I was asked in private to comment about what is shown here with respect to network switches making a difference in a DAC:
Summarizing if you don't want to watch it, he is probing the clock input to a DAC in a streamer. He then puts a power supply next to the Ethernet cable and also uses a current transformer around the Ethernet cable to show that there is a 20% or so increase in clock jitter of the DAC. This is not surprising as there can always be some leakage if you push things hard. I have actually shown the same with respect to Ethernet cables (see: https://www.audiosciencereview.com/...ments-of-emo-en-60kds-ethernet-isolator.1869/)
As we see there, the mains hum that I induced actually appears in the output of the DAC if one doesn't use the isolated cable i was testing.
The key though is its level: we are talking near -120 dB! Such is also the case in the above video. He measures something called "Total Jitter." I don't know what that means but it goes from 130 to 150 picoseconds or something like it. He has another metric he says is "average jitter" and that is in the order of 10 picoseconds. Again, these are variations in the purity of the clock input to the DAC, not the impact it has on its output. But assuming it is, 512 picoseconds of jitter represents the low order bit of a 16 bit audio sample (peak to peak). Given that, what they are measuring is well below the noise level of 16 bit audio -- similar to what I measured above.
In other words, in really contrived examples we can show a measurable difference but it is impossible to make a case of audibility.
When I have measured such devices using the analog output of a streamer, I have not found any difference anyway: https://www.audiosciencereview.com/.../uptone-audio-etherregen-switch-review.10232/
Without:
With fancy switch:
And this is in a very noisy scenario of my workstation feeing from the same source, etc.
I have also done a video on this:
Summarizing if you don't want to watch it, he is probing the clock input to a DAC in a streamer. He then puts a power supply next to the Ethernet cable and also uses a current transformer around the Ethernet cable to show that there is a 20% or so increase in clock jitter of the DAC. This is not surprising as there can always be some leakage if you push things hard. I have actually shown the same with respect to Ethernet cables (see: https://www.audiosciencereview.com/...ments-of-emo-en-60kds-ethernet-isolator.1869/)
As we see there, the mains hum that I induced actually appears in the output of the DAC if one doesn't use the isolated cable i was testing.
The key though is its level: we are talking near -120 dB! Such is also the case in the above video. He measures something called "Total Jitter." I don't know what that means but it goes from 130 to 150 picoseconds or something like it. He has another metric he says is "average jitter" and that is in the order of 10 picoseconds. Again, these are variations in the purity of the clock input to the DAC, not the impact it has on its output. But assuming it is, 512 picoseconds of jitter represents the low order bit of a 16 bit audio sample (peak to peak). Given that, what they are measuring is well below the noise level of 16 bit audio -- similar to what I measured above.
In other words, in really contrived examples we can show a measurable difference but it is impossible to make a case of audibility.
When I have measured such devices using the analog output of a streamer, I have not found any difference anyway: https://www.audiosciencereview.com/.../uptone-audio-etherregen-switch-review.10232/
Without:
With fancy switch:
And this is in a very noisy scenario of my workstation feeing from the same source, etc.
I have also done a video on this:
