On my previous DAC (Pro-Ject Pre Box S2 Digital) the toslink cable mattered quite a lot, the no-name toslink cable I used first sounded much worse than the Supra I bought later, but I realize this depends on the DAC so thats why I ask for the RME DAC.

That is oversimplifying it a bit, yes its digital but without any error correction or error detection, and toslink is also easily affected by jitter:

Poor toslink cables (and/or connectors) degrade the square signal which causes more or less jitter.

Buffering and reclocking the signal certainly helps, and a more accurate clock also helps, but I am unsure if it fully removes all problems. A poor cable can degrade the signal in such a way that it's not possible to reconstruct it. A simple example is how some toslink cables don't work in 192 khz while others do work. It's also not possible to correctly reconstruct a signal where the jitter is bigger than half the width of a clock-pulse (in that case, the buffering and reclocking will put the "bit" in the wrong "slot")

From a sound quality perspective, jitter degrades the sound quality (smears background, loses details, etc), which is why some toslink cables sound better on some DACs.

Which is basically what I am asking in this thread. Can the RME ADI-2 DAC compensate fully for poor cables and degraded signal? If it can't (and most DACs can't), then a higher quality toslink cable will sound better.

If the jitter is so high that the receiver can not read out the data correctly, then the transmitting unit would be no longer working at all - because its clock must be defective. You got all the nice information - why not calculate what jitter amount such a 'wrong bit detected' condition equals? Tip: more than 100 ns...

The example with 192 kHz is not accurate as it is not about jitter, but (usually) transmitters that are not specified to work at such a data rate.

Also, as stated over and over, such a situation would cause clicks. Not a whatever audiophile changed sound stage, depth, veil...

And finally, the ADI would display such a situation with LOCK instead of SYNC.

I have a cable right here, and if I switch to it 192 khz don't work (same transmitter, same DAC). That cable works in 96khz and lower though. And from my understanding, SPDIF is a one-way communication, so no handshaking or "acks" send back from the DAC, which would mean that the cable degrades the signal to much for the receiver to make sense of it.

But that is a little beside the point, my question is if toslink cables sound different, provided they do work with the sample rate? It would actually be pretty amazing if they did, since the DAC would then need to be able to "undo" all degradation that occurred in the cable.

As often with audio, the only way to find out is to try :-)

I ordered one of these, from what I read its a very good cable but without the high-end price (although its not exactly cheap either): http://www.lifatec.com/toslink2.html

That's what I meant (and a bit ridiculous as well). 192 kHz transmitter needs 25 mb/s. 96 kHz 12 mb/s. Yours (if correct) would only work up to 48 kHz.

Normal plastic TOSLINK cable supports 192 kHz up to 15m, even if there is a TOSLINK switcher in between.

Before buying glass fiber cable for TOSLINK I would check with RME whether this has no side effects.
Because glass has not the same dampening like plastic, not that you overload your receiving TOSLINK.

It could even have other advantages to use plastic TOSLINK cables, I could think of that they are more robust and have a better bending radius compared to glass fiber.

The audio material will at the end determine the sample rate. Upsampling won't give you more quality.
All this has been discussed already on this forum.

In the studio area people have technical background and know how our ears and brain work.

In the HiFi area you have usually to face these kinds of problems:
1. people who incorrectly transfer their experiences and procedures from analog technology to digital technology without really knowing how digital technology works (-> audiophile USB cables and alike ...)
2. are very receptive to industry advertising promises
3. The brain can only compare sound differences by fast A/B comparisons (in the range of 2 seconds).
4. Incorrect test methodology because the knowledge about psychoacoustic effects is missing. Wrong subjective impressions only lead to rumours spreading via forums until it becomes a pseudo-truth, which is actually not true at all.
5. Then HiFi is sometimes also a market of vanities. A cable that costs several hundred or thousand euros usually looks more stylish.

The data sheet of the TOTX178 says 6 mb/s and that is Megabit, not Megabyte. Higher bandwidth in this case does not mean less jitter but being able to operate at those data rates at all.

