But that is not the metric that determines audio fidelity. Imagine two scenarios:If you want data, don't you think it is the best to ask them directly? If they can't provide, feel free not to buy this product. Given that the motherboard is customized, no fan and off the shelf nosiy switching power supply, it is very hard to imagine that it is noiser than a general purpose computer crunching intensive calculations. How much better I am not sure.
1. A very busy computer with tons and tons of activity. And let's assume this bleeds into the audio output channel in some manner. The very busy and highly chaotic manner of the computer activity will appear as noise whether induced as jitter or noise alone. So your digital system may go from having 100 db S/N to say, 90 db S/N.
2. A very quiet computer that services different tasks such as receiving packet traffic over Ethernet, USB service interrupts, system timer, etc. All of these events will be in 100s to 1000s of times/sec -- right in the audible band! And without noise to mask them, you will have correlated distortion that is far, far more audible that pure noise. A 100 db S/N ratio with spikes that rise up 10 db will be worse than system #1.
What you want for audio hardware is a system that has constant load and hence constant/random imprint on its audio interface. A quiet system with jumps in activity is the opposite of that.
This is all of course assumes there is such bleeding into the audio interface. With good USB input on a DAC, these differences vanish to levels well below audibility.
BTW, making things small is often the enemy of making things good. With small size, components get placed together and more noise bleeds/couples into adjacent component. I for example cringe at the Ethernet Jack being so close to USB on this device. Why did it need to be so small? Did they start with Raspberry Pi or some other platform like it???
Of course, human nature being what it is, will lead one to think that this little box must be more quiet and hence, that is what is heard subjectivity. We need data to get past that bias so that we can evaluate the device fairly.
I will go ahead and buy one to measure but I sure like to see the manufacturer offer such measurements. If they have it but requires us asking, there is something wrong there.
Again, this may be a good/superior solution. The discussion here is to not give benefit of doubt that goes against how the technology works.