Thoughts on Audio: Digital Coax or Optical? (1 Viewer)

[Drake]

I know it's mostly a matter of opinion, but that's why we're here, right? I've shied away from the optical connections on my progressive DVD and DBS system in favor of the digital coax route. Actually, the DBS system is still RCA because I haven't had a chance to get the digital audio upgrade cables.

Optical connections just don't seem quite as tight to me. Honestly, I haven't even tested if there's a difference in sound quality between the two.

Before I go out and help fuel the economy with new optical cable purchases, what are some of the experiences of forum members using digital coax vs. optical?

Thanks for your feedback...

 

[Philip Hamm]

Though this topic is very compelling as an advnaced topic, I think it would get more and varied answers (and plenty of flames - people behave yourselves!) on the Cables & Interconnects forum, so I'm moving it.

My opionion: It makes no difference. Theoretically Coax is more suseptible to electronic interference so the cable quality makes more of a difference for coax connections. That said I use a dirt cheap 75 ohm coax cable I got for $7 from a harware store in my system for my digital coax connections. The only digital cable where I've ever had real problems is the LD RF connection, which I have a premium AudioQuest cable for and that works fine.

Now, analog connections......

 

[Mark Austin]

My opinion: Coax all the way. Tried a couple of optical cables for several months, and they always seemed harsh, and compressed.
Phillip,
Would you be up for a little experiment?

 

[Ron Boster]

Drake:

Welcome to the forum. The quickest way to find your answers is in the search function. In the last 30 days this question had come up and been discussed. This is always the best place (and quickest place) to find answers to basic HT and HT equipment questions.

Good Luck

Ron

 

[Chriss M]

can someone please explain how it would be possible for one type of digital cable to sound different from another?

It seems to me that it shouldnt make any difference at all how the 1's and 0's get to their destination.

 

[Duncan Barth]

Chriss M:

That question opens up a *huge* can of worms.

In summery: some people think 'bits are bits' and as long as the bits get there, it doesn't matter if you're using a coat hanger (literally) or some super-awesome-pure digital interconnect.

The other camp believes there is a diffrence due to effects of jitter, differences in signal propegation, or just simpily they tried both and one 'sounded better'

I'd run a search through the archives to see the different points of view... this has been debated many times over in the past.

 

[Philip Hamm]

Jitter (a common term for the mistiming of bits in a PCM bitstream) should not be in any way related to the carrier, only the transport.

 

[RicP]

Jitter (a common term for the mistiming of bits in a PCM bitstream) should not be in any way related to the carrier, only the transport.

"should" being the key term there. Unfortunately Philip, the carrier does affect the jitter rate of the signal. Sometimes for the better, sometimes for the worse, sometimes (most times?) that effect is negligible.
BTW, anyone who thinks that 1's and 0's are running down the wire...that's not how it works at all. The signal being transferred is analog, either light or electricity. Last time I checked, electricity is not digital. The result of the electrical signal is a packet or data stream that can be interpreted as digital data, but the signal itself is analog and therefore subject to anything that an analog signal is subject to. The timing clock is also transmitted along with the analog signal across the wire, and the clock is what is affected by jitter.
http://www.jitter.de

 

[BryanEB]

Is there any sort of CRC done on digital audio signals? I would think any sort of error introduced to the signal on the wire would result in a VERY noticeable problem (dropout etc.).

Is this not true? I would think like a modem etc., the resulting digital data has to be perfect for any of it to work.

Bryan

 

[John Morris]

Coax.

 

[Glenn Overholt]

Ok, this is a nice and fairly quiet war. You might want to partially base your answer on the sockets on your equipment, and if you do make purchases in the future, (which we all do), will you still be able to use it?

Most of the new receivers that I have looked at have more optical sockets than coax ones.

As for the loose connection with an optical, I went out a few months backs and picked up an RCA optical, and it was sooo loose in the socket that I could not believe it! I went back and paid more for a Monster optical, but now the fit is tight! I am much happier.

Glenn

 

[Howard_S]

I think the reason many of us don't hear a difference is because we don't have the equipment. As mentioned in either this thread or another thread, most high end components only have coax or have both and never just optical. There is also a reason why every store that sells only high end equipment says coax all the way.

I think the reason receivers offer more optical than coaxial connections is because there are far more components out there with just an optical connection. Lowerend cd players, md players, dvd players etc. not to mention game systems such as xbox.

 

[Philip Hamm]

The timing clock is also transmitted along with the analog signal across the wire, and the clock is what is affected by jitter.

Note, this is only an issue with non-DVD-Audio PCM sound. DTS and DD reclock the digital bits in the decoder. I'm not sure about DVD-A PCM and SACD DSD (is that the correct term?).

If you go back to the "olden" days, there was a laser reading 1s and 0s from a CD. The CD is spun by a motor. The digital PCM bitstream contains a series of 1s and 0s. The timing of the 1s and 0s is a critical component of this signal. And you can believe with 44 thousand samples per second the transport has to be damn precise in order to time them perfectly.

I do not believe that a cable can effect a change in the timing of the bits in a PCM bitstream.

 

[RicP]

Quote:

Pretty much correct although in actuality it is much more sophisticated than that. There is the Solomon-Reed error correction, there is bit interpolation to replace missing bits in the stream, and all of that had to be performed in less than 1/44100 of a second. Incidentally those 44100 samples a second are each 16 bits wide. So every second a CD laser/servo must read 705600 bits and perform error correction and interpolation in addition to clocking them properly.

 

[ChrisMM]

A couple of points or questions I would like to make in this thread.

