Uptone EtherREGEN

[DonH56]

^^^ Well stated, Sir Thomas. Hope you get the weekend off! (I was at 40 hours by early Thursday morning, with at least another 20 or so to go this week, making it hard to practice in the evening. Then I waste time posting nonsense.)

 

[somebodyelse]

or somebody else said...

Deny everything, Baldrick.

 

[DonH56]

This thing is receiving rave reviews on another audiophile site, natch, including the improvement after 30 hours of break-in. There is some serious misunderstanding of the path the signal takes from ethernet to DAC...

 

[Veri]

This thing is receiving rave reviews on another audiophile site, natch, including the improvement after 30 hours of break-in.

Well of course. Gotta break that crazy ugly thing in sigh, audiophiles

 

[Jinjuku]

So you guys never seen the light of the EtherRegen, and yet, you are absolutely certain it does nothing? And willing to bet $10,000 on it? What happened to measurements?

If I have a well buffered playback system and I start playing back a track and have the port shut down or cable removed and can't tell a difference in sound then yes I'm absolutely certain it can't IMPROVE the playback.

 

[dc655321]

If I have a well buffered playback system and I start playing back a track and have the port shut down or cable removed and can't tell a difference in sound then yes I'm absolutely certain it can't IMPROVE the playback.

I wonder what it would take to illustrate that this device cannot produce an audible affect?

Here's a simple experiment (not as simple as your "pull the plug" test, but still...), sort of a lazy null-test:

Download a music (or any other!) file through the EtherRegen to a computer. Save the file.
Replace the EtherRegen with any other switch/router of similar bandwidth and download the same file and save it under a different name.
Now produce a digital fingerprint for each file using md5 or sha-1 or similar software.

If even a single bit is different, the result will be very different message digests for each file.

I know where I would put my money...

 

[amirm]

Download a music (or any other!) file through the EtherRegen to a computer. Save the file.
Replace the EtherRegen with any other switch/router of similar bandwidth and download the same file and save it under a different name.
Now produce a digital fingerprint for each file using md5 or sha-1 or similar software.

If even a single bit is different, the result will be very different message digests for each file.

While the marketing for these products tend to let the buyer believer they reduce errors, their key thesis is that they improve DAC performance somehow by filtering noise, isolating high frequency noise, etc. So to rule this out, or rule it in, the analog output of the DAC needs to be tested. The company is refusing to make this measurement. I am volunteering to measure. Until we do, we can't prove indifference with the test you propose.

That said, I do plan to test the reliability of the product in producing error free switching as we can't take that for granted.

 

[BDWoody]

You’re loosing your time here... Nobody will seriously engage in an open discussion. The first thing they will ask is to show measurements. I’d say my measurements are my one ears. The rest I really don’t care. I received my Etherregen and I am very happy with it. For the others that claim that Ethernet switches don’t do a difference, let them stay in the ignorance. I don’t have patience nor time to try to convince them otherwise.

Uh huh.

 

[dc655321]

Until we do, we can't prove indifference with the test you propose.

I know you know this, but my proposal would provide solid evidence of bit-wise indifference without needing produce an analog signal.
I fully realize that may not satisfy neuroses or ignorance. I'm also certain showing indifference in DAC output would also be met with, well, indifference.

I am volunteering to measure.

That said, I do plan to test the reliability of the product in producing error free switching as we can't take that for granted.

Great. Looking forward to the report.

 

[pkane]

While the marketing for these products tend to let the buyer believer they reduce errors, their key thesis is that they improve DAC performance somehow by filtering noise, isolating high frequency noise, etc. So to rule this out, or rule it in, the analog output of the DAC needs to be tested. The company is refusing to make this measurement. I am volunteering to measure. Until we do, we can't prove indifference with the test you propose.

That said, I do plan to test the reliability of the product in producing error free switching as we can't take that for granted.

That's exactly it. It's not bit errors that the product claims to correct, it's removing the upstream noise from infecting the DAC circuits and ultimately, resulting in a less accurate sound reproduction.

