere that coaxial was better because of the bandwidth, but I can't find that anymore. Was this a freak thing, or something else causing the problem? Just wondering...

trueimage
trueimage is offline  
Sponsored Links
Advertisement
 
post #2 of 14 Old 01-23-2007, 11:25 AM
AVS Special Member
 
trekguy's Avatar
 
Join Date: Nov 2005
Location: West of Mt Diablo
Posts: 2,192
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
You might have a poor quality or very early Toslink cable or module or you have set the PCM sample rate incorrectly for if your player has that option). Optical cable bandwidth is dependant upon manufacturing quality, but a good (not necessarily expensive) Toslink cable can handle DTS--my RS sale cable does just fine.

The data rate for DTS 96/24 is 1.536Mbit/s. Toslink input/output modules for digital audio can carry 15 Mb/s at a 96kHz sampling rate.

The fact that an opinion has been widely held is no evidence whatever that it is not utterly absurd.
-Bertrand Russell
trekguy is offline  
post #3 of 14 Old 01-28-2007, 01:14 PM
Member
 
jhnmdahl's Avatar
 
Join Date: Jan 2002
Location: Minneapolis, MN
Posts: 73
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Optical has a much higher bandwidth, but the optical Toslink connectors on audio products are converted from the signal that gets sent over a coaxial digital audio cable, so if anything will have more jitter and other problems due to the conversion to optical and back. Optical remains attractive for very long runs or in really noisy environments (e.g. some studios or live theater environments), but is otherwise usually not a good idea despite its "bling" factor.

Good luck,

John
jhnmdahl is offline  
post #4 of 14 Old 01-29-2007, 07:11 AM
 
Targus's Avatar
 
Join Date: Jan 2006
Posts: 2,617
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
but the optical Toslink connectors on audio products are converted from the signal that gets sent over a coaxial digital audio cable

nonsense.

Quote:
so if anything will have more jitter and other problems due to the conversion to optical and back.

nonsense.

The "conversion" is done with an LED...it's not difficult.
Targus is offline  
post #5 of 14 Old 04-27-2009, 07:02 PM
Member
 
batagy's Avatar
 
Join Date: Nov 2005
Location: Hungary
Posts: 66
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi!

I just wonder what can be the biggest frequency bandwidth requirement of a digital SPDIF signal (Coax)?

I mean the frequency bandwidth not the data transfer bandwidth.

This page writes that the original SPDIF standard was specified for 6 MHz.
It also write this:
Signal bitrate is 2.8Mhz for sample rate 44.1kHz (I guess because 44.1 kHz * 16 bit * 2 channel = 1.4 MHz data stream).
Signal bitrate is 3.1Mhz for sample rate 48kHz (I guess because 48 kHz * 16 bit * 2 channel = 1.536 MHz data stream).

It's OK, but if I know correctly, SPDIF can transmit maxium 192 kHz 24 bit 2 channel PCM, right?

So it that case, the calculation would be: 192 kHz * 24 bit * 2 channel = 9.216 MHz data stream. So would it require 2*9.216 Mhz = 18.4 MHz bandwidth?

I am correct?

I ask this because I'm thinking about a ground loop isolator for the digital coax. I didn't find any ground loop isolator made specially for SPDIF coax, but I've found ground loop isolator for composite video.
For example This Cables To Go Composite Video isolator is only specified for 8 MHz bandwidth. And this is having 10 MHz bandwidth.

I have found a SPDIF gadget ( an SPDIF balun for CAT5 ), this is specified for 25 MHz, so that seems to be conform to my above calculation, but this is not a ground loop isolator.

Are my calculations good, and does anyone know a ground loop isolator for SPDIF coax?
batagy is offline  
post #6 of 14 Old 05-02-2009, 10:11 AM
Advanced Member
 
Ugly1's Avatar
 
Join Date: May 2007
Posts: 872
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Targus View Post

nonsense.



nonsense.

The "conversion" is done with an LED...it's not difficult.

Nonsense. That the conversion is done at all, and in this case using earthling technology in the form of non ideal components such as LED's, transistors, resistors etc., implies at least some level of degredation in the signal will have necessarily occured.

