SPDIF and Coax - is it the same?

[hybridauth_Facebook_1175383980]

I'm looking for a coax digital cable to plug into my NAD M51.

The local store has this one:

https://www.hifiklubben.dk/kabler/digitalkabel/argon-blue-edition-coax/

However it's a little expensive so I'm looking for something cheaper in a pro audio store.

I'm confused though: Is SPDIF the same as coax ?

That is, will any one of these cables work with the coax connector on my NAD M51:

http://www.thomann.de/dk/mutec_optisches_kabel_05m.htm

http://www.thomann.de/dk/cordial_cpds1_cc_digitalkabel.htm

Sorry for asking such a basic question!

Thanks

 

[fr0g]

Yes. SPDIF is basically a format for transmitting digital data over optical or coaxial cable. Sony Philips Digital Interface Format.

The first link you put is an optical cable, the second a coax.

ie. if you want a coax cable the second link will be fine..

(and they all sound identical, so definitely do not go for some expensive one! )

 

[iMark]

The s/pdif signal can be transmitted over a coaxial cable or a glass fibre cable (a.k.a. Toslink).

http://en.wikipedia.org/wiki/S/PDIF

Since the transmission is digital, the cable has absolutely no effect on the quality of the transmission. In other words, buy the cheapest digital coax cable you can get. I've got a digital coax cable that I bought in John Lewis in the sales for 50p and it works fine.

This cable http://www.thomann.de/dk/cordial_cpds1_cc_digitalkabel.htm is the digital coax. It seems a fair price for a digital coax cable. *good*

 

[hybridauth_Facebook_1175383980]

Thank you so much for helping me out both of you

 

[Alec]

you, but don't get to thinking that they will both fit the same port. I once made that mistake.

Egg/face. Still, it was a cheap cable.

 

[clifford]

adding to this, i have a nad c521bee cd player, which had the following (im guessing coaxial?) digital out, http://i39.servimg.com/u/f39/16/49/99/58/20141116.jpg

Is it possible for me to use this to record digitally into my minidisc players line in? or would it need some sort of converter? At the moment i have the standard toslink to miniplug cable eg, http://www.amazon.co.uk/Fisual-Install-Series-Mini-Toslink-Optical/dp/B003NT6RDO/ref=pd_sxp_grid_i_2_1

cheers

 

[clifford]

trying to figure this out it looks like ill need a converter or a cd player that has a toslink output. oh well. ill carry on my analog method

 

[clifford]

I'm guessing I'll need this, or something similar

http://www.maplin.co.uk/p/digital-audio-optical-phono-to-toslink-converter-l73ba

 

[iMark]

I'm a bit confused about what you're trying to achieve. Do you want to copy CDs to Minidisc? If your MD recorder has got digital coax in, all you need is a digital coax lead from the CD player to the MD deck.

If you want to do something else, please explain. I know quite a few things about Minidisc and what you can do with them. It would help if you give us the model number of the minidisc recorder. Is it a portable or a deck?

 

[clifford]

I'm a bit confused about what you're trying to achieve. Do you want to copy CDs to Minidisc? If your MD recorder has got digital coax in, all you need is a digital coax lead from the CD player to the MD deck.

If you want to do something else, please explain. I know quite a few things about Minidisc and what you can do with them. It would help if you give us the model number of the minidisc recorder. Is it a portable or a deck?

yes i want to copy cds to minidisc. i only have a portable minidisc player (MZ-R500) and i usually just copy via the headphone jack (of my old CD player) to the minidisc line in using 3.5mm to 3.5mm cable.

I recently got this CD player (the Nad) that has that digital out and i didnt know whether it simple to make digital recordings using my current hardware?

 

[davedotco]

I'm a bit confused about what you're trying to achieve. Do you want to copy CDs to Minidisc? If your MD recorder has got digital coax in, all you need is a digital coax lead from the CD player to the MD deck.

If you want to do something else, please explain. I know quite a few things about Minidisc and what you can do with them. It would help if you give us the model number of the minidisc recorder. Is it a portable or a deck?

yes i want to copy cds to minidisc. i only have a portable minidisc player (MZ-R500) and i usually just copy via the headphone jack (of my old CD player) to the minidisc line in using 3.5mm to 3.5mm cable.

I recently got this CD player (the Nad) that has that digital out and i didnt know whether it simple to make digital recordings using my current hardware?

You have a bit of a miss-match here, the digital out on the NAD is rca coax, the digital input on the mini disc is optical toslink.

To connect the two you need a coax cable to a coax to toslink converter (your link is to a toslink to coax converter, you need one the 'other' way round), then a toslink cable and a toslink to mini-toslink converter.

All in all a bit of a nonsense, if you are getting decent results with analogue connections I would stick with it.