I just purchased a Denon 2802. It has only one Digital COAX and 3 optical connections. why 1 vs 3? Are the optical cheaper and hence can add more without effecting the manufacturing cost as opposed to COAX? Also, the argument was made to me that optical requires a tranceiver on either end so thus adds another layer of conversion which can cost you in terms of signal lost or quality. I don't know if this is true.

Last question, why does my 'Ultralink' COAX digital cable indicate a signal flow? That is to imply that the cable must be connected in a certain manner, thus the signal flowing from say the DVD to the receiver. The cable's terminations on either end look identical, even when removing the outer sleave. I just can't understand what difference it would make which way the cable is connected. I mean basically voltage is applied or un-applied to the signal center conductor causing 'bits' to be formed. Are we worry about eddie currents or electron flow?

 

[Phil A]

The speed issue is why DVD-A will likely go a different route as noted below. The other issue with Coax vs. Optical is the connector. A good connector makes a good electrical connection that allows the signal to be transmitted properly. RCA most commonly, or BNC makes a good connection. Good quality RCAs are not that expensive. The more expensive optical cables have ends that are machined several times to make a better connector and these cables will cost in many cases much more than the average coax digital and perhaps not be so durable over time.
http://www.highfidelityreview.com/ne...snumber=402660
"Digital Interface for DVD-Audio and DVD-Video Specifications
The DVD Forum has released guidelines for a DVD-Audio digital interface via IEEE 1394, also known as Firewire or iLink.
The 129 page document, recently published but dated September 2001, details the transmission of both digital audio and video from DVD-Audio and DVD-Video players through an IEEE 1394 (Institute of Electrical and Electronics Engineers Inc.) bus to peripheral equipment, such as receivers, processors and even digital televisions. The data must comply with the MPEG Transport Stream (MPEG-TS) specifications or the Audio and Music Data Transmission Protocol (A&M Protocol).
The DVD Forum guidelines are divided into two halves to best accommodate the differences between DVD-Video and DVD-Audio content. Using MPEG-TS, both video and audio data can be transmitted, which is particularly suited to DVD-Video material where compressed audio content is often used. The A&M Protocol is optimised for high-quality digital audio such as that from DVD-Audio; linear PCM data will be transmitted according to type – either DVD-MBLA or DVD-SMBLA – via the A&M Protocol that will allow up to six channels with a maximum sample rate of 192kHz to be sent from source to processor.
MPEG-TS and the A&M Protocol can be used simultaneously by a player; for the purposes of DVD-Audio, any video content can be carried via partial MPEG-TS (with the applied DVD-Video audio limitations) whereas any ‘high-resolution’ audio content can be transmitted via the A&M Protocol.
However, under the DVD Forum guidelines, the source component retains the responsibility of primary audio decoding for DVD-Audio material. Within the player, Meridian Lossless Packing is decoded before either being passed to the A&M Protocol converter (in the case of ‘high-resolution’ audio) or the MPEG-TS multiplex, in the case of Dolby Digital or DTS and out via the IEEE 1394 interface. It would appear therefore, that processors capable of receiving and decoding a DVD-Audio stream, will not necessarily require Meridian Lossless Packing decoders, but instead an A&M Protocol converter, since the transmission format is linear PCM.
Any copy protection present (according to format) will be carried via the IEEE 1394 interface.
A digital interface between all types of DVD player and processors, receivers and digital televisions will undoubtedly be a turning point for both software and hardware manufacturers, but most importantly for the purposes of DVD-Audio, should dispense with the need to use six analogue interconnect cables between source and receiving equipment. The advantages are clear; bass management and time alignment can be performed within the digital domain by high-end hardware whose digital to analogue converters and analogue stages are superior to those in the player. The move will also give DVD-Audio a significant advantage over the competing Super Audio CD (SACD) format, for which no digital interface is currently planned."

 

[Philip Hamm]

Well seeing as how neither DVD-A nor SACD even have digital outputs it would impossible to speculate as to what the effect would be.

why not?
Ric, jitter is bit mistiming, most likely caused by the transport. In the CD spec the timing of the bits is basically determined by the precision of the motor. If DVD-Audio or SACD have similar timing issues then jitter definitely still is a consideration. It doesn't matter whether an external connection is used, if the bits are mistimed inside the unit you'll have the same jitter situation. I would guess that the (arguably) audible effects of jitter would be similar in these cases.
The only problem with the lack of a digital output for DVD-A and SACD is there's no way to measure the jitter.
Though I have to say for the record, I would find it almost inconcievable that either DVD-A or SACD has any reliance ont he analog bit timing for audio.

 

[Doug_H]

my opinion is that it depends on your system. I have heard either one sound better on some systems. On my system using Coax makes everything sound bright. On a friends system the optical sounded compressed and dull.

On most systems I have not heard a difference so the best thing to do is try them and decide for yourself.

 

[RicP]

Though I have to say for the record, I would find it almost inconcievable that either DVD-A or SACD has any reliance ont he analog bit timing for audio.

SACD I may agree with you, but DVD-A is still just plain old PCM just like CD, only more data. It is still susceptible to the same issues, even though there is far better corrextion and interpolation involved in the DVD-A spec than there was for Redbook CD.

Plus I'm fairly sure that there's a built in buffer in both SACD and DVD-A machines anyway that essentially negates the data retrieval timing issues. DVD-V already uses buffers by spinning the disc faster than necessary and reading ahead, exactly how Orange Book CD-ROM data is read.

 

[Chip E]

I use coax whenever possible.. As my forum compadre Keith H. likes to say "Their more robust" and i agree whole hartedly... Actually, at this point, the only piece i have that only offers TosLink is a Dish 6000 receiver. No biggie i suppose.