One of the most complained about degradations typical of commercially available audio gear is the increased clock jitter lower quality implementations of this electrical -> optical -> electrical type of conversion yields. Unfortunately the sonic effects of clock jitter introduced to a spdif signal chain are identifiable and measurable with both listener ears and lab test equipment.

It is a fact that spdif signal clock jitter on gear with both electrical and optical spdif interfaces is typically going to be a bit worse for the optical implementation as compared to electrical due to the fact that there is more electronics this signal must propagate through. It has been shown that clock jitter in spdif signals changes sonic attributes enough to be verified experimentally.
Ugly1 is offline  
post #7 of 14 Old 05-02-2009, 10:26 AM
 
duvetyne's Avatar
 
Join Date: Dec 2008
Posts: 1,635
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:


Nonsense. That the conversion is done at all, and in this case using earthling technology in the form of non ideal components such as LED's, transistors, resistors etc., implies at least some level of degredation in the signal will have necessarily occured.

Then you would agree that a coax or balanced AES/EBU interconnect will also be plagued with the same issues. Mainly due to the earthling made transistor driving the inexpensive earthling made pulse transformer in a typical coax interface?
duvetyne is offline  
post #8 of 14 Old 05-02-2009, 10:56 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,290
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 680 Post(s)
Liked: 1144
Quote:
Originally Posted by Ugly1 View Post

Nonsense. That the conversion is done at all, and in this case using earthling technology in the form of non ideal components such as LED's, transistors, resistors etc., implies at least some level of degredation in the signal will have necessarily occured.

Your error here is presuming that coaxial cable is any better in these regards.

Typically, a SP/DIF output is driven through a transformer, and on the receive end there is a possibilty that the digital signal will be partially or completely corrupted by noise that is picked up by the coax, grounding problems, etc.

In difficult environments, optical is clearly preferred. Optical eliminates a commonly-reported problem with coax, which is muting of the received signal when an appliance such as refridgerator turns on or off.

Optical also eliminates potential grounding problems that may show up as ground loops that cause hum and noise on regular analog inputs or outputs.
[/quote]


Quote:


One of the most complained about degradations typical of commercially available audio gear is the increased clock jitter lower quality implementations of this electrical -> optical -> electrical type of conversion yields.

While jitter is frequently complained about, actually indicting it as the source of an audible problem is a whole different thing.

Jitter in a digital link is far from being an uncorrectable problem. It's the job of any effective digital line receiver to make jitter simply go away. This sort of thing has been done effectively inside CD players for almost 30 years. In this case the jitter is due to non-uniform rotation and reading of the CD disk itself.

The "magic bullet" against jitter has been known about for at least 50 years - its a large buffer and a precise clock.

Quote:


Unfortunately the sonic effects of clock jitter introduced to a spdif signal chain are identifiable and measurable with both listener ears and lab test equipment.

Been there done that, and have shown to myself that even fairly inexpensive modern surround receivers have what it takes to make jitter picked up on a digital interconnect simply go away.

Quote:


It is a fact that spdif signal clock jitter on gear with both electrical and optical spdif interfaces is typically going to be a bit worse for the optical implementation as compared to electrical due to the fact that there is more electronics this signal must propagate through.

Pshaw. In fact jitter is more likely on a piece of coax because its susceptibility to picking up interferance and noise due to grounding problems.

As has been pointed out, the *extra components* in an optical link are composed of a LED and a phototransistor. These components have negligable inherent jitter and it doesn't matter anyway because the digital circuitry that follows it has been tasked with the responsibility of making jitter go away through buffering and reclocking.

[/quote]
It has been shown that clock jitter in spdif signals changes sonic attributes enough to be verified experimentally.[/quote]

If you actually check the publication dates on the articles about this problem, the bulk of them were written in the early 1990s, when digital line receivers weren't nearly as sophisticated as they are today. Again, I've done extensive bench and listening tests of equipment from that era, and yes they had a problem with jitter. The same tests performed on relatively inexpensive equipment from the early part of this millenium showed complete resistance to jitter, short of jitter so grotesquely strong that it was impossible to recover correct data.