The ADI-2 DAC has a 25 Megabit/s (Mbps) receiver and doesn't need glas fiber at all (see Ramses comment). We did tests with 10 m standard POF from ADI-2 Pro into a simple switch/repeater (no re-coding of the signal) into another 10 meters to the DAC at 192 kHz. Works flawlessly. So there is no need for uber-expensive solutions.

Thanks Curt.

If you did not understand my points, then it's clear why you have problems to understand our good meant proposals.

It also tells how good the lobby and advertisement work of the HiFi industry is. They only need to use the adjective "audiophile" in the advertisement (see the cable that you finally choosed) and when you read this you are automatically biased. The Mutec TOSLINK cable which I proposed is also perfectly fine, but it seems not fine enough for you, because it's simply missing some buzzwords.

And you really want to tell me here that you would be able to make a list for the opposite ?
If it's based on all this overhyped advertizements then I can only hope for you that you do not believe that yourself.

Magnus, please do not take it personal in any way. You are free to ask whatever you want.

But I am wondering. Why do you ask then in a vendor forum with professional people if you do not trust their expertise ?
Then it would be better to ask in one of the typical HiFi or analog forums where your expectations are better satisfied.
This would avoid a lot of unnecessary discussions that simply waste time for no reason.

Another thing, if the standards in the studio sector which produces the music that you hear would not be good enough, then you would have a much bigger problem on the consumer side than a "cable problem". This would mean that you whole HiFi components would be simply oversized (if studio would be that bad).

Here you go!!  The "Clever Clock"    :)

For Years, Cables have been a Diversionary Tactic.  A Red Herring of Sorts.

Finally!!  The Secrets are Revealed! 

http://www.machinadynamica.com/machina41.htm

"Signature Edition" no less!    Who's Signature though I wonder...P.T. Barnum?

Thanks for this nice discovery!! I think I will make some purchases to enlarge my... soundstage.

Amen... You said everything.
I'm really glad digital domain and internet time allow us to think and to stop believing all the BS they sold to us in the 90's audiofool's magazines...
Those snake oil digital cables make me sick... Audiophile USB or Ethernet cables, golden plated TOSlink cables, people really don't think...

BTW I love this dialogue between a member of the Audioholics forum and the boss of Audioquest snake oil cables

Digital cables matter, but of course they are still bit-perfect. There are two things that matter beside being bit-perfect:
1. Electronic noise that traverse the cable and affect the analog stage of the DAC. This can easily be heard (and measured) by connecting a USB DAC directly to a normal computer with fans and stuff, compared to for example USB connected to a well engineered ethernet transport.
2. Jitter, which despite being bit-perfect (in a sense) causes the bits to align to wrong clock-slot. All cables degrade the signal (digital cables are analog, 1 are represented by +V or a light pulse), and the more a cable degrade the signal the more risk of jitter (or in extreme cases bit-loss).

USB cables are affected by both, ethernet by 1 (and in a roundabout way 2), toslink by 2, coax by both. How much a DAC is affected by this depends on the DAC, 1 can be partially solved with USB being galvanic isolated, 2 can be partially solved by rebuffering and reclocking internally in the DAC, or by using the DACs clock for the sender like in asynchronous USB.

This is why I personally prefer toslink connected to the DAC, since at the very least it completely solves 1, and 2 is the reason I started this thread. In my system, USB sounds worse than toslink (and I can toggle between them on preamp so easy to compare and do blind tests).

Surely you so called "studio professional" knows about this :-)

I'm waiting for the day when some phony comes up with a can of audiophile air for the ultimate wifi transmission delivering unprecedented clarity and detail with a soundstage so deep and wide no cable can ever convey.

Way to go, you might learn something new if you keep reading and that would be horrible.

And btw, bit-perfect and 100% free of jitter is not the same thing. Bit-perfect means all bits arrives as they should and in proper order, but not that they arrived in correct time.

Take toslink for example: it depends on the senders clock, no easy way to influence that (toslink is a one-way street), so if the sender is a little to slow or a little to fast than you get either buffer underrun or buffer overrun, unless the DAC adjusts its internal clock in which case you will get at the very least some jitter. Smaller variation in the senders clock can be handled by the DACs internal clock and a buffer, which I think is what the SteadyClock FS does, but it will never remove 100% jitter.

