digital coax or optical cables?

[GMichael]

My system will let me use digital coax or digital optical cables for audio to and from: Yamaha RX-V2500. Is one any better than the other? My first thought would be to use the optical because it can not be affected by electro-magnetic fields around it. But the manual says that if I were to connect both, the receiver will default to the coax and ignore the optical. To me, this suggests that the coax are better. Has anyone ever tested these and found one to be better than the other?

[noddin0ff]

Either should be fine. I think coax is cheaper and easier. Some say that cheap optical cables are inferior (cheap plastic optical cable is not as robust for transmitting light). But a digital signal is a digital signal. It shouldn't matter how you connect, especially over typically short distances.

[hermanv]

The inexpensive optical cables are plastic fiber as opposed to the expensive glass fiber cables. Plastic cables have far more time smear (results in jitter) than glass. Coax has very little time smear.

If your digital to analog converter includes jitter correction it will make little or no difference but if it doesn't, the coax offers a quality advantage. This advantage is fairly small but nevertheless real.

So (IMHO) for inexpensive cables you get the best quality per dollar for coax. If you buy video cables like those used for composite video (the ones with the yellow RCA jacks) they seem to be cheaper than the ones marked "digital" both are coax . Unlike others, I've heard no improvement with increased prices for coax, but my D to A has an excellent jitter recovery scheme.

Most people (including me) that use these forums are concerned with small or even miniscule improvements, why not get the small improvement coax offers?

[GMichael]

Thanks guys. I'll look into getting a coax cable. Too bad I already let the guy at Tweeters talk me into the optical cable. Monster cable ta boot. (yuck) Forgive me for I knew not what I was doing.

[noddin0ff]

It's my understanding from what I've read that jitter correction might have been a problem years ago. But now all DAC's are better and cheaper and all have sufficient jitter correction. It is true that cheap plastic optical doesn't transmit light as well and is more prone to light loss at the ends, but in a typical home set up, I don't think this makes any difference. Every now and then someone claims to get good digital connections with wire coathangers. That may be urban legend, but I tend to believe it could be true.

[kexodusc]

The coat-hanger expirement is worth reading up on...

Most material I've seen on the issue of jitter suggests it isn't a problem for cable lengths less than hundreds of feet as we're talking about infinitely small numbers for jitter. Also, that most DAC's have correction for this now, and that the information is processed slower than the time it takes to travel anyway, making the reality of jitter all but zero.

I won't pretend to know what any of this fangled stuff means, though.

Who knows? I've never heard an optical cable that sounded "not right"...they either work or they don't.

Wanna play it safe, get the coaxial cable.

[noddin0ff]

I think a lot of the jitter concerns with optical cables was not necessarily due to length but due to a mismatch of refractive index at the ends. The plastic cable having one RI and the receiving optical device on the component it is attached to having another. This can cause reflections and result in whatever timeshifting is. Having the RI well matched through out the signal path is ideal. I assume this is all standardized now, but that would be an area where a cheap optical cable might be less than perfect. Either way jitter is not suppose to be a problem (if it ever was) because error correction takes care of it.

[hermanv]

Unless the jitter is so large that one bit time spills into another bit time and corrupts the value of the adjacent bit, error correction does not fix jitter.

Most digital to analog conversion circuits have a clock recovery circuit that restores the correct timing, but not all clock recovery circuits are equally effective.

This is hard in words but easy with graph paper: Digital (Redbook CD) sound places a dot on the graph paper in one of 65,536 absolute levels 44,100 times per second for each channel. The reconstruction filter then connects the dots converting the quantized digital signal back into a continously varying analog voltage. If the dot is placed a little early or a little late the connect the dots waveform will be changed in much the same way it would have been if the dot were placed a little high or a little low.

It is a matter of degree and there is obviously some point beyond which the effect is no longer detectable. But for me at least I hear differences between plastic optical cable and coax cable that my training tells me is most likely due to jitter. When I bought a jitter correction box the differences I heard between the plastic and the coax effectively dissapeared.

My newest DAC simply writes the data into a RAM buffer and reads it out with a very stable clock. With this piece of equipment I can hear no audible change between optical and coax signals.

[noddin0ff]

Unless the jitter is so large that one bit time spills into another bit time and corrupts the value of the adjacent bit, error correction does not fix jitter.

I think you meant to say that IF the jitter is large enough to corrupt the adjacent bit THEN error correction cannot fix the data. Which is true. But I believe this would be very very rare. It would require bits to be nearly perfectly out of phase...

Most digital to analog conversion circuits have a clock recovery circuit that restores the correct timing, but not all clock recovery circuits are equally effective.

Are you referring to reclocking? I'm not any expert so my knowledge is limited and could be wrong. I think it is now common for DACs to reclock the data. I presume they do this whether jitter is present or not. So I'd think a bad reclocking circuit, if audible, would sound bad regardless of jitter, if any, in the signal. Even though CD's encode 16 bit music info, the actual data stream is 32-bit words. In these words are bits that contain synchronization info that can be reclocked. Jitter should rarely, if ever, obliterate this info. So as you said the D/A conversion can reclock the data. This link was interesting if your interested. It's where I read about the 32 bit word structure.
http://www.epanorama.net/documents/audio/spdif.html

This is hard in words but easy with graph paper: Digital (Redbook CD) sound places a dot on the graph paper in one of 65,536 absolute levels 44,100 times per second for each channel. The reconstruction filter then connects the dots converting the quantized digital signal back into a continously varying analog voltage. If the dot is placed a little early or a little late the connect the dots waveform will be changed in much the same way it would have been if the dot were placed a little high or a little low.

