• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Toslink switch and TosLink/Coaxial SPDIF converter - Review and measurements

JamesYeomans

Member
Forum Donor
Joined
Apr 28, 2020
Messages
79
Likes
83
Location
UK

DrCWO

Active Member
Audio Company
Forum Donor
Joined
Nov 14, 2020
Messages
247
Likes
356
the most effective way to observe digital is to do it in digital...
I'll try to clarify by example:
I have several digital acquisition cards and no need to use a DAC ouput etc
some from the professional world like the rme are heavily processed in input and offer, like the rme, practically constant results whatever the quality of the incoming signal...
but I also have a general public card that is very poorly processed in digital input..
it allows observations, comparisons jitter mesurment that are truly differentiated according to the quality of the sources..

conclusion for this type of observations of digital sources only the modest one is useful and allows us to sort things out and avoid major misunderstandings.... .
;-)
I wonder why today quality degradation in digital signals is still in discussion :facepalm:
Sure this was an issue with elder DACs that had weak PLLs to recover the clock. These times are over. Get a State of The Art DAC and forget about digital signal quality and also about WooDoo cables, signal enhancers and other stuff associated with this theme.
In any case, I only think about the quality of digital audio signals when I hear dropouts.
 
OP
R

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,657
Likes
4,193
Location
Liège, Belgium
OP
R

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,657
Likes
4,193
Location
Liège, Belgium
There is no delay, because the data is not processed. Or more precise: the delay is below one sample aka microseconds. Lip-sync is several milliseconds.
So I'll try and check.
If there is no delay, that will confirm what you say.
(I have no doubt, be assured. I just like when we can see the confirmation)
 

JamesYeomans

Member
Forum Donor
Joined
Apr 28, 2020
Messages
79
Likes
83
Location
UK
I think I'll try to get one, to compare.
(It's quite expensive, though)

Unfortunately, I don't have the equipment to assess such thing as resistance to Jitter in digital domain, like Matthias has done.

But we may still see (measure) a difference.
Here's the direct link to the RME recommended SPDIF switch, in case anyone missed it in that forum thread: https://www.tindie.com/products/Beni_Skate/automatic-spdif-opticalrca-audio-switch/
I have 3 of them and they all function well. Never any drop-outs etc. I never use 192kHz sample rate though (don't see the point), so haven't tested that aspect.
I am not affiliated with the maker/seller.
 

morillon

Major Contributor
Joined
Apr 19, 2022
Messages
1,270
Likes
250
I wonder why today quality degradation in digital signals is still in discussion :facepalm:
Sure this was an issue with elder DACs that had weak PLLs to recover the clock. These times are over. Get a State of The Art DAC and forget about digital signal quality and also about WooDoo cables, signal enhancers and other stuff associated with this theme.
In any case, I only think about the quality of digital audio signals when I hear dropouts.
that's not the point..
- we can decide to stop the observations of this kind of signals..
- we say "there's that, there's only to buy recent dacs" (we can also estimate that not useful if we don't want to have decent resolutions) but above all...
but
if the idea is to observe what is happening at the first level... then yes, you have to take certain precautions... otherwise it makes no sense...
and that's just the point ...
after you can indeed explain to rja4000 that his review was useless ... it's your point which in the end is not addressed to me ...
My point is just how to make an interesting observation.....of its precise signals...
 
Last edited:
  • Like
Reactions: MCH

EJ3

Major Contributor
Forum Donor
Joined
Dec 10, 2019
Messages
2,082
Likes
1,601
Location
James Island, SC
I'm sorry..
but experiments using the rme's digital inputs (except perhaps in extreme cases) are of no interest...
their powerful "steadyclock fs "input processing levels absolutely everything... so it is normal that you observe...nothing...
You need practically unprocessed inputs to be able to make interesting observations.
I have a lot of respect for your efforts...but there they were in vain
;-)
(This is a situation that sometimes comes across on ASR, typically also in observation in the digital domain...and which must be clearly explained to avoid misunderstandings, or irrelevant measurements-observations..I had even one day gently asked “gentleman rme” if it would be possible to disengage him for this type of observation..
but it was not well understood I think)
BUT it is good to know that nothing happens to that throughput so now I can say that I want/need an RME.

Just because the measurements are of no use to you does not mean that they are not useful to others.
 

morillon

Major Contributor
Joined
Apr 19, 2022
Messages
1,270
Likes
250
BUT it is good to know that nothing happens to that throughput so now I can say that I want/need an RME.

