Age old debate Dig. Coax or optical

[mrohde]

My Krell HTS 7.1 has just arrived to go with my Krell showcase 7 amp and B&W 803’s (running a 7.1 system) speakers. I looking to optimize audio performance anywhere I can. So, having said that, and at the risk of staring one of those ago old battles of opinions here is the question:
Which is better in regard audio quality and bandwidth (if that is even an issue) digital coax or optical for audio from the cable box and my DVD player (Denon 5910CI).

Before anyone asks, I am not using HDMI at this time. My usage is music 30%-40% music (2 channel and 5.1-7.1) and movies the rest of the time.

All opinions are welcome.

Thanks

[David Meek]

Digital coax for me. Toslink (optical) has always imparted a harshness or "grain" that I find unpleasant. I have noticed that true glass Toslink has much less of this character than the plastic version.

I'm sure the "a bit is a bit" types that don't hear qualitative differences will join in shortly - and their opinions are just as valid - but that's what I hear in my system, and have heard in others, also.

Edit: Having said all the above, I just ordered a glass Toslink as my HD-Dish tuner doesn't have a digital coaxial out - HDMI and Toslink only, and the HDMI is a direct run from the tuner to the plasma. Necessity. . . .

[Alaric]

My HTIB is pretty weak sound-wise , so I probably couldn't tell if the sound was mailed to my receiver! My reason for Toslink usage is a shortage of digital coax inputs.

[whoaru99]

If you are not using Denon Link (or HDMI??), you will also need to use 5.1 analog interconnects for the hi-res multichannel audio from SACD/DVD-A. Info provided just in case...

[alebonau]

not sure the need for an age old battle. you got the gear there try it both ways and decide for your self. if one is better than the other you shoudl hear the difference. if you cant hear a difference maybe there isnt any ?

[Chris D]

Hey, man! I personally use coax whenever I can, without getting into the really dirty details. Sometimes there's only optical, and that works fine for me without degradation, but if I have a choice, it's coax.

5910ci, huh? Cool stuff.

[George Bellefontaine]

Same as Chris.

[spyboy]

With that kind of equipment you probably should go with coax. I have a mix of coax and optical and I personally can't tell the difference.

[aud19]

Same as Chris.

Me too.

Generally speaking, you can get either better quality coax cables for your money than optical or equivelant quality costs you more money. Not too mention optical is more fragile and the additional conversions from electrical to light and back again.

Given similar quality there should be little to no difference between them though.

FWIW, forum sponsor Cat cables has a VERY nice silver digital coax with rave reviews as well as quality real glass optical cables available

[Chris D]

optical is more fragile and the additional conversions from electrical to light and back again.

Yup. Without touching the nitty gritty details, that's the essence of why I choose coax when I can.

If we're bringing up Cat Cable's silver digital coax, I have three of those in my system, and they're positively stunning in every way.

[RobP]

FWIW, forum sponsor Cat cables has a VERY nice silver digital coax with rave reviews as well as quality real glass optical cables available

:agree:


Coax for me.

[Amphiprion]

Optical cables have one huge advantage - electrical isolation. You won't get any ground loops with two pieces of equipment connected with toslink.

And as for converting to light and back to electrical signals causing degradation, I garauntee you it does NOT happen. I use 100Mbps digital optical isolators at work that go electrical -> optical -> electrical solely for the optical isolation on circuit boards. I have voltage spikes >500V on one side of the isolator and everything works hunky dory, no problems, no errors, nada. This is with your standard 5V logic stuff.

But if you don't have ground loops or any problems like that, I'd go coax. The toslink connectors suck. But you won't hear any difference between the two, so long as the toslink cable doesn't get bumped/nudged/pulled out of the socket.

[Kevin P]

And as for converting to light and back to electrical signals causing degradation, I garauntee you it does NOT happen. I use 100Mbps digital optical isolators at work that go electrical -> optical -> electrical solely for the optical isolation on circuit boards. I have voltage spikes >500V on one side of the isolator and everything works hunky dory, no problems, no errors, nada. This is with your standard 5V logic stuff.