It is a matter of degree and there is obviously some point beyond which the effect is no longer detectable. But for me at least I hear differences between plastic optical cable and coax cable that my training tells me is most likely due to jitter. When I bought a jitter correction box the differences I heard between the plastic and the coax effectively dissapeared.

My newest DAC simply writes the data into a RAM buffer and reads it out with a very stable clock. With this piece of equipment I can hear no audible change between optical and coax signals.

I think you're right here. I don't have any listening experience in this arena, but I'll accept that people with the right ears and right gear can hear the effects of jitter. You say your newest DAC has a buffer and a clock for reclocking. I think that with the forward march of electronics, a data buffer and clock are pretty standard for all but maybe some very cheap components. I'll hazard a guess that after the last few years, the problems you were able to hear exist no longer because the equipement better. But I'll admit that better quality DAC and clock would produce better signals. But since they are getting better/cheaper fast, quality differences should soon surpass the human ability to distiguish the differences.

[hermanv]

All clocks have jitter its a matter of how much. Re clocking data from a CD requires a clock circuit that can track the exact clock rate coming from the CD player, they will not be absolutely exact and even a small error would eventually cause the receiver to either run out of bits or to overrun the incoming bits so these clocks are designed to phase lock to the incoming stream.

This very phase lock process means that the clocks can not be as stable as we would really like them to be. In general I agree, the stuff is much better than it used to be enough so that outboard D to A converters are becoming uncommon.

I still feel that this function is important enough that I don't want to pay for it twice. I mean so many devices are digital today and each has a D to A, it seems a little silly. I rather pay for one known good one and I think that it belongs in the pre-amp. That way you can switch analog sources or digital sources all with the box you usually think of as the source selector. The receiver people have figured this out and are doing it but I think except for Meridian most high end people are not including the D to A function in their pre-amps

First class D to A chips and clock chips are still not cheap enough that you will find them in that $100 CD player.

[vr6ofpain]

so what was this about coat hangers?

[theaudiohobby]

It is a matter of degree and there is obviously some point beyond which the effect is no longer detectable. But for me at least I hear differences between plastic optical cable and coax cable that my training tells me is most likely due to jitter. When I bought a jitter correction box the differences I heard between the plastic and the coax effectively dissapeared.

which jitter correction boxes did you look at. thanks.

[hermanv]

I was offred a Monarchy model (sorry don't remember the number) that has the optional extra long coax driver and the BNC connector for I think it was $150, new. If memory serves that was about half the going price.

You should be able to find a used one for that amount or less as fewer people are using outboard DAC these days. Good time to pick up a first class DAC cheap too, most of them have jitter fixes built in. If you buy an external box remember you'll need two digital cables now.

I still prefer the outboard solution, I've worn out three CD players, medium priced. I have a decent system and reasonably good ears and can hear no difference from one player to the next once the digiital signal is properly conditioned and externally decoded. I realize not everyone agrees with this assesment.

[JBMAudio.com]

My system will let me use digital coax or digital optical cables for audio to and from: Yamaha RX-V2500. Is one any better than the other? My first thought would be to use the optical because it can not be affected by electro-magnetic fields around it. But the manual says that if I were to connect both, the receiver will default to the coax and ignore the optical. To me, this suggests that the coax are better. Has anyone ever tested these and found one to be better than the other?

Simple rule of thumb here. For runs under 100 feet, go for Coax, it is cheaper, more durable, and just plain easier.

[noddin0ff]

Simple rule of thumb here. For runs under 100 feet, go for Coax, it is cheaper, more durable, and just plain easier.

Are you saying for runs OVER 100 feet go for optical? Or just being humorous in suggesting to always go Coax. Because a long run of optical would be very expensive and likely impractical...

[GMichael]

Over 100 feet? Would this be for a remote location? DVD player in one room that is zone 2 for the receiver? Or do some people just have larger rooms than I'm used to seeing? LOL.

[JBMAudio.com]

Are you saying for runs OVER 100 feet go for optical? Or just being humorous in suggesting to always go Coax. Because a long run of optical would be very expensive and likely impractical...

Because of the distance, optical would be a better choice for runs over 100 feet, but since hardly anyone ever has these long of runs, coax is the way to go

[Monstrous Mike]

Are you saying for runs OVER 100 feet go for optical? Or just being humorous in suggesting to always go Coax. Because a long run of optical would be very expensive and likely impractical...

The Toslink optical interface found in home audio is only designed for maximum runs of 25 to 30 feet. It is not a normal optical connection but was designed by Toshiba to allow for the use of inexpensive (plastic) cables with inexpensive connectors. (The challenge in real glass fiber optical cables is the glass to connector manufacturing.) Real optical cabing and interfaces would be too expensive for home audio users.