Network switch, FMC, cable upgrades worthwhile when router can't be co-located?

[UltraFast69]

Ed, Have you considered running Fiber the 45' between your switches instead of coax? I've replaced my 100' of ethernet with Fiber between two Startech fmc's. I may try the Sonore Optical Module fmc's instead of the Startech also.

I was thinking about that, but we recently did a complete overhaul to the power foundation - it is off the charts good, no joke (will post this).

So with that I pulled the dedicated circuit from the original sub panel and added it to the new panel and relocated the digital foundation (mode, etherREGEN, LPS and ADD POWR pieces 45’ closer but still in the other room.

Instantly you could hear things more open, dynamic and amazing sense of presence. After 6 hours I shut things down, paused 4/5 hours and just started up the system.

I have given up saying things can’t get better, this is off the charts after 30 minutes.

The AQVox SE has been pulled from 2 channel duty, and repurposed into home. I simply reversed the no longer needed Ghent 45’ Belkin cable coming out etherREGEN to the AQVox SE.

All these changes by better reconfigurement I ended up with another Audioquest Vodka Ethernet, Stillpoints Ultra SS, and a Ghent DC Oyaide Cable proving less is more - but in all fairness I had a head start with the addition of the panel.

 

[Willco]

That leaves out streaming though. That being said, streaming (Tidal, etc.) is like the buffet at The Golden Corral. Plenty to eat, but I wouldn’t call it good. Your ripped CD’s will sound better and a CD playing through the same DAC, better yet again. I would say:

CD = 10/10
Ripped CD = 8-9/10
Streaming = 7/10

There are quite a few people using the widely acknowledged sota Taiko Extreme server who report that files from it's Pcie drives to be of higher sq than sota cd transports. But this is not first hand experience on my part as I don't own or plan to own a cd player.

Qobuz/Tidal streaming on the Extreme may be a different story, but I can say streaming sounds very good to me-- depending on the source. And here is what, in my opinion, makes the disc/streaming debate somewhat irrelevant to me: the recording, how it was mastered and everything else it's gone through before it's turned into a cd or streamed into your system far outweighs the delivery system be it lp, cd, hard drive or stream. A poorly sourced cd is going to sound like crap and a well sourced stream is going to sound very, very good. I put my energy in "collecting" on Roon well sourced streams and given the vast amount of music available to me, it is not difficult to put together a large collection of excellent streamed music.

The next step for me is to explore the comparison between the streamed files and the same files downloaded or ripped to the Extreme's hard drive.

Given the quality of the server I already have and the likely future tech improvements that will come to that server, spending a lot of money on a high end cd player just doesn't make sense to me.

 

[Mike]

There are quite a few people using the widely acknowledged sota Taiko Extreme server who report that files from it's Pcie drives to be of higher sq than sota cd transports. But this is not first hand experience on my part as I don't own or plan to own a cd player.

Qobuz/Tidal streaming on the Extreme may be a different story, but I can say streaming sounds very good to me-- depending on the source. And here is what, in my opinion, makes the disc/streaming debate somewhat irrelevant to me: the recording, how it was mastered and everything else it's gone through before it's turned into a cd or streamed into your system far outweighs the delivery system be it lp, cd, hard drive or stream. A poorly sourced cd is going to sound like crap and a well sourced stream is going to sound very, very good. I put my energy in "collecting" on Roon well sourced streams and given the vast amount of music available to me, it is not difficult to put together a large collection of excellent streamed music.

The next step for me is to explore the comparison between the streamed files and the same files downloaded or ripped to the Extreme's hard drive.

Given the quality of the server I already have and the likely future tech improvements that will come to that server, spending a lot of money on a high end cd player just doesn't make sense to me.

Not to put too fine a point on things, but we can assume that the streaming services either got their files from ripping the CD’s (maybe early on?) or the digital file from the record company. This is the same digital file used to make the CD, so it’s essentially the same. The difference is that the CD electronic stream isn’t traveling thousands of miles beneath the ocean.

Don’t get my wrong, I like the streaming services and I rip all my music. But if someone is looking for the absolute digital best from an album, my exhaustive findings has been that the CD beats the rip (barely), but definitely beats the stream.

THAT BEING SAID, we are seeing so much high res now being streamed and this puts the CD and a huge disadvantage. In that case, it comes down to high res download Vs stream.


Sent from my iPhone using Tapatalk

 

[Mike]

I was thinking about that, but we recently did a complete overhaul to the power foundation - it is off the charts good, no joke (will post this).

So with that I pulled the dedicated circuit from the original sub panel and added it to the new panel and relocated the digital foundation (mode, etherREGEN, LPS and ADD POWR pieces 45’ closer but still in the other room.

