"Audiophile Grade" Ethernet Switches - The new generation

[Flashman]

@Flashman

That is good news but may I ask a question. What resolution files are you streaming? I don't think Tidal has anything above 48 kHz, at least I haven't noticed anything including their NQA files which unpack to higher resolutions but stream at 48 kHz or less in my experience.I understand that DSD 64 is equivalent to around 88.2 or 96 kHz and DSD 128 would be twice that.

I'm streaming over a gigabit connection via a Cisco switch and don't get problems unless I do something like having Roon upsample everything to 24/192 which definitely triggers issues for me but DSD 64 and 24/176.4 and lower aren't issues for me. I'm not certain what the issue is but Roon say it's a buffer handling problem on Devialet's part. High res files can certainly trigger it but I don't think it's so much the resolution as the fact that more data is involved, the buffer size is measured in milliseconds so higher res files means more data gets held in the buffer and I wonder whether how much data is held in the buffer is part of the issue, but given that some people have problems streaming at CD resolution and I don't, I think there are other factors involved as well. It seems as if changing switches has been very beneficial for you but I've tried a number of nion-qaudiophile switches in my system over the last 10 months (switches from Netgear, D-Link, and Cisco, and also a Netgear Orbi hub's ethernet ports, and had similar good luck with all of them. My experience is that any standard switch works just as well for me as any other. I'm happily enjoying my success but I wish I knew what was going on and why what I'm doing works for me but doesn't for others. I'd also like to know why changing from one gigabit switch to a different gigabit switch worked for you and @Elztalbiker when changing to other switches hasn't worked for other people here.

Having the Roon Ready connection work is really great but it seems to be a hit or miss thing. I don't think either of us have solved the problem, we've each just found our own way to avoid it in our individual setups. What I really want is to see the problem solved for everyone with no need to resort to limiting connection speeds or the file resolutions we stream or having to change switches or the like. I just want to see Devialet fix the underlying problem.

David, I probably need to do some more experimenting.  Unfortunately, I have shut Roon down as I am leaving home and won't be back until late January.  At that point, I can do additional testing.  I am likewise puzzled as to why I can stream gigabit Ethernet in my setup, where previously it would not let me do so unimpaired.  I am pretty sure that that I upsample to 24/192 in Roon as a practice but I would need to check that when I return.  More later...

David, I am back from my long trip (our daughter just had our first grandchild, so I was out of pocket for one month) and have checked on some higher resolution files with my Devialet Expert 220 Pro connected by Ethernet to the Bonn N8 switch.  Everything plays nicely without a glitch.  I tried quickly going from one song to another and playing long stretches of music and had no issue.  The highest resolution files I have are 352.8kHz/24 bit.  Typically, for files beneath 192kHz/24 bit, I upsample to this resolution.   I have no idea why the Bonn N8 plays so nicely with the Devialet and Roon RAAT but it does.  A nice surprise!

[David A]

@Flashman

Are you suggesting that grandchildren make us out of pocket? Strange, I've noticed that too :-)

The whole thing about streaming high res files with RAAT over a gigabit connection has me scratching my head too. Roon have said there's a problem with how the Devialet handles the queue and I have no reason to doubt that but that raises the question off why the problem itself is intermittent. I could trigger it by streaming 192/24 content but sometimes the problem would start almost immediately when I started streaming at that resolution, sometimes it would take an hour or so, and the regularity with which it would occur when it started wasn't regular either. I never really had problems at lower resolution and others do. I can accept queue handling being the problem but that still leavess the question of what makes the queue harder to handle at some times than at others when using a gigabit connection. I don't have an answer to that and now, just 3 weeks short of a year after we got the Roon Ready firmware update from Devialet, I have even fewer ideas of what the problem is than I did when it first started being reported.

