advertisement


Audiophile Network Switches for Streaming ... really ?

Unless there is some 'bad blood' in a particular case for some cause that seems an unnecessary norm. Often when I was more active in audio discussions on AudioKarma I and others would reference resources available elsewhere, which was often quite helpful.

To be fair I don’t think that was the issue as I’ve seen other links to ASR etc that weren’t culled. I was just guessing tbh.
 
If you capture the before and after signals and send them to me :)

@Fourlegs DAC and associated power supplies is probably worth £25k (correct me if I am wrong). There are two massive power supplies that each looks like a Krell power amp feeding a Chord Dave and M scaler. Presumably you would need an A to D converter of comparable quality to get meaningful results from the test you propose.
 
I called it a flight of fancy because that's exactly what it is. Theres no measurable difference, the difference only exists in the mind of the listener.

If it didn't the manufacturers of all these add ons, all of them, clock, switches, reclockers, fancy psu would be supporting their sales with measurements, because who'd want to miss out on the opportunity that would offer to bolster sales.
 
Last edited:
@Fourlegs DAC and associated power supplies is probably worth £25k (correct me if I am wrong). There are two massive power supplies that each looks like a Krell power amp feeding a Chord Dave and M scaler. Presumably you would need an A to D converter of comparable quality to get meaningful results from the test you propose.

I think the Dave was £8k when I bought it (they are £10k now) and the Sean Jacobs power supply is £6.6k so that total that I paid for the Dave set up is £14.6k. But there is mscaler and power supply so £20k all in is more like it. So your guesstimate is a bit toppy but your point is fair.

More to the point is that even if I did use an ADC to capture the analogue output from the Dave @chiily would have to use a device to output that digital file(s) to his DAC and then measure the output from his DAC thereby needlessly introducing other factors in the process. He would no doubt claim that he would be looking for differences in the two files so outputting through his DAC should be the same for both of them but it is entirely possible that the distortion from his system could swamp the difference in my two files thereby negating the validity of the process. There is also the question of what device he would be using to do the distortion measurements. I would suggest it needs to be of the calibre of an APx555. I certainly do not have one of those and I doubt that he has one although I am happy to be corrected if I am making a wrong assumption. Anyway, the fact remains that the only proper way to do any measurements is without introducing variables by trying to measure from recordings and I am happy to participate in direct measurements if @chiily wants to bring his measuring kit to me (it would be useful to know though what he has that is capable of measuring the suspected IMD caused by RF noise). I am borrowing an AfterDark modified S100 switch for a couple of weeks in January and we could throw that into the mix to measure the system with it against the PhoenixNET (from my experience it certainly sounds different to the Phoenix so measuring would be interesting) and of course we would measure the system without either of them.
 
I called it a flight of fancy because that's exactly what it is. Theres no measurable difference, the difference only exists in the mind of the listener.

If it didn't the manufacturers of all these add ons, all of them, clock, switches, reclockers, fancy psu would be supporting their sales with measurements, because wood wants to miss out on the opportunity that would offertory bolster sales.
Will leave to the security of this fantasy
 
I think the Dave was £8k when I bought it (they are £10k now) and the Sean Jacobs power supply is £6.6k so that total that I paid for the Dave set up is £14.6k. But there is mscaler and power supply so £20k all in is more like it. So your guesstimate is a bit toppy but your point is fair.

More to the point is that even if I did use an ADC to capture the analogue output from the Dave @chiily would have to use a device to output that digital file(s) to his DAC and then measure the output from his DAC thereby needlessly introducing other factors in the process. He would no doubt claim that he would be looking for differences in the two files so outputting through his DAC should be the same for both of them but it is entirely possible that the distortion from his system could swamp the difference in my two files thereby negating the validity of the process. There is also the question of what device he would be using to do the distortion measurements. I would suggest it needs to be of the calibre of an APx555. I certainly do not have one of those and I doubt that he has one although I am happy to be corrected if I am making a wrong assumption. Anyway, the fact remains that the only proper way to do any measurements is without introducing variables by trying to measure from recordings and I am happy to participate in direct measurements if @chiily wants to bring his measuring kit to me (it would be useful to know though what he has that is capable of measuring the suspected IMD caused by RF noise). I am borrowing an AfterDark modified S100 switch for a couple of weeks in January and we could throw that into the mix to measure the system with it against the PhoenixNET (from my experience it certainly sounds different to the Phoenix so measuring would be interesting) and of course we would measure the system without either of them.
That's the bit about all other things being equal. The only change to the audio chain is the switch. All the other constants cancel out.

