Optical Cable

[tbewick]

I did say that I thought the issue, which indeed I raised, was trivial earlier on, so I think it's slightly unfair saying that I've blown things out of proportion. In saying that, the diagram I drew I'm sure exaggerates the level of distortion a typical cable would introduce in a home hi-fi set up. My point was simply that such distortion is possible because of the cable, but can be rejected. This is unlike a signal that is not digitally coded.

 

[tbewick]

I have to hold my hand up over this one - you would probably get no significant distortion over the short lengths of optical cable involved in home theatre/hi-fi. So it is reasonable to say that you either get a perfect signal or not. I also doubt whether over long runs the signal would decay into a sine wave - this is probably only applies to electrical signals.

As for the book I read, it was the only one at the library which had anything about digital audio. It wasn't a particularly easy read.

 

[Hipnotic4]

This is totaly irrelavent but just wondering if they make any digital opt cables with a clear jacket?

 

[MDS]

tbewick,
Which Watkinson book did you read? The Art of Digital Audio or An Introduction to Digital Audio? The Art of Digital Audio is a bit more dense and similar to Principles of Digital Audio by Pohlman. Watkinson and Pohlman are probably the two best known authors on the subject.

If you really want your head to spin, pick up Digital Audio Processing by Doug Coulter. I admit that I didn't really follow alot of what he was saying in that book and I am familiar with the low level computer gorp.

A Programmer's Guide to Sound by Tim Kientzle (out of print) is alot easier to read. Digital Audio with Java by Craig Lindley is pretty good too. Although those two books show alot of code (which may be lost on non programmers) they both cover the basics in an approachable manner. Reading books like that first will make it easier to follow the more advanced books. If you started with The Art of Digital Audio, you jumped in head first!

Mastering Audio: The Art and Science by Bob Katz is a mix of technical and practical advice. He explains Jitter particularly well. He is a well known mastering engineer (mostly for his work with Steely Dan).

 

[jaxvon]

Mastering Audio: The Art and Science by Bob Katz is a mix of technical and practical advice. He explains Jitter particularly well. He is a well known mastering engineer (mostly for his work with Steely Dan).

I'll need to pick up some of these books. I have a correction though, the producer that did a lot of work with Steely Dan and Fagen was Gary Katz, not bob.

 

[Hipnotic4]

Nicee i gotta get some of those books 2

 

[tbewick]

Hi MDS,

I read the 'Introduction to Digital Audio' book. The book doen't offer much analysis of the explanations, so you could read it and leave without any idea how much you need to spend on equipment. I suppose such a book that did offer such information would be titled something like 'Practical Digital Audio'. Out of interest, seeing as this is such a regular sort of question, do you know of any links to sites that do proper physical tests comparing cables? This, to me, would offer a more rigorous coverage of the topic, rather than just going on the accepted principle of using short lengths of cable.

It's rather interesting with Bob Katz. I've seen his site, Digital Domain, and he seems to offer seemingly contradictory advice. For example, on an article about jitter, he initially seems to dismiss the importance of jitter, but then goes on to say how jitter can introduce (subjective) distortion on playback. Again, I think that his site, though interesting, doesn't really offer much practical advice. One of his conclusions is to buy a DAC that has high jitter reduction. I've never seen this sort of performance information on any CD player or digital receiver.

 

[WmAx]

1st thing: would a moderator please move this thread to the cables forum?

e.

It's rather interesting with Bob Katz. I've seen his site, Digital Domain, and he seems to offer seemingly contradictory advice. For example, on an article about jitter, he initially seems to dismiss the importance of jitter, but then goes on to say how jitter can introduce (subjective) distortion on playback.

On one part of his site, Katz also discusses analog audio and how it's audibly superior to most digital systems, going on about the subject with all sorts of typical audiophile subjective descriptions of sound. He may be a fine mastering engineer, but without supporting perceptual research to back up claims of audiblity, one can not take such claims seriously.

Katz is a mastering engineer; he may be referring to potential additive jitter in a long line of processes, if multiple AD/DA or non-master clock linked digital audio devices are used in conjunction with each other.