All I can say at this point is that it is now clear that there are at least workarounds for people experiencing the problem, namely connecting the Devialet to the network using either a 100T connection or wifi. I suspect that Devialet themselves could eliminate the problem with a firmware update that changed the ethernet port settings to 100T and no one would notice anything other than the disappearance of the problem. We don't need gigabit speeds for music streaming.

[mdconnelly]

@David A - It is a mystery. I am quite curious as to how "audiophile" switches such as the Bonn N8 have made the problem go away for those that were experiencing it. It does make sense with the EtherRegen since the port to the Devialet is at 100 mbps. We may never get a straight answer on it, much less see it fixed.

For awhile I suspected it might be a hardware issue on some but not all Expert Pros. That could be a costly fix they they didn't want to swallow just for Roon RAAT. Then again, your comment about them just doing a firmware update to make the ethernet port 100mbps would make so much sense.

I may just give one of these switches a try after awhile.

[Flashman]

@Flashman

Are  you suggesting that grandchildren make us out of pocket? Strange, I've noticed that too  :-)

Our first grandchild -- Daniel Patton Reed -- will empty our pockets, too!  We're more than willing! ;-)

[Flashman]

@David A - It is a mystery.  I am quite curious as to how "audiophile" switches such as the Bonn N8 have made the problem go away for those that were experiencing it.  It does make sense with the EtherRegen since the port to the Devialet is at 100 mbps.  We may never get a straight answer on it, much less see it fixed.

For awhile I suspected it might be a hardware issue on some but not all Expert Pros.  That could be a costly fix they they didn't want to swallow just for Roon RAAT.  Then again, your comment about them just doing a firmware update to make the ethernet port 100mbps would make so much sense. 

I may just give one of these switches a try after awhile.

@David A@mdconnelly It's a puzzler, no?  The suggestion that Devialet simply provide a firmware update to make the Ethernet port 100Mbps is brilliant but Devialet will never do it.  The fact that a different switch solved my issue may be one of the reasons why Devialet may not have been able to isolate the problem and thus provide a fix.  I am being generous to our good friends at the company.

@mdconnelly The Bonn N8 is also available in an upgraded form -- better low-noise power supply and aluminum chassis -- via NuPrime (Omnia SW-8) for USD449.  That's only USD50 above the Bonn N8 so seems like a good deal.

[K4680]

@David A - It is a mystery.  I am quite curious as to how "audiophile" switches such as the Bonn N8 have made the problem go away for those that were experiencing it.  It does make sense with the EtherRegen since the port to the Devialet is at 100 mbps.  We may never get a straight answer on it, much less see it fixed.

For awhile I suspected it might be a hardware issue on some but not all Expert Pros.  That could be a costly fix they they didn't want to swallow just for Roon RAAT.  Then again, your comment about them just doing a firmware update to make the ethernet port 100mbps would make so much sense. 

I may just give one of these switches a try after awhile.

@David A@mdconnelly It's a puzzler, no?  The suggestion that Devialet simply provide a firmware update to make the Ethernet port 100Mbps is brilliant but Devialet will never do it.  The fact that a different switch solved my issue may be one of the reasons why Devialet may not have been able to isolate the problem and thus provide a fix.  I am being generous to our good friends at the company.

@mdconnelly The Bonn N8 is also available in an upgraded form -- better low-noise power supply and aluminum chassis -- via NuPrime (Omnia SW-8) for USD449.  That's only USD50 above the Bonn N8 so seems like a good deal.

Something to read!

[Petter]

https://www.stereo.net.au/forums/topic/2...periences/

Some intresting listing of the diffrent switches and opinions ,for what it is wort

[Greg]

I’ve not read all of this thread (dipped in and out because I’m a sceptic) so apologies if this link has already been previously posted. Completely objective but interesting reading.

https://www.audiosciencereview.com/forum...iew.10232/

[no32]

I’ve not read all of this thread (dipped in and out because I’m a sceptic) so apologies if this link has already been previously posted. Completely objective but interesting reading.

https://www.audiosciencereview.com/forum...iew.10232/

I hear big positive deference listening to tidal through UpTone.
I read many articles that explain why it’s all one big BS.
I assume that in my case it’s all psychosomatic since going through the measurements there’s no way I hear any deference.