And yes, you'll find still some DIY modification and tweak hardware supplier sites that are beating this old myth to death. Just because someone is trying to mine urban myths and amateur paranoia for cash or influence doesn't mean that it is really a problem at this time.
arnyk is offline  
post #9 of 14 Old 05-03-2009, 02:40 AM
Advanced Member
 
Ugly1's Avatar
 
Join Date: May 2007
Posts: 872
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by arnyk View Post

Your error here is presuming that coaxial cable is any better in these regards.

Your error would be presuming what I have presumed, but you guessed correctly. I would presume that coax would be a bit easier to modulate well , especially on a budget, as compared with optic links. My argument would likely revolve around superior jitter performance of not having to do voltage to current conversions (and back) to drive LED's or laser diodes and there photodiode or phototransistor counterparts, and the maturity of the controlled impedance PCB, pulse transformer, and etc markets. All these big junction geometry high capacitance current hogging slush boxes add up to sacrificed speed accuracy/performance from a circuit design standpoint.

Don't get me wrong since I never once said that electrical is always superior. I just happen to have the opinion that in this market we are more likely to get better performance from the electrical circuits. There's obviously exceptions to every rule.

Regarding the obviously electrical problems associated with this appliance generated signal muting (which I am completely unfamialiar with since it doesn't affect me) or ground loops...seriously? I personally make it my mission to design out these kinds of concerns from my own setup and I really can't be bothered with this particular incompetence on the parts of the audio industries lack of a true universal grounding standard and the fact that probably 95% plus of setup problems could be eliminated by users pulling their heads out a little more. On the one hand it's sad that obviously so many do have electrical problems but a competent setup tech should be able to work through it and it is then a non issue.

Quote:
Originally Posted by arnyk View Post

While jitter is frequently complained about, actually indicting it as the source of an audible problem is a whole different thing.

I never indicted anything just pointing out that I'm told that some have a big problem with jitter. The only problem I have persoanlly had with jitter was due to data loss in with skips and pops but I'm not going to discount others results with more subtle jitter problems.

Quote:
Originally Posted by arnyk View Post

The "magic bullet" against jitter has been known about for at least 50 years - its a large buffer and a precise clock.

Too bad manufacturers don't get to use magic. Understanding that jitter sucks and truly removing it from circuitry are two very different things. Jitter will be with us forever in electronics even as we are making great strides with technology to reduce it as applications keep getting tougher over time.

Quote:
Originally Posted by arnyk View Post

Pshaw. In fact jitter is more likely on a piece of coax because its susceptibility to picking up interferance and noise due to grounding problems....These components have negligable inherent jitter

I generally disagree with these statement. I have to come back to my old argument from above that it's much more difficult to drive a square current wave through an LED than prevent noise from being a problem in a well laid out/grounded systems 75 ohm data links....on coax no less. I admit that with more money the LED drive circuitry can be better, but then again at the same time this same money could be used improving the electriomagnetic interface as well.
Ugly1 is offline  
post #10 of 14 Old 05-03-2009, 02:48 AM
Advanced Member
 
Ugly1's Avatar
 
Join Date: May 2007
Posts: 872
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by duvetyne View Post

Then you would agree that a coax or balanced AES/EBU interconnect will also be plagued with the same issues. Mainly due to the earthling made transistor driving the inexpensive earthling made pulse transformer in a typical coax interface?

I certainly would, as I would hope you'd agree electrical trancievers offer quite bit more bang for the buck as compared with all the optoelctronics and fancy drive circuitry necessary for achieving equal levels of jitter performance.
Ugly1 is offline  
post #11 of 14 Old 05-03-2009, 02:35 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,290
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 680 Post(s)
Liked: 1144
Quote:
Originally Posted by Ugly1 View Post

I would presume that coax would be a bit easier to modulate well , especially on a budget, as compared with optic links.

Not true.

Quote:


My argument would likely revolve around superior jitter performance of not having to do voltage to current conversions (and back) to drive LED's or laser diodes and there photodiode or phototransistor counterparts, and the maturity of the controlled impedance PCB, pulse transformer, and etc markets.
[\\quote]

Please get back to me when you are certain about that argument, instead of your current situation fo not knowing whether you want to make that argument or not.


Quote:


All these big junction geometry high capacitance current hogging slush boxes add up to sacrificed speed accuracy/performance from a circuit design standpoint.

