SPDIF Optical Cables, can they perform different?

[solderdude]

But, I was always curious. I don't know the SPDIF protocol, but why didn't they include error correction? Would it have made the transceiver too expensive?

SPDIF was never intended for long distances. TOSLINK is the exact same signal except electrical H = light on, electrical low = light off.
1mm POF cable is not the same as the transmission of data Ray is talking about. This is single mode transmission with a light core of 9micron.
They can reach high dataspeeds by using very narrow band high power lasers of various wavelengths (all infrared) together and using optical narrow band filters.
SPDIF requires a higher bandwitdth because the clock is also included with the data incl. some bits and spaces for syncing.

 

[VintageFlanker]

Thanks, @solderdude

the transmission of data is the same though

That's all that matters to me.

 

[kchap]

SPDIF was never intended for long distances. TOSLINK is the exact same signal except electrical H = light on, electrical low = light off.
1mm POF cable is not the same as the transmission of data Ray is talking about. This is single mode transmission with a light core of 9micron.
They can reach high dataspeeds by using very narrow band high power lasers of various wavelengths (all infrared) together and using optical narrow band filters.
SPDIF requires a higher bandwitdth because the clock is also included with the data incl. some bits and spaces for syncing.

It's more fun talking about DWDM; back to the world of audio. It seems to me there needs to be a new standard for digital audio interfaces. For TOS link and S/PDIF it's issues of distance, data rates, poor jitter characteristics and lack a data/control channel, preferably bidirectional.

USB is more flexible at far as data rates and a data/control channel but, is only good for very short distances. The other problem with USB is
scalability especially asynch USB. Asynch USB is only good for 1 PC/Streamer and 1 DAC. There are standards like AES50 but, they seem to expensive and too proprietary

 

[AnalogSteph]

There are standards like AES50 but, they seem to expensive and too proprietary

Though AES50 is supposedly an open standard...

My goodness, I had no idea how many Audio over Ethernet protocols there are. Interesting stuff. Makes me wonder what Voicemeeter uses, I somehow doubt it's something entirely new and proprietary.

 

[kchap]

Though AES50 is supposedly an open standard...

My goodness, I had no idea how many Audio over Ethernet protocols there are. Interesting stuff. Makes me wonder what Voicemeeter uses, I somehow doubt it's something entirely new and proprietary.

As the Wiki url you refer to shows there are companies that base there product on AES50 but the all seem to have a proprietary element to them.

 

[BitPerfect_]

It doesn't sound different.
Unless the cheap one is so bad data is being lost cauing pops/crackles, then the sound is identical.
[...] and then only if the optical fibre in the cheap cable is so crap that it significantly distorts the optical signal.

And if the optical signal is distorted, we will hear pops/crackles or the sound will be distorted?

I'm asking because this seems to be an interesting subject. Until recently, I've read in different places that an optical cable can't do Hi-Res 24/192 and only some glass fiber cables will be able to carry such amount of data. I've bought a cheap 2m optical cable (a few euros) and BAM! the DAC shows 24/192, and there are no pops/crackles. I have to mention that the source and the dac could deliver and receive 24/192 via optical output/input according to the documentation.

At his point I am asking if the sound could be different if I will use a glass fiber cable that's why I am asking how the distortion will be noticed.
Thanks,

 

[antcollinet]

And if the optical signal is distorted, we will hear pops/crackles or the sound will be distorted?

I'm asking because this seems to be an interesting subject. Until recently, I've read in different places that an optical cable can't do Hi-Res 24/192 and only some glass fiber cables will be able to carry such amount of data. I've bought a cheap 2m optical cable (a few euros) and BAM! the DAC shows 24/192, and there are no pops/crackles. I have to mention that the source and the dac could deliver and receive 24/192 via optical output/input according to the documentation.

At his point I am asking if the sound could be different if I will use a glass fiber cable that's why I am asking how the distortion will be noticed.
Thanks,

If the optical signal is transferred just well enough that the receiver can understand the bits without getting them wrong, then the sound will be identical regardless of what quality or length of cable/fibre is used.