Sent from my iPhone using Tapatalk

[David A]

I’ve not read all of this thread (dipped in and out because I’m a sceptic) so apologies if this link has already been previously posted. Completely objective but interesting reading.

https://www.audiosciencereview.com/forum...iew.10232/

The link has been posted before and discussed before. The review is interesting but hardly objective.

For my comments on a number of the issues with the review, refer to post #409 in this thread. The problem with all tests is they tell you something about what they measure and they tell you nothing about what isn't measured. A comparison with a single "control" switch tells us nothing about how the EtherREGEN compares to switches generally. The "control" switch in unnamed and we are given no data to show that it is representative of switches in general. If switches make a difference, then the "control" may be really good, better than most other switches, in which case that would also mean that the EtherREGEN is also really good. Then again if the "control" is worse than most other switches then the EtherREGEN will also be worse than most other switches. Since we know nothing about the control, we know nothing about the EtherREGEN other than it measured the same as the "control" on the tests conducted and we can draw no conclusions about whether or not either switch is better, worse, or similar than any one of the numerous other switches available. You can't say that the EtherREGEN can't make a difference without knowing anything about the many switches that weren't tested, or anything about the "control" switch other than it measured identically on certain tests.

We have no idea whether the tests conducted were the correct tests to show any differences because Uptone Audio haven't provided any test data to show that it is better than other switches.

As for the listening test in the review, it was a sighted test conducted under circumstances which were problematic in my view. Note that if you conduct your own listening test and believe you hear a difference the reviewer states as follows"

"You hear otherwise? Well, I tested it and there is no audible difference either. I suggest you repeat my test. If you do hear a difference, have someone switch cables behind your back. You can do this quickly, or wait days or weeks to switch. Just don't look and keep a log of 10 trials. If you can tell better than 8 out of 10 times that there is an audible difference with EtherRegen, then you have conquered the impossible!"

So the reviewer believes he can be certain there is no difference on the basis of a sighted listening test in which he reports a swap from "control" to EtherREGEN and then from EtherREGEN to "control". If you don't hear the same as he does, he wants you to do 10 trials on a blind test basis, not the 2 trials he did on a sighted basis, and he says you will then come to the same conclusion as he did. That implies that he thinks he can be absolutely certain on the basis of a very simple sighted test but he wants anyone who doesn't hear what he reports when they do the same as he did to do a much more stringent test which he believes he has no need to do himself because he can get it right first time. Now that is hardly objective. One of the reasons why test procedures are outlined in detail in scientific papers is so that others can perform exactly the same test procedure to verify the original results. If you can't verify the original results using identical procedures then something is going wrong, either the original results were mistaken or the test procedure isn't a good procedure. There are many other reports of sighted tests where listeners have reported hearing a difference. Either there is a difference which the reviewer did not hear, or sighted tests are not a good test procedure in this case and if they aren't a good test procedure for other people then they can't be a good test procedure for the reviewer either.

In summary, the test results reported in the review are interesting but they fail to establish what is claimed for them.

The onus of proof does fall on the claimant so it's up to Uptone Audio to demonstrate that their product does what they claim but if someone wants to produce evidence that purports to show that it doesn't do what is claimed then there is an onus on them to actually produce evidence which really does hold up to analysis and this review does not do that. At best it shows that the EtherREGEN performs identically to another switch on some tests. That does not establish that it does not produce measurable improvements in areas which weren't tested. The listening test proves nothing, especially given the suggestions provided for anyone who hears a difference. If a listener is not allowed to say there's an audible difference on the basis of a couple of sighted comparisons then the reviewer can't be allowed to say there is no audible difference on the basis of a couple of sighted comparisons. The test procedures need to be identical for all listeners or the results are equally worthless no matter what they indicate.