 

[fr0g]

Just a word of caution even if and when you get a cable that connects the two...

When I owned a Sony Minidisc unit, it would not copy from CD ... This of course was to do with the legality of digitally copying music. I wouldn't be surprised if this block was still in place...

Not to mention that Minidisc is a dead, pointless format these days! You can replace the portable minidisc player with an MP3 player that will hold hundreds of albums, for under 50 euros. It will sound just as good and you won't need to carry around the discs...

 

[clifford]

Ah yes I did read about that being a problem with digital copying. My analogue recordings are good so I'll stick with them. No worries.

I just like minidisc as an object I find it rather beautiful and I'm happy to stick with it for the time being.

Cheers.

 

[SteveR750]

Not to mention that Minidisc is a dead, pointless format these days! You can replace the portable minidisc player with an MP3 player that will hold hundreds of albums, for under 50 euros. It will sound just as good and you won't need to carry around the discs...

And for a bit more, you can get one that can make telephone calls!

 

[iMark]

I have never had any problems copying CDs to MD. In fact I have a machine in the attic that copies them at 4x speed: the Sony MD-DX3. You do get problems though if you want to make a digital copy of the MD. (I also have a Sony MZ-G750 in the attic. Haven't used it in years.)

It's a shame that the NAD CD player only has a digital coax out. I agree with others that investing in a coax to toslink converter is a bit much for an almost dead format. If you're really interested in making the most of MD's I suggest buying a second hand MD deck. A couple of years ago I bought a second hand Sony MDS-JE 520 for EUR 50. (My MD-DX3 unfortunately doesn't have an optical output. I never thought I would need that when I bought the machine in 2000). I have used it to transfer MDs digitally to my Mac. (Most MD decks have optical outputs and my old MacBook has an optical input.) I have nearly finished that project.

Now I'm mainly using the MD deck for digitizing vinyl. Actually, I'm using the ADC (Analogue Digital Converter) in the MD deck, which is connected to the old MacBook. I then take the hard drive with recorded vinyl to my Mac Mini and clean up the sound with Vinyl Studio. Another interesting use for an old MD deck is to use it as DAC. Most of them have a coax and an optical input, so you could plug in any transport (CDP, DVD or BDP) and a TV or STB.

I really liked and still like the Minidisc technology. But I find ripping CDs into the iTunes library and then tranferring music and podcast to my iPod Nano much more convenient than a portable MD player. I would argue that my Sony MZ-G750 sounds a little bit better than my iPod but convenience rules when I'm travelling. I can put 16 GB of music and podcasts on my Nano. That would equal a case of over 30 MDs. And it has FM radio, just like my Sony MZ-G750. Just get better earphones than anything Apple supplies.

For details about inputs and outputs on Minidisc machines I can really recommend this website: http://www.minidisc.org/equipment_browser.html

 

[andyjm]

Since the transmission is digital, the cable has absolutely no effect on the quality of the transmission. In other words, buy the cheapest digital coax cable you can get. I've got a digital coax cable that I bought in John Lewis in the sales for 50p and it works fine.

As anyone who has read my ramblings would know, I am not a 'cable believer', but I am afraid the above is technically incorrect.

Posters assume S/PDIF is purely a serial data link. It is not. It carries a robust stream of serial data, and a very fragile sample clock superimposed on the data (if you care, google 'Manchester biphase mark encoding'). If your DAC is sensitive to jitter on the S/PDIF clock, then the type of cable can matter.

Toslink is generally poorly implemented and has a (relatively) low bandwidth in comparison to coax. Limited bandwidth S/PDIF can introduce 'code correlated jitter', which is the most easily detected type of jitter - much more objectionable than random jitter.

So, if your DAC slaves itself to the S/PDIF clock and is sensitive to jitter ( to be clear - most modern DACs aren't), then all other things being equal, you would probably do better with decent coax.

 

[kmlav]

I would second the go for a Coax vote's.

 

[fr0g]

Vote for optical here

Most DACs these days are pretty immune to jitter, so ignoring that, which I do, there is the issue of electrical noise...there is no electrical connection with optical so that is that sorted

 

[SteveR750]

Since the transmission is digital, the cable has absolutely no effect on the quality of the transmission. In other words, buy the cheapest digital coax cable you can get. I've got a digital coax cable that I bought in John Lewis in the sales for 50p and it works fine.

As anyone who has read my ramblings would know, I am not a 'cable believer', but I am afraid the above is technically incorrect.

Posters assume S/PDIF is purely a serial data link. It is not. It carries a robust stream of serial data, and a very fragile sample clock superimposed on the data (if you care, google 'Manchester biphase mark encoding'). If your DAC is sensitive to jitter on the S/PDIF clock, then the type of cable can matter.

