Yes, easy to overlook

I added numbers to the plug for more clarity.

https://www.dropbox.com/scl/fi/pooohmt1 … 3&dl=1

What exactly does not work, what is the card connected to?

Glad it works

But that Meridian thing I would take out of the house, it is electricity consuming for nothing.

From the product description :

HDSP 9632 debuts SteadyClock (TM), a newly developed clock technology, combining professional features like maximum jitter suppression at full varipitch capabilities and software controlled sample rates.

If that would be true, it would be possible to measure it but thats not the case, both digital outputs carry the same music from your routing.
The signal path afterwards seems to be not equal, one is a fraction louder that makes you think its significant better.
Or your routing in Totalmix was not the same, you know you can double-click the fader to let it jump up to the zero-position exactly.
Also check out ADI-2 DAC FS, regarding  jitter

i would never ! These stuff belongs to electro recycling, cables separate and electronics separate, thats all valuable stuff.

Btw your photo of the XLR plug shows a quality of soldering that is the worst. My teacher would have cut off that and shout at me.

The signal is the same, but transmitters and receivers of the signal are different. And the cables are very different.

None of these can change the audio content of the digital signal, otherwise that signal would be corrupted.

They don't change a signal, they change sound character. Idk how, but different digital cables affect sound differently. It's about very subtle things in very high resolyrion systems.

Sorry, but this is nonsense. Zeros and ones have no sound or sound character.
The signal doesn't have any waveform. It's a simple stream of digital states, either 0 or 1.
You seem to fool yourself with expectation bias.

Such "psychoacoustic effects" are known and have been researched.
In a properly conducted blind or double-blind test at the same volume levels, you would be unable to recognize a difference.

One thing upfront: in the RME forum, we try to provide information that is as honest, accurate, and technically sound as possible.

For this topic, proper test methodology is critical. Without it, listening comparisons easily lead to different—and often wrong—conclusions. There is nothing wrong with conducting listening tests; the problem only arises when they are not done correctly, allowing bias to influence the result.

Below is a brief overview of the relevant points.

Psychoacoustic limits and test methodology

If you want to conduct serious listening comparisons, you must consider well-known psychoacoustic thresholds and practical limitations. Otherwise, you will very quickly “hear” differences that do not physically exist.

Level differences

A level difference of only ≈0.2–0.3 dB is already sufficient for a signal to be perceived as clearer, more open, or more detailed.
From around 0.5 dB, this effect becomes obvious to many listeners.

This is why precise level matching is mandatory. Rough adjustment “by ear” is not sufficient for a valid comparison.

Temporal aspects / switching speed

The auditory short-term memory for fine spectral and dynamic details is limited to a few seconds—often significantly less.
As the time interval between two listening events increases, the brain increasingly replaces actual memory with expectation and reconstruction.

In practical terms, switching must occur almost immediately (typically < 1 second, ideally much faster), otherwise the comparison is no longer based on perception but on assumptions.

Multiple passes, not a single impression

A single listening impression is not meaningful.
You need multiple repetitions to rule out coincidence and random hits.

Test setup considerations

Depending on what exactly you want to test, setting up a comparison that meets all requirements (especially fast switching and level accuracy) may be more or less difficult.

Example setup (Bit Test)

A very simple and robust test is the Bit Test of the ADI-2 series reference converters, ideally combined with an RME audio interface and a PC audio player that supports ASIO (Windows).

Play the test pattern via different digital paths (AES, S/PDIF optical, S/PDIF coaxial, USB)

Ensure:
- the audio player outputs bit-perfect data,
- all faders in TotalMix FX are set to 0 dBFS.

Overview of such a setup

PC------------------------------------------+
|                                                          |
| USB2                                        USB2 |
|                                                          |
UCX II---AES-------------------------->ADI-2 Pro FS
             SPDIF(o)------------------------/   /
             SPDIF(c)---------------------------/

The test pattern (a specific digital sequence of 0s and 1s) is transmitted losslessly up to the DSP of the ADI-2 Pro FS.
The DSP detects this unique pattern and confirms on the display that the bit test has succeeded.