As soon as bits start to get flipped then you will hear gross distortions (pops/crackles/drop outs). There is no circumstance in which digital signal distortion can cause subtle changes to the quality of the sound that you have to listen carefully for, assuming the receiving DAC is halfway competently designed (competent local clock generation successfully eliminating jitter artefacts).

 

[MC_RME]

Fun stuff: grab the TOSLINK connector and pull it out of the socket VERY SLOWLY. You might reach a point where the sound starts to detoriate, like getting noisy, then shuts down. Do the same when inserting. You then know how much 'safety' you have when fully inserting the TOSLINK connector.

 

[eyes-on-you]

Quote from an audiophile site:

—q—

“Not all optical Toslink cables are the same and neither do they have the same kind of filament. Common plastic fiber filament Toslink has a bandwidth of about 6 MHz and this is what has given Toslink such a bad reputation because it chokes out the harmonics of the 3.3 MHz fundamental which needs bandwidth out to 10X that of the fundamental on order to form a nice square wave.

Back in 2002 I bought an 30 MHz Bandwidth Audioquest Optilink 4 Glass Toslink cable which has the same Fused Silica Glass filament that they use in their ATT/ST Glass cables. Over the last 10 years I have compared it to many coax cables up to a price-point of $600 and I have yet to find one that has the absolute transparency that the Audioquest Optilink 4 Fused Silica Glass Toslink has in that it has a 30 MHz bandwidth which is sufficient to allow the full development of the Digital signal's harmonics which is essential for the best sounding digital music playback.

Most recently I used my old Musical Fidelity V-Link to compare my .99999 silver Illuminations D-60 coax to the Optilink 4 Fused Silica Glass Toslink and although the Illuminations D-60 sounds absolutely great on its own when compared to the Fused Silica Glass Toslink it immediately becomes apparent that a lot of low level detail is being completely glossed over by the Illuminations D-60 and a portions of the harmonic structure of the Music is not being fully fleshed out.

Those who have never used or heard a 30 MHz bandwidth Fused Silica Glass Toslink will continue to assume that Toslink is vastly inferior to coax and will never be bothered with what their Music might sound like using a 30 MHz bandwidth Fused Silica Glass Toslink cable like the one John Atkinson used to allow the V-Link add so much additional fidelity to the Benchmark DAC in his Stereophile Magazine Review of the V-Link...

"I then changed to the V-Link, had it feed the Benchmark via a 1m length of AudioQuest Optilink-5 glass TosLink cable, and did not touch the Benchmark's volume control. The violins in the Sibelius were now slightly less steely, the soundstage a tad wider and deeper. More important, the sounds of individual instruments, such as the horns at the start of the first movement, and the timpani and plucked double basses at the start of the second, were slightly more of a piece with the surrounding acoustic."

—q—

Is this possible?

Does each optical toslink have a different working bandwidth?

 

[Doodski]

Quote from an audiophile site:

—q—

“Not all optical Toslink cables are the same and neither do they have the same kind of filament. Common plastic fiber filament Toslink has a bandwidth of about 6 MHz and this is what has given Toslink such a bad reputation because it chokes out the harmonics of the 3.3 MHz fundamental which needs bandwidth out to 10X that of the fundamental on order to form a nice square wave.

Back in 2002 I bought an 30 MHz Bandwidth Audioquest Optilink 4 Glass Toslink cable which has the same Fused Silica Glass filament that they use in their ATT/ST Glass cables. Over the last 10 years I have compared it to many coax cables up to a price-point of $600 and I have yet to find one that has the absolute transparency that the Audioquest Optilink 4 Fused Silica Glass Toslink has in that it has a 30 MHz bandwidth which is sufficient to allow the full development of the Digital signal's harmonics which is essential for the best sounding digital music playback.