Are you sure about that?

The transformers used to drive SP/DIF coax *intentionally* sacrifice speed and performance in order to avoid creating illegal amounts (FCC Part 15) of EMI. Toslink line drivers don't need to do that because they can't possibly create EMI.

Quote:


Don't get me wrong since I never once said that electrical is always superior. I just happen to have the opinion that in this market we are more likely to get better performance from the electrical circuits. There's obviously exceptions to every rule.

You get to be wrong.

Quote:


Regarding the obviously electrical problems associated with this appliance generated signal muting (which I am completely unfamialiar with since it doesn't affect me) or ground loops...seriously?

Big time.


Quote:


I personally make it my mission to design out these kinds of concerns from my own setup and I really can't be bothered with this particular incompetence on the parts of the audio industries lack of a true universal grounding standard and the fact that probably 95% plus of setup problems could be eliminated by users pulling their heads out a little more.

How condescending of you.

Quote:


On the one hand it's sad that obviously so many do have electrical problems but a competent setup tech should be able to work through it and it is then a non issue.

Plan B: Use optical and avoid these setup problems completely.



Quote:


I never indicted anything just pointing out that I'm told that some have a big problem with jitter.

You were also told that optical is unlikely to have no problems with most common cause of jitter with SP/DIF.

Quote:


The only problem I have persoanlly had with jitter was due to data loss in with skips and pops but I'm not going to discount others results with more subtle jitter problems.

Switch to optical, and the skips and pops will probably go away.

This is pretty rich - you have audible problems that optical will probably cure, but you are avoiding optical because you fear audible problems that have been generally solved.

Quote:


Too bad manufacturers don't get to use magic. Understanding that jitter sucks and truly removing it from circuitry are two very different things.

Yes, but they both happened in the past. No magic required.


Quote:


Jitter will be with us forever in electronics

Minute amounts of jitter yes, but audible amounts of jitter, no.

Quote:


even as we are making great strides with technology to reduce it as applications keep getting tougher over time.

How many times do you have to be told that audible jitter is solved problem before you will believe it?



I generally disagree with these statement. I have to come back to my old argument from above that it's much more difficult to drive a square current wave through an LED than prevent noise from being a problem in a well laid out/grounded systems 75 ohm data links....on coax no less. I admit that with more money the LED drive circuitry can be better, but then again at the same time this same money could be used improving the electriomagnetic interface as well.

arnyk is offline  
post #12 of 14 Old 05-03-2009, 03:15 PM
AVS Special Member
 
11001011's Avatar
 
Join Date: Aug 2004
Location: N.W. Washington
Posts: 1,361
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by trueimage View Post

Here is why: On my HTPC both optical and coaxial output worth fine for everything. Excpet a DTS track. Only the coaxial gives me sound. Optical gives no sound at all.


I think this is a problem in your HTPC setup or configuration.

My DVD player has both coax and optical outputs and my HTS has both coax and optical inputs.

I can use either one and get full DTS sound.

I have hooked both up and done A/B comparisons.

I can not hear any difference between the two.

A 'phile and his money are soon parted...
11001011 is offline  
post #13 of 14 Old 05-04-2009, 04:12 AM
Advanced Member
 
Ugly1's Avatar
 
Join Date: May 2007
Posts: 872
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by duvetyne View Post

Statements like this:



Indicate that you really don't have much of a clue.

I note the total avoidance of bringing the tech and resorting to ad hominem on your part. Just to point it out in case it isn't clear. Can you present any technical support for your argument as I attempted within my post?



Quote:


You don't seem to be familiar with common problems or the existence of ground loops....very telling!

Haha! Yeah well, some of us learned to understand and deal with these problems already. Why put up with ground loops unnecessarily? One might also say your statement is telling. You're reasoning that since I don't have to deal with the lameness that comes with ground loop problems since I've reduced the presence of them in my equipment to an insignificant level somehow implies that I don't have any understanding of what a ground loop is would be completely illogical BTW.

So James Clerk Maxwell...why don't you give us the technical rundown on the mechanisms behind which how refrigerators cause muting with electrical spdif implementattions. This should entertain.
Ugly1 is offline  
post #14 of 14 Old 05-04-2009, 05:24 AM
Advanced Member
 
Ugly1's Avatar
 
Join Date: May 2007
Posts: 872
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by arnyk View Post

Not true.