Toslink is generally poorly implemented and has a (relatively) low bandwidth in comparison to coax. Limited bandwidth S/PDIF can introduce 'code correlated jitter', which is the most easily detected type of jitter - much more objectionable than random jitter.

So, if your DAC slaves itself to the S/PDIF clock and is sensitive to jitter ( to be clear - most modern DACs aren't), then all other things being equal, you would probably do better with decent coax.

Andy, is there any difference in the way the data is streamed to the DAC, i.e. async, adaptive? I'm guessing that a DAC that send timing data to the device will therefore be somewhat dependent on cable quality, whereas adaptive / data reclocked in the DAC is far less so?

One thing I don't understand, is why DACs don't simply buffer a chunk of data, and stream from that source? At the moment, I have set J River to play from the memory, surely a DAC could include at little extra cost a couple of Gbs worth of SSD from which to stream each file? This is effectively what is happening in the PC -> internal soundcard?

 

[SteveR750]

Since the transmission is digital, the cable has absolutely no effect on the quality of the transmission. In other words, buy the cheapest digital coax cable you can get. I've got a digital coax cable that I bought in John Lewis in the sales for 50p and it works fine.

As anyone who has read my ramblings would know, I am not a 'cable believer', but I am afraid the above is technically incorrect.

Posters assume S/PDIF is purely a serial data link. It is not. It carries a robust stream of serial data, and a very fragile sample clock superimposed on the data (if you care, google 'Manchester biphase mark encoding'). If your DAC is sensitive to jitter on the S/PDIF clock, then the type of cable can matter.

Toslink is generally poorly implemented and has a (relatively) low bandwidth in comparison to coax. Limited bandwidth S/PDIF can introduce 'code correlated jitter', which is the most easily detected type of jitter - much more objectionable than random jitter.

So, if your DAC slaves itself to the S/PDIF clock and is sensitive to jitter ( to be clear - most modern DACs aren't), then all other things being equal, you would probably do better with decent coax.

Andy, is there any difference in the way the data is streamed to the DAC, i.e. async, adaptive? I'm guessing that a DAC that send timing data to the device will therefore be somewhat dependent on cable quality, whereas adaptive / data reclocked in the DAC is far less so?

One thing I don't understand, is why DACs don't simply buffer a chunk of data, and stream from that source? At the moment, I have set J River to play from the memory, surely a DAC could include at little extra cost a couple of Gbs worth of SSD from which to stream each file? This is effectively what is happening in the PC -> internal soundcard?

 

[iMark]

To be perfectly honest, I think you're perpetuationg myths that stem from the continuous stream of audiophool drivel. It may not be totally scientific, but if you hear an untinterrupted stream of sound (without any clicks and pops) your cabling is OK. It doesn't matter whether your cables are digital coax or optical.

The bandwidth of S/PDIF is more than enough for it not to introduce jitter.

 

[unsleepable]

As anyone who has read my ramblings would know, I am not a 'cable believer', but I am afraid the above is technically incorrect.

Posters assume S/PDIF is purely a serial data link. It is not. It carries a robust stream of serial data, and a very fragile sample clock superimposed on the data (if you care, google 'Manchester biphase mark encoding'). If your DAC is sensitive to jitter on the S/PDIF clock, then the type of cable can matter.

Ok. But just to clarify, the reliance on the source clock applies to both TOSLINK and coaxial—unless as you said, a re-clocking mechanism is in place.

Toslink is generally poorly implemented and has a (relatively) low bandwidth in comparison to coax. Limited bandwidth S/PDIF can introduce 'code correlated jitter', which is the most easily detected type of jitter - much more objectionable than random jitter.

Poor implementation is not really an argument against the technology itself. Asynchronous USB is also often poorly implemented in modern DACs, and this is no reason to recommend DACs with synchronous USB ports.

Besides, this whole thing is poor TOSLINK implementation is a bit old in my opinion. It would be good to have specific examples of current devices with badly implemented TOSLINK ports in comparison with digital coaxial ones.

As for bandwidth, both TOSLINK and coaxial can provide more of it than required by S/PDIF. And it's unused. So I don't understand your proposition here. I would understand that the average error when sampling at, let's say, 100MHz is smaller than when sampling at 10MHz—but this is not how S/PDIF works. You could argue that some DACs accept different resolutions on TOSLINK and coaxial ports, but then again this is a matter of implementation rather than technology, and I'm not even sure it's the most common case.

So, if your DAC slaves itself to the S/PDIF clock and is sensitive to jitter ( to be clear - most modern DACs aren't), then all other things being equal, you would probably do better with decent coax.