Based on the research I have read(I refer to blind test research using multiple test subjects, not the worthless joke whitepapers that are referencable from a site like stereophile), the amount of jitter-error to become audible in music playback is massive, relative to the amount induced by common digital electronics or cable transmission problems. To become audible on pure sine wave test tones, a very sensitive listener, under lab conditions, using headphones for maximum sensitivity, may detect 3 ns of jitter at just the right jitter frequency. This is worst case scenario, under un-realistic conditions, and this is a higher value than an average DAC(though some very poorly designed units with odd functionality/design may induce slightly more than this value) will produce. With music, using selections that contain isolated sections of instruments, to increase sensitivity, the most sensitive listener under the most sensitive conditions(headphones, etc.) could detect 20 ns in one instance.

Refer to an old thread which was concerning optical cable jitter, and in which I provided some result graphs of Dolby Labs research:

http://forums.audioholics.com/forums/showthread.php?t=4547

-Chris

 

[tbewick]

Thanks Chris for the link.

The Audioholics article on optical cable suggests that buying a good quality optical cable is worthwhile -

"The second concern, data error, is easier to understand - data arriving at the D/A converter is either missing or wrong. This would be very detrimental to sound quality and often results in loud popping noises or even muted silence. Data interpolation, the reconstruction of damaged data, would result in less obvious, but similarly compromising, effects on fidelity."

Is the necessity for data interpolation a common occurrence with Toslink connections? This contradicts the idea of having either a perfect signal or no signal at all. The book I read on digital audio said that in a properly working digital system, the need for data interpolation due to errors should be minimal.

"Today the TOSLINK optical interface has reached its sonic performance potential. What changed? There are three factors that draw the line separating mediocrity from excellence: purity of the optical conductor, clarity of the optical termination (lens), and flexibility without signal loss. When each of these parameters is optimized, and the interconnect is used with quality components, the TOSLINK connection is audibly indistinguishable from coaxial S/PDIF."

Wouldn't this mean that it's better value buying a cheap coaxial cable rather than shopping around for a high quality optical cable? The fact that a contributor to the article is a company that manufactures cables does lead to the feeling of possible bias.

 

[WmAx]

Wouldn't this mean that it's better value buying a cheap coaxial cable rather than shopping around for a high quality optical cable? The fact that a contributor to the article is a company that manufactures cables does lead to the feeling of possible bias.

Please refer to the jitter values claimed in said article concerning cheap optical cable. It's a bogus[unsubstantiated] claim that these have any relevance to audibility in a typical length distance connection.

-Chris

 

[markw]

So, I'm confused...

Has someone here claimed to have heard the effects of jitter with either expensive or inexpensive toslink or coaxial cables? ...or is this simply arguing about the number of angels that can dance on the head of a pin.

If I read wick's last post, it basically says that when all is right with a toslink connection, which involves several parameters, it's just as good as a plain vanilla coaxial connection.

Again, what IS the point of this thread?

 

[jonnythan]

Has someone here claimed to have heard the effects of jitter with either expensive or inexpensive toslink or coaxial cables? ...or is this simply arguing about the number of angels that can dance on the head of a pin.

If I read wick's last post, it basically says that when all is right with a toslink connection, which involves several parameters, it's just as good as a plain vanilla coaxial connection.

Again, what IS the point of this thread?

The point, ultimately, in response to the original query, is that there is no difference between a $100 Monster Cable toslink cable and a $5 Cables For Less one.

 

[tbewick]

That's cleared it up for me. The only point that confused me was the possible need for data interpolation due to errors. I'm guessing that if there was something seriously wrong with the Toslink cable necessitating a lot of data interpolation, then clicks would give it away. Hopefully the rest of the time, data interpolation either isn't required or is so infrequent as to be inaudible. In other words, you get an effectively perfect signal or you don't get a signal at all.

 

[markw]

In other words, you get an effectively perfect signal or you don't get a signal at all.

This is true for any cable, toslink or coaxial. I would think this was also made clear by the number of words expended in htis thread by now.