Bits are the same, but the extra noise alters the analog output of a DAC. The process might be the actual noise mixing or modulating the analog output, or, maybe some of this noise affecting the DAC clock resulting in jitter at the output. Both should be measurable from the analog output of a DAC.

Pulling the plug or saving the files to disk will not prove anything to the believers, because they mostly already agree that bits are not altered in the transmission.

 

[KSTR]

It's very simple. Anyone who has the slightest understanding of how ethernet works, how computers work, digital audio works, and is capable of just little bit of critical thought, will conclude there is no mechanism for this to make any improvement.

It's a con. We don't need to see or measure it to draw an accurate conclusion. It's about not believing in fairies , not blindly believing the marketing claims and using existing knowledge to inform yourself.

Full Ack.

Nonetheless I can see a very(!) faint mechanism where an electrically better switch could make a difference, and that would be some extreme RF-susceptibility of the DAC. The digital section might be producing more jitter on the final DAC-chip's data/clock lines, and/or it sneakes directly into the analog signal path via demodulation), DAC probably part of a streamer etc. Though it would be really tough to design such a piece of sh!t on purpose, as a design excercise, haha....

EDIT: ah, crossposted with Paul ;-)

 

[dc655321]

Pulling the plug or saving the files to disk will not prove anything to the believers, because they mostly already agree that bits are not altered in the transmission.

Do they? I've not seen evidence of rational thought process by proponents, just "you have to listen!".

 

[amirm]

Pulling the plug or saving the files to disk will not prove anything to the believers, because they mostly already agree that bits are not altered in the transmission.

Pulling the plug is a good test though. Say I am monitoring the distortion and noise in real time with my analyzer and I pull the plug as the streamer keeps feeding the DAC. Surely the measurements of the DAC output would change if the upstream switch is doing some good.

 

[DonH56]

The argument is being made that the clock is much cleaner and that leads to better sound from the DAC. Users are saying this in various reviews. That tells me that (a) they have a lack of understanding of the data (and clock) path from the Ethernet port to the actual DAC, and (b) marketing works. Oh, and (c) so does cognitive bias...

 

[KSTR]

To defend the product (not the company nor their strategy) one could argue that the device is useful in the lab where the high isolation allows for things like impressing exactly known RF common-mode noise patterns on the downstream ethernet cable and see whether this gives measureable/audible artifacts.

 

[amirm]

To defend the product (not the company nor their strategy) one could argue that the device is useful in the lab where the high isolation allows for things like impressing exactly known RF common-mode noise patterns on the downstream ethernet cable and see whether this gives measureable/audible artifacts.

Good point but they need to provide measurements to show all of this is true. Saying it is not enough.

 

[KSTR]

It's not an industrial or medical device and therefore it won't happen, IMO.

 

[amirm]

The argument is being made that the clock is much cleaner and that leads to better sound from the DAC. Users are saying this in various reviews. That tells me that (a) they have a lack of understanding of the data (and clock) path from the Ethernet port to the actual DAC, and (b) marketing works. Oh, and (c) so does cognitive bias...

We all suspend disbelief when watching Star Trek show that there is such a thing as photon torpedos. The jargon they use with clock and phase noise like photon torpedos, sound real enough to audiophiles so they take it as true. Shame that they (audiophiles) don't allow us to explain to them such things are no more real than photon torpedos are.

 

[Deleted member 65]

Having built national & international Cisco networks as PM/Network Manager for a major bank I’m stunned by the lack of competence from the provider of this product. Probably searched for a new niche since USB has been covered.

Only issue the bank found when connecting the countries and datacenters was the speed of light, no errors whatsoever in the network between the four countries. Due to the speed of light restriction (see latency and competition with other players) it was decided to centralize Markets to one country.

 

[DonH56]

We all suspend disbelief when watching Star Trek show that there is such a thing as photon torpedos. The jargon they use with clock and phase noise like photon torpedos, sound real enough to audiophiles so they take it as true. Shame that they (audiophiles) don't allow us to explain to them such things are no more real than photon torpedos are.

Wait, photon torpedoes aren't real? Next you'll be saying no warp drive, no transporters, no Spock... I swear if you say there's no Santa Claus I'm going to get Thomas and we're heading to your place to straighten you out!