I'm sure your digital optical isolators work great, but I'll bet you $5 that they're quite a bit more sophisticated (and expensive) than the 5 cent LED and 5 cent photodiode encased in plastic connectors joined by a (usually) plastic piece of fiber cable that passes our precious digital audio from component to component in your system.

Sounds like you're some sort of electronic tech... try making an opto-isolator out of a cheap LED and a cheap photodiode and see if it performs as well as your 100Mbps parts...

[Race Car Driver]

I feel like the dumb kid in the smart class right about now..

Or maybe the dumb kid in the dumb class?

optio-isolator photodidoe? hrmm.. I should know this.

Oh well, continue...

[spyboy]

If the toslink connectors are not disturbed, there is no difference in sound quality between coax and even inexpensive toslink cables, IMHO.

[mjb]

Optical cables have one huge advantage - electrical isolation. You won't get any ground loops with two pieces of equipment connected with toslink.

Seconded! Optical is my choice for this reason. But, really it makes no difference. A bit stream is a bit stream however its transported.

[Kevin P]

It's not just that the bits get from point A to point B, the timing of the bits is important too. In most cases, coax beats optical for sound quality, because for Toslink the bits go through a 5 cent LED that has a high slew rate (slow response time). This causes the sharp edges of the waveform (edge transitions) to become more sloped. Reflections within the plastic cable, and the cheap photodiode on the receiving end contribute to this as well. This then gets passed to a comparator or Schmitt trigger which converts the rounded-off transitions back into sharp ones, but the timing of the transitions can be affected by this conversion depending on the threshold of the trigger, efficiency of the optical components, sensitivity of the photodiode, etc. This results in jitter, which can and does effect the sound quality.

Coax suffers from the same "problem", but to much less degree on average compared to inexpensive Toslink interfaces. Consider how people report the differences using silver coax (Silvercat Digital) vs. copper coax. This is due to the silver cable transmitting the edge transitions more precisely.

If the optical interfaces were like the ones telecom companies use for their fiber links (laser transmitters and glass fiber), then optical would indeed be the superior format.

It would be interesting for someone with a dual-trace oscilloscope to connect both the coax and optical from a source to a receiver, and then connect the scope to the points on the digital board on the receiver to compare the signals from the coax and the optical, to see how they differ.

[spyboy]

It's not just that the bits get from point A to point B, the timing of the bits is important too. In most cases, coax beats optical for sound quality, because for Toslink the bits go through a 5 cent LED that has a high slew rate (slow response time). This causes the sharp edges of the waveform (edge transitions) to become more sloped. Reflections within the plastic cable, and the cheap photodiode on the receiving end contribute to this as well. This then gets passed to a comparator or Schmitt trigger which converts the rounded-off transitions back into sharp ones, but the timing of the transitions can be affected by this conversion depending on the threshold of the trigger, efficiency of the optical components, sensitivity of the photodiode, etc. This results in jitter, which can and does effect the sound quality.

Coax suffers from the same "problem", but to much less degree on average compared to inexpensive Toslink interfaces. Consider how people report the differences using silver coax (Silvercat Digital) vs. copper coax. This is due to the silver cable transmitting the edge transitions more precisely.

If the optical interfaces were like the ones telecom companies use for their fiber links (laser transmitters and glass fiber), then optical would indeed be the superior format.

It would be interesting for someone with a dual-trace oscilloscope to connect both the coax and optical from a source to a receiver, and then connect the scope to the points on the digital board on the receiver to compare the signals from the coax and the optical, to see how they differ.

Good thoughts. Please explain exactly what jitter sounds like. As far as I know jitter is not really an issue.

[chrispy35]

'audible modulation' according to an AES paper.

Chris P.

[spyboy]

Thanks chrispy

My oscilloscope is in the shop

In the mean time, I will stick with any old optical cables that I have.

Actually, I am waiting for my beryllium coax cables to come in

[audioqueso]

I've said it before... and I'll say it again...

