What happens to an audio signal from media player to dac

[Julf]

Originally Posted by abraxalito

I was speaking of my experience (IME expands to 'in my experience'), not my theory.

You have experienced some effects and possibly made some observations by making measurements. You are postulating an explanation for those effects. I would call that a theory.

[abraxalito]

You're welcome to call it what you like In case it has escaped your notice, in science there's only ever falsification, never proof.

[mr_push_pull]

Originally Posted by Julf

You have experienced some effects and possibly made some observations by making measurements.

exactly what I wanted to say. I'd guess that common mode noise is not differentiated from jitter by listening alone, so maybe abraxalito has some measurements. also, aren't there any more effective solutions for CM noise than cables (low capacitance?)? and if a good cable is the only solution, is there any esoteric technology involved that justifies the prices?

[abraxalito]

Yes there are ferrites for CM noise on cables. Or CM chokes for balanced cables like USB - Steve Nugent (Empirical Audio) is even selling some CM chokes of his own design for a tidy sum Personally I would recommend spending as much on ferrites as on the cable - no more than single digit $.

[mr_push_pull]

but then again, doesn't this mean that all those waxing lyrical about cable differences are describing noise flavors? aren't there readily-available USB cables with ferrites already fitted? I think I have one that came with some gadget at home.

[mr_push_pull]

one more thing I meant to add, unrelated to the above. the accepted wisdom is that it's not the jitter amount alone that matters but its characteristics. random jitter is said to be less problematic than correlated jitter.

[abraxalito]

Originally Posted by mr_push_pull

but then again, doesn't this mean that all those waxing lyrical about cable differences are describing noise flavors?

Intermodulation distortion caused between the audio and the HF noise yes. That would be my hypothesis. The solution that arises from this hypothesis is to provide better CM filtering on cables.

Here's a theory for you Julf as you're seemingly interested in those: The audio industry at large doesn't like the SQ differences to be down to something tangible like CM noise, they'd prefer to keep selling stuff that fixes up non-existent jitter problems. Like picosecond clocks for example. What do you reckon?

aren't there readily-available USB cables with ferrites already fitted? I think I have one that came with some gadget at home.

Yes but probably only to meet EMC regs, not to give low IMD.

[Julf]

Originally Posted by abraxalito

Intermodulation distortion caused between the audio and the HF noise yes. That would be my hypothesis. The solution that arises from this hypothesis is to provide better CM filtering on cables.

The USB data cables are differential / balanced, so high common mode rejection should be easy to achieve. Ground/power is more problematic - but if you don't use the USB power, but have an independently powered DAC, galvanically isolating the connection is trivial.

This of course doesn't apply to optical - no common mode issues there.

Here's a theory for you Julf as you're seemingly interested in those: The audio industry at large doesn't like the SQ differences to be down to something tangible like CM noise, they'd prefer to keep selling stuff that fixes up non-existent jitter problems. Like picosecond clocks for example. What do you reckon?

I agree that you probably have a point about the high end of the audio industry not being too keen on the answer being too easy, but I don't think that applies to the whole audio industry.

[mr_push_pull]

Originally Posted by abraxalito

Intermodulation distortion caused between the audio and the HF noise yes. That would be my hypothesis. The solution that arises from this hypothesis is to provide better CM filtering on cables.

worth trying as long as it's in the one figure range.

Originally Posted by abraxalito

Here's a theory for you Julf as you're seemingly interested in those: The audio industry at large doesn't like the SQ differences to be down to something tangible like CM noise, they'd prefer to keep selling stuff that fixes up non-existent jitter problems. Like picosecond clocks for example. What do you reckon?

picoseconds you say? how about the $10k femtosecond MSB clock?

Originally Posted by abraxalito

Yes but probably only to meet EMC regs, not to give low IMD.

worth giving it a try though.

(nice sig BTW. but JMK is defunct as of now which makes for an interesting problem )

[abraxalito]

Originally Posted by Julf

The USB data cables are differential / balanced, so high common mode rejection should be easy to achieve. Ground/power is more problematic - but if you don't use the USB power, but have an independently powered DAC, galvanically isolating the connection is trivial.

Haven't seen a trivial (i.e off-the-shelf) way to isolate UAC2 yet but I had heard that ADI was working on a solution. You know of a trivial solution?

I agree that you probably have a point about the high end of the audio industry not being too keen on the answer being too easy, but I don't think that applies to the whole audio industry.

Fair point - consider my post edited to qualify the word 'audio' with 'high end'.