Most recently I used my old Musical Fidelity V-Link to compare my .99999 silver Illuminations D-60 coax to the Optilink 4 Fused Silica Glass Toslink and although the Illuminations D-60 sounds absolutely great on its own when compared to the Fused Silica Glass Toslink it immediately becomes apparent that a lot of low level detail is being completely glossed over by the Illuminations D-60 and a portions of the harmonic structure of the Music is not being fully fleshed out.

Those who have never used or heard a 30 MHz bandwidth Fused Silica Glass Toslink will continue to assume that Toslink is vastly inferior to coax and will never be bothered with what their Music might sound like using a 30 MHz bandwidth Fused Silica Glass Toslink cable like the one John Atkinson used to allow the V-Link add so much additional fidelity to the Benchmark DAC in his Stereophile Magazine Review of the V-Link...

"I then changed to the V-Link, had it feed the Benchmark via a 1m length of AudioQuest Optilink-5 glass TosLink cable, and did not touch the Benchmark's volume control. The violins in the Sibelius were now slightly less steely, the soundstage a tad wider and deeper. More important, the sounds of individual instruments, such as the horns at the start of the first movement, and the timpani and plucked double basses at the start of the second, were slightly more of a piece with the surrounding acoustic."

—q—

Is this possible?

Does each optical toslink have a different working bandwidth?

No, it is not possible. The way the digital circuitry responds to square waves and distorted square waves means they either work as a zero or a 1 or they don't. There is no in between where things can sound different. The explanation given here is made of half truths and stretching facts and theory to fit a narrative that is false.

 

[antcollinet]

No, it is not possible. The way the digital circuitry responds to square waves and distorted square waves means they either work as a zero or a 1 or they don't. There is no in between where things can sound different. The explanation given here is made of half truths and stretching facts and theory to fit a narrative that is false.

More correctly - yes they can have different bandwidth - in terms of how accurately they can represent a 3+MHz square wave.

However it makes no difference to the sound as long as the innaccurately transferred waves are still detected correctly at the receiver, as you described.

 

[Bob from Florida]

Mapleshade's Omega Mikro cable line made this SPDIF digital cable made of copper ribbon terminated to their homemade RCA plugs. The ground side of the ribbon was directly connected while the signal ribbon was segmented by small value surface mount capacitors. The idea was the caps would round off the square waves enough that any reflected signals would not sync on the DAC and cause issues with playback. They call it a "jitter reduction network". The more caps the more effective up to the point where the DAC would not sync at all. Now there are issues with this design for sure from an engineering standpoint. A friend of mine used one of these cables for a while between a CD player and a DAC. The cable did work - as in the DAC would output music. It did have this annoying habit of momentary sync loss whenever the air conditioner kicked on....

The 75 ohm coax cable I brought over for my friend to try also worked and did not loose sync when the air came on.

 

[LTig]

More correctly - yes they can have different bandwidth - in terms of how accurately they can represent a 3+MHz square wave.

This is correct for a Coax cable. An optical cable uses light for signal transmission so the band width of the cable certainly is much much higher than 6 MHZ. Whoever claims otherwise has absolutely no clue about physics.

 

[LTig]

Mapleshade's Omega Mikro cable line made this SPDIF digital cable made of copper ribbon terminated to their homemade RCA plugs. The ground side of the ribbon was directly connected while the signal ribbon was segmented by small value surface mount capacitors. The idea was the caps would round off the square waves enough that any reflected signals would not sync on the DAC and cause issues with playback. They call it a "jitter reduction network". The more caps the more effective up to the point where the DAC would not sync at all. Now there are issues with this design for sure from an engineering standpoint. A friend of mine used one of these cables for a while between a CD player and a DAC. The cable did work - as in the DAC would output music. It did have this annoying habit of momentary sync loss whenever the air conditioner kicked on....

View attachment 162376

The 75 ohm coax cable I brought over for my friend to try also worked and did not loose sync when the air came on.