I can't believe this thread was started on THIS forum.

Unless your equipment has an optical connection that is inferior to the coax (or vice versa), you are not going to hear the difference between coax and optical in such a short run in most common setups. Best thing is just to try it out for yourself.

Being 100% honest, it's a psychological thing more than anything else. I prefer coax for the simple reason: It's one less processing step... and that's it. Taking the 0s and 1s and changing it to an optical signal, then converting it back. Coax is straight 1s and 0s from equipment to equipment. No one can truthfully hear that processing difference until their equipment is crapped (as mentioned above about one connection/convertor being inferior), but I like coax because I know in the back of my mind it's one step shorter in processing.

[audioqueso]

http://www.htguide.com/forum/showthread.php4?t=16791&page=1&pp=35&highlight=coax+optical

[spyboy]

Does anyone have a 8,000 foot optical cable that I can use to try to hear the difference?

[audioqueso]

Does anyone have a 8,000 foot optical cable that I can use to try to hear the difference?

You'll hear a difference. Optical will have audio. Coax will not (since it won't carry a signal that far). :B

[Race Car Driver]

I have a coiled 8000ft optical cable out on a flatbed in the driveway, you pay for shipping you can have it

[spyboy]

I have a coiled 8000ft optical cable out on a flatbed in the driveway, you pay for shipping you can have it

How much would it be to overnight it?

[mrohde]

You have to love this thread! Thanks for the info and the entertainment!

[Victor]

It's not just that the bits get from point A to point B, the timing of the bits is important too. In most cases, coax beats optical for sound quality, because for Toslink the bits go through a 5 cent LED that has a high slew rate (slow response time). This causes the sharp edges of the waveform (edge transitions) to become more sloped.

Actually ‘high slew rate’ would result in faster response time. It is however not important here, as both coax and optical interfaces do use edge conditioning in the same or similar manner. The sound quality is not an issue here also because the THD+N is virtually identical regardless of which interface you use. That is to say that for a given length of cable you can get the same bit rate.

Reflections within the plastic cable, and the cheap photodiode on the receiving end contribute to this as well. This then gets passed to a comparator or Schmitt trigger which converts the rounded-off transitions back into sharp ones, but the timing of the transitions can be affected by this conversion depending on the threshold of the trigger, efficiency of the optical components, sensitivity of the photodiode, etc. This results in jitter, which can and does effect the sound quality.

It is true that jitter-related distortion might be higher with Optical interface. It has nothing to do with Schmidt trigger, since it is used in both interfaces. The signal reflection due to the impedance mismatch is more evident with optical interface, although I must admit that it is not true all the time and some optical interfaces are actually implemented quite well.

However, if no jitter control is used, the differences in Signal-to-Noise ratio between the coax and optical interfaces are only of academic importance with coax having a marginally better specification. Clearly this difference can never be heard by humans, as it is infinitely difficult to measure reliably and repeatedly. Very few labs, and you can count them on one hand on North America, can do this trick. Nearly all of them report to the State Department and do not publish their findings for obvious reasons.

On the other hand if jitter control is used then it is an open field here, because the end result would 100% depend on the quality of the electronics used. In this case the interface is irrelevant as absolutely identical results are produced.

Coax suffers from the same "problem", but to much less degree on average compared to inexpensive Toslink interfaces. Consider how people report the differences using silver coax (Silvercat Digital) vs. copper coax. This is due to the silver cable transmitting the edge transitions more precisely.

People do report all kinds of things all the time. Nothing can be done about that. Good science, on the other hand, must be repeatable and statistically significant.

Well, - there is no statistical significance in sound quality issues between two interfaces in the context of a typical commercial audio equipment.

I use coax because it offers better connectors, less signal reflections, less potential noise and better reliability. I do not consider the sound quality as one of the issues.

[spyboy]

Thanks Victor

Your last statement makes sense.

I am still using some crappy old optical cables. If I had the hearing of a bat, it might make a difference if I used coax.

PS Did you design a couple of amps some years ago? If not, thanks anyway.