This proves that the digital audio stream is transmitted losslessly, regardless of whether AES, S/PDIF optical, or S/PDIF coaxial is used.

The final D/A conversion is performed entirely inside the ADI-2 Pro FS using its own DAC.
The result is identical for all digital inputs—including USB and ADAT—and independent of cable choice.

This is simply digital data transmission, which is either correct or incorrect; when correct, it is lossless.

Clocking and jitter

Technologies such as SteadyClock FS ensure that digital data arrives with minimal jitter at the device performing the D/A conversion. The ADI-2 Pro FS is a special case, as it always uses its internal femto-second clock for the final D/A process, even when operating as a clock slave.

Knowledge matters

Proper testing is finally the same as you proposed, "practice", but with the important difference of using proper methodologies.
A solid understanding of how the involved components and technologies work together greatly helps to interpret results realistically and avoid false conclusions.

The proposed bit test is maybe not something that you would belief, it needs perhaps too much knowledge of how it works. It is just an offer how you can proof quickly that the different digital transfer technologies do not alter the bits, in short: there can not be any sound difference, if the transport is "bit perfect".

Whatever you prefer, you can also do listening tests, but done in a proper way, otherwise you come to the wrong results.
If you trust your ears, then you can not have anything against such a test which excludes any bias, so please do so.

Thanks, but sorry, this seems to be the same issue that leads to wrong results.

I assume that it takes a little time to go to the HiFi amp, switch cables, return to your seat, and start playback again.

Therefore, we have two issues here:
1. You know what cable you plugged in, so you have an expectation bias.
2. Your brain cannot remember the sound for so long.

This is not a proper test methodology, excluding psychoacoustic effects.

I find that all devices and headphones I own sound astonishing, even all may mobiles including w800 sony walkman phone. All my headphones sound different but all extremely good.All my DA's. My studio speakers sky system one sounds fantastic, my self build speakers sound great. Even 10$ earbuds sound amazing. My old speakers, cheapish Tannoys don't sound that good and what really sounds awfull is my old pioneer headphones from the 80"s.
My mood, for lack of a better word, has the most influence on my listening experience. I understand that buying something new and shiny can put you in the right mood, but so does some meditation or *****.
Cheers

Tonal balance and dynamics of a music piece do change when you change the power cable ?
Why should studio-guys mixing engineers take care of tonal balance and dynamics at all ?
Makes no longer sense to produce music with effort, music becomes music thru Furutech cable, right ?

yes, exactly, otherwise it would be measurable.

Wait, do you claim that digital outputs of a random motherboard and a hdsp 9632 sound the same?

I reject this insinuation.

This is not about drawing a line between ‘studio professionals and audiophiles’ or denying anyone's listening experiences. The topic is exclusively about testing methodology and the question of under what conditions perceptions can be reliably separated from psychoacoustic effects.

If it is claimed that different digital transports (AES, S/PDIF), USB cables or power cables cause differences in sound (when devices are working correctly) then this is a technical claim. Such claims must be measurable or reproducibly verifiable. Without this, they remain subjective impressions – nothing more and nothing less.

Pointing this out is not an ‘ideological standpoint’, but elementary diligence. This is precisely why the RME forum regularly refers to clean level equality, blind tests and controlled conditions. Not to discredit anyone, but to avoid false conclusions arising from expectations, context or comparison errors.

No one is prohibiting personal listening or individual preferences. But as soon as tonal differences are presented as technically real, different standards apply than those used for purely describing personal impressions.

That was the point – nothing else.

What exactly do you find disgusting in that sentence? It's pure logic, isn't it? I personally don't care even if wisdom words comes from a discredited character.
And it seems like only you know THE TRUTH. 
Let me guess.. you are young and not so smart, am I right? It's ok, I'm not so smart too.
Anyway, thank you for attention and, let be honest, not the most interesting comment.

Honestly, the first paragraph was excessive.
The rest was poetic and wague at the same time.
Ok, data is the same, so then the point is about jitter!
The difference between motherboard and 9632 digital outputs is very noticeable!