Interesting read, and the quoted sentence certainly indicates that as long as a toslink cable works it will sound the same. I also saw a measurement on RME ADI-2 DAC that displayed jitter for USB and toslink, and toslink had less (not sure I believe that though, more likely some electronic noise on USB caused it to look that way).

That would mean that toslink is without doubt the best digital connection for this DAC due to galvanic isolation, unless you need higher sample rates that only USB can provide.

It will be fun to compare my current cheap toslink to the Lifatec one I ordered, see if theory and real-life matches (which is not always the case).

MagnusH,

Man. I feel Bad for you.   Audio, and all of it's Accessories, and Tweaks are Terrific Fun...but only when they make sense.

You came to this Forum where excluding myself, there are many, many knowledgable people to help you better understand the Technology, and thus provide you with the Tools to extract the BEST Performance from your Set Up. 

It seems though that you stopped by not with an interest in Learning more, but rather with an assortment of preconceived notions that you have Pontificated upon. and debated Ad Nauseum, whilst really only seeking Affirmation, and for the RME forum members to Rally around you as a Support Group. to Applaud, and Celebrate your every Purchase.   You know, this is a Classic instance where that is highly unlikely to happen. 

The Group tried to provide guidance.   Sorry it didn't work out for You.

Good Luck, and Enjoy your Cables.

First of all, take a chill pill :-)

Secondly, jitter does not make a zero of a one (or the opposite), jitter is when the 1 or 0 is placed in the wrong time-slot. And that happens all the time, in pretty much all digital transmissions, to a varying degree. In fact, with a 1ns jitter and a femto clock, that would mean the bit is placed 1.000.000 slots away from its "correct" place, but in this case it won't matter much since its such a small time-error.

I think the word time slot is not correct here. It suggests it takes the place of another time slot. There are no slots at jitter levels. There are slots if you want to call them that at sample level. At worst you could call it not exactly at the right time, but even this suggests a much to big problem. If you wiggle your head a little you create much much bigger variations.

The timing of the bits is irrelevant. It is the full sample wich is 2X24 bits plus extra data, which needs to be played back or recorded at the right time.

Someone should make a jitter plugin, that simulates jitter (yes that is possible, cause jitter is right level at wrong time, which is equal to wrong level at right time) and then people could AB it to hear which jitter level is perceivable.

How perceivable jitter is depends a lot of the audio system and room acoustics, but on my previous DAC I noticed a difference in sound quality when trying different toslink cables, and the only thing that could cause changes in SQ is jitter.

Is it not be possible to add a jitter test in the DAC, similar to the bit-perfect test that already exists? Maybe clock-jitter is hard to measure, but jitter up to the DAC chip should be possible to measure if you know the source stream.

That would make the question I started this thread with really easy to answer, and it would be a great tool to optimize sound quality with (which both audiophiles and studio professionals should be interested in).

Nahh, I never fool myself :-)

Just kidding, I know about expectation bias and placebo, but one way to get around that is to focus on a very small detail in the background of the music, and compare that at same volume. In this case I selected a faint background sound in the beginning of Jennifer Warnes Joan of Arc, 40 seconds into the tune if I recall correctly. That detail was not even possible to hear with the worst toslink cable (I would guess increased jitter masked it so it became part of the background noise).

Closing your eyes and trying to look in the direction of the leftmost sound, and then open eyes and measure distance, is a fairly objectively way to determine stereo width (not perfect, but better than just "listen").

And you think expectation bias can not cause you to not hear something that is there?  Did you read the article MC linked? One of the most interesting points is that even if the cable could make a difference with the ADi2 even then you could just as well like the one with more jitter best. Our ears  and brain have no purity measuring device build in...It is an absolute fact that people like distorted audio better then clean if it is not to much. Even that distortion makes details more audible. So the better cable you liked most could be worse jitter wise.
I hope you understand what I am trying to say...It is not pro's vs audiophiles. It is simply that ears and brains are not a measuring device. No matter what trick you try.