Shunyata AES/EBU & S/PDIF Required Lenghts

[Parabellum]

I am not too worried going with shorter lenghts of cable with USB (even desirable AFAIK) but I am wondering what are the minimum required lenghts for Shunyata AES/EBU & S/PDIF digital cables. From what I know, 1.5M is often cited as the minimum for best performace but I see some Alpha and Sigma in shorter 1M lenghts. Is this detrimental to performance on these cables?

Also, are AES/EBU & S/PDIF cables Hi-Res 24/192 capables?
In my case, I may not actually need USB since I do not plan to play DSD files. Only ripped CDs and Qobuz.
I read some reviews where AES/EBU & S/PDIF is claimed sound more natural than USB generally speaking.

[SCAudiophile]

FWIW, I read many articles and whitepapers years ago that claimed the same minimum length. They were thorough and made sense at the time so my practice is to keep 50, 75 and 110ohm digital cables of no less than 1.5m and I have also had 2m and 3m lengths.

The thinking behind these papers all centered around cables of at least 1.5m would minimize signal reflections back to the transmitter side which would muck up the integrity of the digital signal and results in increased jitter.

Too long a digital cable can result in it becoming a transmission line.

Transmission-line effect comes into play when signal cable lengths are so long that the signal delay down the cable length and back becomes longer than the waveform's rise or fall time.

I don't know about USB but given it is about the worst cable spec ever IMHO, was intended only for computer to printer quick connects originally and was never meant for audio, I don't use it for audio.

I'm sure Caelin can give any more current scientific truth behind any true minimum length of a SPDIF, AES or other digital cable as well as max lengths.

[tdelahanty]

Originally Posted by SCAudiophile

FWIW, I read many articles and whitepapers years ago that claimed the same minimum length. They were thorough and made sense at the time so my practice is to keep 50, 75 and 110ohm digital cables of no less than 1.5m and I have also had 2m and 3m lengths.

The thinking behind these papers all centered around cables of at least 1.5m would minimize signal reflections back to the transmitter side which would muck up the integrity of the digital signal and results in increased jitter.

Too long a digital cable can result in it becoming a transmission line. Transmission-line effect come into play when signal cable lengths are so long that the signal delay down the cable length and back becomes longer than the waveform's rise or fall time.

I don't know about USB but given it is about the worst cable spec ever IMHO, was intended only for computer to printer quick connects originally and was never meant for audio, I don't use it for audio.

I'm sure Caelin can give any more current scientific truth behind any true minimum length of a SPDIF, AES or other digital cable as well as max lengths.

I was very pleased by your remarks about reflections, delay and transmission line theory. It's possible many people don't realize these digital signals are transmitted at very high frequencies, up in the megahertz ranges, and these theories make perfect sense.

[SCAudiophile]

Originally Posted by tdelahanty

I was very pleased by your remarks about reflections, delay and transmission line theory. It's possible many people don't realize these digital signals are transmitted at very high frequencies, up in the megahertz ranges, and these theories make perfect sense. [emoji106]

Thank you!

[Parabellum]

Originally Posted by SCAudiophile

FWIW, I read many articles and whitepapers years ago that claimed the same minimum length. They were thorough and made sense at the time so my practice is to keep 50, 75 and 110ohm digital cables of no less than 1.5m and I have also had 2m and 3m lengths.

The thinking behind these papers all centered around cables of at least 1.5m would minimize signal reflections back to the transmitter side which would muck up the integrity of the digital signal and results in increased jitter.

Too long a digital cable can result in it becoming a transmission line. Transmission-line effect come into play when signal cable lengths are so long that the signal delay down the cable length and back becomes longer than the waveform's rise or fall time.

I don't know about USB but given it is about the worst cable spec ever IMHO, was intended only for computer to printer quick connects originally and was never meant for audio, I don't use it for audio.

I'm sure Caelin can give any more current scientific truth behind any true minimum length of a SPDIF, AES or other digital cable as well as max lengths.

Shunyata offers 1M lenghts on their website along the pricing, so if no minimum required lenght of 1.5M is specified, then I believe I can thrust they made significant testing in offering 1M lenghts. I would like to have their opinion on this however.

Regarding USB, you were right, this was developped I believe mainly for those who used their compters (laptops) as a streaming device. In my case, I have a Bryston BDP-2. It is basically a computer with a motherboard, CPU, RAM and USB ports as well. But, unlike a computer, there is also AES/EBU and S/PDIF outputs too. Based on an article over HiFi Advice, every time the BDP when connected with AES/EBU sounds better than USB. I was all set for a new USB cable but now I am thinkering a bit more that AES/EBU might be better in my case. Especially so that I don't care about DSD.

[PHC1]

I don't use USB for the bulk of my listening on the main system. It is streamed to a Roon end point streamer and then gets to the DAC over the SPDIF.

The USB however IS used when I pump DSD high resolution files that I have purchased from the Apple MacBook Pro. Now, these recordings are truly high resolution, recorded that way with minimal processing.

Holy Smokes! I have never experienced sound like that with CD/SACD/Vinyl.

If there is a deficiency in USB, let's just say, let me have more of it!

[SCAudiophile]

Originally Posted by Parabellum

Shunyata offers 1M lenghts on their website along the pricing, so if no minimum required lenght of 1.5M is specified, then I believe I can thrust they made significant testing in offering 1M lenghts. I would like to have their opinion on this however.

Regarding USB, you were right, this was developped I believe mainly for those who used their compters (laptops) as a streaming device. In my case, I have a Bryston BDP-2. It is basically a computer with a motherboard, CPU, RAM and USB ports as well. But, unlike a computer, there is also AES/EBU and S/PDIF outputs too. Based on an article over HiFi Advice, every time the BDP when connected with AES/EBU sounds better than USB. I was all set for a new USB cable but now I am thinkering a bit more that AES/EBU might be better in my case. Especially so that I don't care about DSD.

Yes, definitely. The reflection issue is reported to not occur when the transmitter, receiver and cable are all exactly matched and working exactly at the stated 75ohm or 50ohm spec.

As for USB, luckily audio and cable manufacturers have gone to great pains to overcome the shortcomings in the baselime spec to make things much for tenable.

[crion2]

Aurender put an OCXO circuit near their AES/EBU and S/PDIF outputs. So one actally need to check what the streamer/DAC manufacturers put most thought into.

I had 1.5m minimum on most cables it was also more practical. I specified a 1M clock cable just because it is only connected to one unit. Not sure reflection on a v2 cable would cause a mandate 1.5m.

[tdelahanty]

Hope this explains reflections. Lets pretend your 1m digital cable is a one meter long piece of railroad track and there is a box car sitting on the track. As the car travels the length of track it will pass thru null points ( zero reflections). The trick is to choose cable length to equal a null point at the chosen frequency of operation. I did this experiment in lab class (about a thousand years ago) and it is easily demonstrated.

[SCAudiophile]

Originally Posted by tdelahanty

Hope this explains reflections. Lets pretend your 1m digital cable is a one meter long piece of railroad track and there is a box car sitting on the track. As the car travels the length of track it will pass thru null points ( zero reflections). The trick is to choose cable length to equal a null point at the chosen frequency of operation. I did this experiment in lab class (about a thousand years ago) and it is easily demonstrated.

Good post!