Rounding off square waves in a digital circuit is a recipe for disaster. To prevent reflections both source and sink impedance need to be identical and match the cables impedance. Inserting caps into the transmission path breaks this match. Such a cable is not suited for transmission of digital signals.

 

[antcollinet]

This is correct for a Coax cable. An optical cable uses light for signal transmission so the band width of the cable certainly is much much higher than 6 MHZ. Whoever claims otherwise has absolutely no clue about physics.

Good point - no idea what I was thinking.

 

[Bob from Florida]

Rounding off square waves in a digital circuit is a recipe for disaster. To prevent reflections both source and sink impedance need to be identical and match the cables impedance. Inserting caps into the transmission path breaks this match. Such a cable is not suited for transmission of digital signals.

Not sound engineering but apparently sold some cables due to appropriate use of "jitter reduction" buzzwords.

 

[solderdude]

Does each optical toslink have a different working bandwidth?

Not each but there are differences between POF cable (Plastic Optical Fibre) and bundled glass optical fibers.
Also in POF cables there are differences in damping/km.

POF cable has a bandwidth of 5MHz/km. This means a 100m cable has a Bandwidth of 50MHz and a 10m cable 500MHz.
The problem is not bandwidth but attenuation.
POF cable has an attenuation of about 150dB/km so runs of over 10m means there is already a lot of attenuation.
There are some POF cables that do better.
The more attenuation there is the harder it becomes to get an not to noisy signal out again.
As TOSLINK receivers commonly are used for asynchronic transmission protocols doesn't help either.

Glass bundled fibers have much higher bandwidths and a lot less attenuation/km.

So yes, glass fiber is much better than POF and there are differences in POF cables.
Bandwidth is not the problem though... attenuation is.
The biggest culprit in the FO transmission, however, is the receiver bandwidth and sensitivity as well as speed of its comparator and internal photo diode and pre-amp.

 

[antcollinet]

Not each but there are differences between POF cable (Plastic Optical Fibre) and bundled glass optical fibers.
Also in POF cables there are differences in damping/km.

POF cable has a bandwidth of 5MHz/km. This means a 100m cable has a Bandwidth of 50MHz and a 10m cable 500MHz.
The problem is not bandwidth but attenuation.
POF cable has an attenuation of about 150dB/km so runs of over 10m means there is already a lot of attenuation.
There are some POF cables that do better.
The more attenuation there is the harder it becomes to get an not to noisy signal out again.
As TOSLINK receivers commonly are used for asynchronic transmission protocols doesn't help either.

Glass bundled fibers have much higher bandwidths and a lot less attenuation/km.

So yes, glass fiber is much better than POF and there are differences in POF cables.
Bandwidth is not the problem though... attenuation is.
The biggest culprit in the FO transmission, however, is the receiver bandwidth and sensitivity as well as speed of its comparator and internal photo diode and pre-amp.

Given the other comments upthread about bandwidth what does bandwidth mean in terms of optical cables/fibres?

 

[solderdude]

The maximum frequency of the light pulses that aren't smeared in such a way that one cannot retrieve the original light pulses any more basically.
Of course this is dependent on the construction of the light guide (it is all about breaking indexes), length and used light color (wavelength).

Then you need to have a driver than can switch a LED, VCSEL, FP or other type of laser on and off in a quick enough and a detector that is fast enough (followed by an amplifier with enough bandwidth)

 

[xaviescacs]

This is correct for a Coax cable. An optical cable uses light for signal transmission so the band width of the cable certainly is much much higher than 6 MHZ. Whoever claims otherwise has absolutely no clue about physics.

Which is that physical fundamental difference you are referring to? As I see it, an optical signal and an electrical signal are both electromagnetic waves, so in terms of their capacity of transmitting information using its frequency (energy/power is a different matter), they behave fundamentally in the same way. The difference is given by the medium: a conductor is much more complex and has much more side effects that a "transparent" medium for infrared EM, in which as far as I know the main problem is dispersion of light. Is there something I'm missing?