[Victor]

PS Did you design a couple of amps some years ago? If not, thanks anyway.

...it is off topic here, but, yes, I did design power amps in the past. I no longer do it.

[Chris D]

Really? Sounds like an interesting discussion. Perhaps you could tell us all about it in a new thread?

[gianni]

Chris,

If you search his previous posts, you will find that he has shared some information as to some of his background. He did design work for a very well known high end brand (which he mentions). You can search his previous posts or perhaps he'll share if he desires.

I will say that from his posts I draw the following conclusions:

1) he knows his stuff and is very objective. If he can't explain the science behind it, and does not have direct experience, he's not going to venture a guess.

2) He is very careful to differentiate between his opinions and fact - something any forum could use more of.

3) He is an asset to this forum and I hope he sticks around.

Chris, If you can convince him to share a little about his experience with amp design/performance all in one place such as a separate thread or sticky, my hat is off to you. Of course, that may ignite a fire around here with those that don't agree with him. From reading his posts, I'd say Victor does not believe in or waste time with audio snake oil, black magic or hocus pocus. Could be interesting....but I'd be surprised if he bites.

[Victor]

Chris,

Although my comments have not been in general favor here and I am known for attracting the wrath of Gods, still I’ll be happy to discuss the aspect of power amplifier design as I see it. In fact it does not have to be a power amp topic, it can be any topic that I can constructively discuss, - that does not make it anything, but I’ll try. We can do it in this thread or in a new one, - what would you like to talk about?

Vic

[Chris D]

Well, tell you what--what would you think, if for starters, you started a new thread where you just tell us all about yourself? We're always interested in members telling us about them, where they're from, what their interests are, what their current A/V setup consists of, what they've owned in the past, etc. If you have a past in the A/V industry, perhaps you could share with us what your education is, how you ended up there, the kind of work you've done, etc.

I think that kind of stuff would be interesting and useful for others. Just as long as we avoid pitfalls like things that would be specific product advertising for companies. How does that sound?

[spyboy]

Hi Victor

This may sound strange, however, I found my Nakamichi to have something I call "harmonic correctness". The sound of real instruments, (as opposed to electric guitars), just sounded more natural to me. On the other hand, the Nakamichi seemed to have slightly less bass than other amps.

My Denon AVR-2500 has a lot of "slam", but is not quite as "natural" sounding.

Now I have a Denon 5803A that got great reviews from Gene DellaSalla (sp?). I wonder if a Sunfire Cinema Grand would sound better than the Denon 5803A?

My Citation 7.1 (425 X 2 @ 4 ohms), sounds just slightly better than my Kenwood VR-6070 which tested at 92 X 6. I think the 5803A tested at 117 X 6 (7?).

Thanks in Advance

[Victor]

Hi Victor

This may sound strange, however, I found my Nakamichi to have something I call "harmonic correctness". The sound of real instruments, (as opposed to electric guitars), just sounded more natural to me. On the other hand, the Nakamichi seemed to have slightly less bass than other amps.

My Denon AVR-2500 has a lot of "slam", but is not quite as "natural" sounding.

Now I have a Denon 5803A that got great reviews from Gene DellaSalla (sp?). I wonder if a Sunfire Cinema Grand would sound better than the Denon 5803A?

My Citation 7.1 (425 X 2 @ 4 ohms), sounds just slightly better than my Kenwood VR-6070 which tested at 92 X 6. I think the 5803A tested at 117 X 6 (7?).

Thanks in Advance

spyboy,

I like Sunfire products a lot. It is clearly an interesting and unique design. It is not a ‘me too’ design for sure. Keep in mind though that Bob Carver never tried to create an extremely clean amp, - he wanted the most powerful and ergonomic design with reasonable THD specs. So the THD is not stellar, but ok.

I cannot comment on the sound of Nakamichi or Denon product as I have no idea what makes them tick. But my guess is that you are trying to drive them too hard by demanding more power then they can give before the distortion sets in. If used at lower power levels I think the problems you describe might disappear.

Victor