Digital coaxial might suffer from two other issues about which TOSLINK is immune. Unsteady voltage or ripple can be a source of jitter in itself, and the connection can also transmit noise independent of the signal.

 

[andyjm]

A bit technical, but here goes.

The bandwidth requirement of a S/PDIF link isn't set by the speed of data transfer, but by the need to maintain sharp edges at the bit transition points. This implies a bandwidth many times higher than the data rate.

A wrinkle of the encoding methodology is that '1s' are sent at effectively twice the frequency of '0s'. Depending on the type of discriminator circuit used to establish when a '0' to '1' transistion has taken place, if the bandwidth of the link is limited and the pulses have rounded edges, '0's can be detected at slightly different points than '1s'. This means that '1111' will tend to slow the PLL clock down slightly, '0000' will speed it up slightly. The result is that phase jitter is introduced into the recovered clock, correlated with the data on the link. The worse the bandwith of the link, the more rounded the pulses, the worse the effect.

There is an AES paper by Dunn and Hawksford "Is the AESEBU / SPDIF digital interface flawed" which goes into this problem in some depth.

This problem is avoided if jitter mitigation techniques are used. Resampling, ring buffer storage, or the best of all, a flow control protocol that allows the clock to be situated next to the A2D chip in the DAC (where it belongs) not in some other box entirely. Async USB and wordclock outputs are techniques that allow flow control to be used.

 

[unsleepable]

A bit technical, but here goes.

The bandwidth requirement of a S/PDIF link isn't set by the speed of data transfer, but by the need to maintain sharp edges at the bit transition points. This implies a bandwidth many times higher than the data rate.

That as you go down the protocol layer you need more bandwidth is something that happens with pretty much any protocol. S/PDIF encodes the signal at OSI layer 2 using BMC (biphase mark code), which is a quite widespread modulation system. The required bandwidth is double of the data transmitted plus some additional overhead—so it's not "many times", but actually of the same order, and it does depend on the bandwidth of the transmitted data.

A wrinkle of the encoding methodology is that '1s' are sent at effectively twice the frequency of '0s'. Depending on the type of discriminator circuit used to establish when a '0' to '1' transistion has taken place, if the bandwidth of the link is limited and the pulses have rounded edges, '0's can be detected at slightly different points than '1s'. This means that '1111' will tend to slow the PLL clock down slightly, '0000' will speed it up slightly. The result is that phase jitter is introduced into the recovered clock, correlated with the data on the link. The worse the bandwith of the link, the more rounded the pulses, the worse the effect.

In BMC, each bit is transmitted over two clock cycles—whether a 0 or a 1 is sent. Then there is always a change of signal at the end of each bit. This change marks when the data must be interpreted, and also sets the clock.

Therefore, the transmission rate is constant. And there is no confusion as to when a bit is read—when the signal changes beyond the sampling distance.

There is an AES paper by Dunn and Hawksford "Is the AESEBU / SPDIF digital interface flawed" which goes into this problem in some depth.

This problem is avoided if jitter mitigation techniques are used. Resampling, ring buffer storage, or the best of all, a flow control protocol that allows the clock to be situated next to the A2D chip in the DAC (where it belongs) not in some other box entirely. Async USB and wordclock outputs are techniques that allow flow control to be used.

This paper is from 1992—more than twenty years ago. And it seems to analyse data transmission over "bandwidth-limited links" in the context of the technology that was available at the time, and issues arising from non-steady changes in voltage or signal intensity. Seriously, how is this relevant now?

 

[andyjm]

One of the interesting things as you get old, is how knowledge gained by older generations is somehow considered of less value than the current crop of researchers. So, running through your points (in a slightly different order):

Just because the paper was written in '92 does not make it any less valid. Ampere, Volta and Ohm were all born in the 1700s, their analysis is still going strong.

You are confusing data rate with the bandwidth required to preserve a decent pulse shape. Instead of thinking of the clock as a digital signal, think of it as an analogue waveform whose risetime needs to be protected. To keep a passable edge, the link needs to be able to carry at least the first 3 or 4 odd harmonics.

In biphase mark encoding, a '0' has a transition at the start of a bit period, a '1' has an additional transition in the middle of a bit period. A repeated '1111111' will be a square wave with twice the frequency of '000000'.

You are confusing bit errors in the data with derived clock phase errors. It is not a question of when the bit is sampled, it is a question of when the flywheel in the phase locked loop gets its kick from the clock transition detection circuitry - this point differs for '1's and '0's, hence the introduction of phase jitter.

By definition, every link is band limited, and this problem is still real today. The standard toslink chipset has relatively poor bandwdith, and does introduce phase jitter. Whether it is audible or not depends on the magnitude and spectrum of the jitter, and what the DAC does with it.