Just record the audio output at sufficient resolution one with the switch, one with out, keep everything else constant, easy.
 
That's the bit about all other things being equal. The only change to the audio chain is the switch. All the other constants cancel out.

Just record the audio output at sufficient resolution one with the switch, one with out, keep everything else constant, easy.

Can you humour me and let me know what measuring device you will be using?
 
Can you humour me and let me know what measuring device you will be using?
I'm just going to diff (subtract) the two signals in the digital domain. Take one value from another, nothing fancy at all really.

I have a copy of Adobe audition somewhere, that would do it. Just needs to be the same bit of audio to align them correctly.

You could try it yourself using https://www.audacityteam.org/
 
I'm just going to diff (subtract) the two signals in the digital domain. Take one value from another, nothing fancy at all really.

I have a copy of Adobe audition somewhere, that would do it. Just needs to be the same bit of audio to align them correctly.

You could try it yourself using https://www.audacityteam.org/
Even easier, generate a checksum for each file and compare them. Something like md5 or sha256.
 
I'm just going to diff (subtract) the two signals in the digital domain. Take one value from another, nothing fancy at all really.

I have a copy of Adobe audition somewhere, that would do it. Just needs to be the same bit of audio to align them correctly.

You could try it yourself using https://www.audacityteam.org/

But you need to get the output from the DAC back to digital with an A to D converter which has clocks and power supplies of comparable quality to those in the DAC. Such a device would likely cost 10’s of thousands of pounds. Rob Watts has been working on one for years.

You would need to do the comparison at a far higher bit rate than than the original digital file. Something like 768 kHz/24b.

If you are talking of comparing the digital output from the switches, that argument was dismissed on page one of this thread. No-one is suggesting that the different switches output different data in the digital domain. In other words they are all bit perfect.
 
@Fourlegs
@andrewd

You are both making things far too complicated....
Any recordings made from the DAC analogue output will do ... you are simply looking for a difference in the analogue waveform with or without switch 'cleansing'.
It really doesn't matter what additional noise the A-D converter itself might add (if any) because this would be a constant across the two states.

Even Audacity would do fine for this job ... and it's free ;)
 
That's the bit about all other things being equal. The only change to the audio chain is the switch. All the other constants cancel out.

Just record the audio output at sufficient resolution one with the switch, one with out, keep everything else constant, easy.

Exactly!!!
You would almost think folks are trying to make it seem impossible (or really expensive) to do :D
 
@Fourlegs
@andrewd

You are both making things far too complicated....
Any recordings made from the DAC analogue output will do ... you are simply looking for a difference in the analogue waveform with or without switch 'cleansing'.
It really doesn't matter what additional noise the A-D converter itself might add (if any) because this would be a constant across the two states.

Audacity would do fine for this job ... and it's free ;)

Even simpler is to have an assistant swop unsighted between switches say 10 times. When my friend with the S100 switch and Antipodes Oladra streamer visited he was adamant that he preferred the music streamed over Qobuz rather than played from the Antipodes Oladra internal SSD. 'Just for fun' as they say I had him swop between the two options in a random way unsighted by me. I just wanted to see if there was an audible difference. I correctly identified the Qobuz option 8 out of 10 times (it was the same qobuz file that had been downloaded from qobuz and saved to the SSD or was being streamed 'live').

We didn't bother to do the same test with the switches because to my ears that was a bigger difference than the Qobuz vs SSD source changes but of course it could easily be done and with so much less faff than what is being proposed.
 
We didn't bother to do the same test with the switches because to my ears that was a bigger difference than the Qobuz vs SSD source changes but of course it could easily be done and with so much less faff than what is being proposed.

Double blind testing = 'less faff' than recording the output twice ???

You really are avin a larf :D
 
I'm just going to diff (subtract) the two signals in the digital domain. Take one value from another, nothing fancy at all really.

I have a copy of Adobe audition somewhere, that would do it. Just needs to be the same bit of audio to align them correctly.

You could try it yourself using https://www.audacityteam.org/
But the difference isn’t in the digital domain! How many times does this need to be stated?
 
But the difference isn’t in the digital domain! How many times does this need to be stated?

And the recording isn't from 'the digital domain' either ... How many times does this need to be stated?

The recording is taken from the 'ANALOGUE' output of the DAC !!!!
 
And the recording isn't from 'the digital domain' either ... How many times does this need to be stated?
Too quick on the keyboard there: the post to which I was responding which was about comparing the digital files. Here for your ease of reference:

I'm just going to diff (subtract) the two signals in the digital domain. Take one value from another, nothing fancy at all really.

Apology accepted.
 


advertisement


Back
Top