Care to offer any reasoning? Clearly it would be easier to at least understand why you disagree if you were to address some of the technical supporting arguments I've raised such as the difficulty of designing circuits to drive anything like a high speed square current pulse through the optotransmitters being used as compared to driving a voltage pulse onto a controlled impedance transmission line.

Let's talk about some circuits now since we are at this level of the conversation now. I do happen to have extensive direct design experience with these types of circuits and am having trouble seeing what your point is. I think I've given you enough about my point of view to come up with a technical rebutal, if not please let me know.

Quote:


Please get back to me when you are certain about that argument, instead of your current situation fo not knowing whether you want to make that argument or not.

Please see my technical argument above which is the second time I've now spelled it out for you. Will you choose to ignore it yet again?

Quote:


Are you sure about that?

As I have been saying...yes, this is my experience. What is your experience that would cause you to disagree with what I've seen.

Quote:


The transformers used to drive SP/DIF coax *intentionally* sacrifice speed and performance in order to avoid creating illegal amounts (FCC Part 15) of EMI. Toslink line drivers don't need to do that because they can't possibly create EMI.

I would ask that you please provide some tech which backs this assertion. What about the transformer designs is being done which sacrifices speed for emi reduction? This goes contrary to my eperience which shows inexpensive pulse tranformers from various low balling manufacturers I've used in data transmission are usually quite insignificant in their EMI profile (at least from a Part 15, EN 55022 perspective) as compared with the transmission lines they are driving. The biggest concerns I've always had are related to the impedance control, bandpass characteristics, leakage, isolation etc.

And the bit about Toslink circuits not possible radiating...well let me tell you something...any circuit can radiate in the correct hands, or should I say incorrect hands. It's all a matter of the physics of the circuit involved: for example if this LED driver happened to be dumping half of the intended LED energy into a tuned antenna...well you get the idea. Don't laugh, it is identifying these unintentional tuned antennae that PCB layout dudes are so fond of trying to sneak into the layouts thats a big part of keeping me interested in circuit design.

Quote:


You get to be wrong.

This doesn't appear to even make any sense in the context it was quoted. It appears to be a lame attempt at ad hominem. Did I miss the beauty of this gem? If so please help me out here. What exactly am I supposed to be wrong about here?

Quote:


Big time...How condescending of you.

Well, I aready apologised didn't I? I used to deal with the symptoms caused by gross ground loops and it sucked. I found a way that works for me, and like magic many problems were gone...and who knows how many were fixed that I hadn't even identified or correlated grounding problems. There is a better way which by design necessarily eliminates most of what seems to generate much of the chatter on sites like this. There are great tutorials on this very site revealling all the secrets.

Quote:


Plan B: Use optical and avoid these setup problems completely.

I certainly have no problem with plan B and use it extensively in my own system currently. It's not my favorite solution which is technically possible but works pretty well for me. Again, if you think I have been rallying against using optical links I'm sorry you missunderstood my point. I have merely been pointing out my distaste for blanket statements being made in this thread such as when I made my first reply to Targus or when you say things like
Quote:


Jitter in a digital link is far from being an uncorrectable problem.

which is blatantly false. If you don't think jitter in modern digital links is a problem try designing the sampling stages for a state of the art digital network analyzer. Jitter will always be a problem to be dealt with it is only a matter of what is necessary to accomplish a given application that changes the answer.

Quote:


How many times do you have to be told that audible jitter is solved problem before you will believe it?

What are you talking about man? I already said that I don't have any identified jitter issues. I am only not going to go out on a limb and try to say the others who claim to have audible jitter problems are wrong as you seem willing to do. I'm sorry if you think I am discounting what you are saying, because I'm not, but a good objectivist bases his theories on more than what just one internet poster tries to convince him of. I am not really intersted in searching out this data since as I say jitter is a nonissue for me, but as soon as I see data I trust that agrees with you then I'm all in. Until then I'm attempting to keep an open mind.
Ugly1 is offline  
Reply Audio theory, Setup and Chat

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off