Just because the measurements are of no use to you does not mean that they are not useful to others.
:oops:
is not a review of rme and the steadyclock impact :oops:
(it's especially drcwo which suggests that all this is of no interest.... me just an exchange on the method)
 
Last edited:

DrCWO

Active Member
Audio Company
Forum Donor
Joined
Nov 14, 2020
Messages
247
Likes
356
that's not the point..
- we can decide to stop the observations of this kind of signals..
- we say "there's that, there's only to buy recent dacs" (we can also estimate that not useful if we don't want to have decent resolutions) but above all...
but
if the idea is to observe what is happening at the first level... then yes, you have to take certain precautions... otherwise it makes no sense...
and that's just the point ...
after you can indeed explain to rja4000 that his review was useless ... it's your point which in the end is not addressed to me ...
My point is just how to make an interesting observation.....of its precise signals...
I don't want at all call @Rja4000's work useless.
It shows very nicely that detecting edges and then amplifying and sending out the signal causes phase noise. It also shows that there are DACs like the ADI-2/4 that don't care. This is how it should be with every modern DAC. But then there are (especially older) DACs that react sensitively to this. This is also what his research shows. Great :)
I just wanted to say that this is mostly irrelevant with modern DACs so for me a “Don’t care anymore”.
Maybe I was a bit harsh and apologize for that. But I have followed the entire DAC development as an engineer since the 90's and until the development of digital double PLLs (e.g. Apogee Rostera 200) this was a more or less severe problem in every DAC. Hence also the audible differences between digital cables.
I am just happy that this is solved now
@Rja4000 thank’s for posting this
 

gumblag

Member
Joined
May 17, 2022
Messages
26
Likes
34
Location
The Netherlands
The same plug, isn't it? Any suggestion for a correct name in video applications? +
I'm Dutch. In the Netherlands they call this kind of connector, apart of the kind of signal which goes through it, also a 'tulp connector' literally translated into English "tulip connector'. A tulip is a flower as you might know, something one can see so I would suggest 'tulip connector' for video signals.
 
Last edited:
OP
R

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,657
Likes
4,193
Location
Liège, Belgium
There is no delay, because the data is not processed. Or more precise: the delay is below one sample aka microseconds.

I confirm: There is NO delay at all.

I've set the 2 plots to different amplitude, otherwise they are 100% superposed.
Red is through the TosLink switch, Black is straight.

One square is 10µs. Sampling frequency is 96kHz. SIgnal is at 182.5Hz.

2023-09-11 12_59_38-Multi-Instrument Pro 3.9.8.1   -   [+3DP+DLG+LCR+UDP+VBM+DHS]   -   _Dante...png


Zoomed out..

2023-09-11 13_04_35-Multi-Instrument Pro 3.9.8.1   -   [+3DP+DLG+LCR+UDP+VBM+DHS]   -   _Dante...png
 
Last edited:

eliash

Senior Member
Joined
May 29, 2019
Messages
407
Likes
209
Location
Bavaria, near lake Ammersee
...always wanted to know how my ADI-2 fs performs in terms of jitter. If it does like this it´s stunning...hats off...and Thanks for the measurements (even with a different perspective on them)!
 

MCH

Major Contributor
Joined
Apr 10, 2021
Messages
2,507
Likes
2,142
I ordered one.
Stay tuned ;-)
At 5 meters, i would be curious to know what will you be actually testing: the cable, the transmitter or the receiver. All toslink cables attenuate the same? I would be surprised if yes.
 
OP
R

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,657
Likes
4,193
Location
Liège, Belgium
At 5 meters, i would be curious to know what will you be actually testing: the cable, the transmitter or the receiver. All toslink cables attenuate the same? I would be surprised if yes.
That's part of the issue.
Here, those devices seem to be mere repeaters.
They detect a signal, don't care about its format, and repeat it.
The longer (or badder) the cable, the worst the result will be.
I wonder
1. If repeating the signal will help for longer distance
2. If there is a point they won't work anymore, anyway
3. If the RME will be able to work whatever the cable length or if, as it is likely, it will just stop working at a given attenuation (where no DAC would work anyway)

There are a lot of unknown here.
 
  • Like
Reactions: MCH

MC_RME

Addicted to Fun and Learning
Technical Expert
Audio Company
Joined
May 15, 2019
Messages
843
Likes
3,533
All TOSLINK POF (plastic optical fibre) indeed have a similar damping on the optical signal. And the longer the more damping, but 5 meter is still in normal condition and expected to work. Other materials like glass fibre, which is not the standard TOSLINK connection, have nearly no damping and can be used at much higher lengths, but can cause other problems due to internal different reflection and overdriving the receiver chip.

The issue with 192 kHz is typically the receiver. The transmitter, even if not specified for that data rate, often works ok.
 
OP
R

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,657
Likes
4,193
Location
Liège, Belgium
All TOSLINK POF (plastic optical fibre) indeed have a similar damping on the optical signal. And the longer the more damping, but 5 meter is still in normal condition and expected to work. Other materials like glass fibre, which is not the standard TOSLINK connection, have nearly no damping and can be used at much higher lengths, but can cause other problems due to internal different reflection and overdriving the receiver chip.

The issue with 192 kHz is typically the receiver. The transmitter, even if not specified for that data rate, often works ok.
I ordered 5m and 10m.
We'll see what this will teach us ...
 
Top Bottom