Instantly you could hear things more open, dynamic and amazing sense of presence. After 6 hours I shut things down, paused 4/5 hours and just started up the system.

I have given up saying things can’t get better, this is off the charts after 30 minutes.

The AQVox SE has been pulled from 2 channel duty, and repurposed into home. I simply reversed the no longer needed Ghent 45’ Belkin cable coming out etherREGEN to the AQVox SE.

All these changes by better reconfigurement I ended up with another Audioquest Vodka Ethernet, Stillpoints Ultra SS, and a Ghent DC Oyaide Cable proving less is more - but in all fairness I had a head start with the addition of the panel.

Great to hear and that was my point: less is more. I brought this all up because so many of my, shall we say, experienced in life but less experienced in technology clients get themselves all turned up in knots with what they read online (think: computer audio site, which by the way, do those guys ever listen to music or do they just play with their dingleberries? LOL). But I digress...Many end up calling me thinking they need six boxes or “digital dingleberries” between their router and their digital transport or DAC. I literally have to talk them off the ledge. LOL.


Sent from my iPhone using Tapatalk

 

[grubble]

Not to put too fine a point on things, but we can assume that the streaming services either got their files from ripping the CD’s (maybe early on?) or the digital file from the record company. This is the same digital file used to make the CD, so it’s essentially the same. The difference is that the CD electronic stream isn’t traveling thousands of miles beneath the ocean.

Would be great to know if services like Qobuz or Tidal has co-located servers that are regionally distributed and serving music streams that are semi-close to their audience. It would NOT make sense if I'm streaming them via Qobuz from some data center in France. Services like these aren't designed to serve hi-rez music streams if they aren't set up this way.

 

[Jack]

Qobuz uses AWS content servers in the US.

 

[asindc]

Love or hate ASR as you like, this following quote from the Uptone review is technically correct.

“Perhaps the biggest issue with claims of audio improvement is that your DAC is so far removed from Ethernet that little you can do upstream can impact it. Ethernet has a clock but that is used for communication on the wire. Once a packet (chunk) of data arrives, it is put in memory in the operating system. At that point, it no longer has any timing information much less a clock. It is the responsibility of the application to associate timing with it. And such software notion either works, or doesn't. If it doesn't your music will stop or drop out. None of that timing has any relationship whatsoever with the clock that the DAC eventually uses to play data sent to it. It is the audio application together with the DAC (and or Operating System) which determine timing.”

What really matters is keeping any noise away from your ultra-quiet analog electronics. Lots of ways to do that in renderers, streamers or add-ons. If you use Isochronous Asynchronous USB the only clock that matters is the one in your DAC. If you are using S/PDIF or AES/EBU then the source clock jitter does matter, although many DAC designers have optimized the input circuitry to make this a second level issue. My streamer is battery operated but does use WiFi, so i’ve focused on reducing transmit power to a bare minimum and making the 5 GHz path very high impedance with ferrite chokes.

This would not apply, of course, to DACs with streaming built in, like my Cary DMS-600 or a Lumin T2 or MSB DAC with a renderer module installed.

 

[drunkenspyder]

Different, not better IMO. Upsampled/filters is not native. It can sound different. But when you compare 16/44 ripped CD to the CD, using the same DAC, you can hear the superiority of the CD, but only slight. The big improvement comes with the CD Vs Streaming.

No doubt the ripped CD’s/SACD’s provide amazing convenience. That’s what I do with all mine!

Sent from my iPhone using Tapatalk

I am not sure I agree with the blanket assertion that physical CD is always better than a ripped file into same DAC, though I realize it is conventional wisdom. In my headphone rig, I have both an MSB Select II DAC and the MSB Universal Media Transport. They are connected via MSB's proprietary I2S. The DAC also contains a Network Renderer 2. The DAC has a GigaFoil 4 right behind it [tried an ER, but preferred the GF]. The headphone rack has its own Cisco Catalyst 3560 switch. It is connected over fiber to the main Cisco switch. All files served by a Roon Nucleus+. The tiered [SSD drives for music] Netgear NAS is located in the main switch closet quite some distance from both audio systems. The NAS has a 4-port LAG connection to the main switch, and all the audio network traffic is kept on a separate subnet and with QoS priority. All audio components are powered either by external LPS or a PSA P10 [2 for speaker system] or P5 [2 for headphone rack] regenerator. My primary critical listening headphone is a Stax SR-009S powered by a custom T2 amp. It is resolves fairly well. It's not an A/B/X blind setup, but it's fairly effective for SQ comparison. All SACDs are ripped using either a pre-3.5 Playstation, or with an Oppo 105 or the MSB using iso2DSD. All Redbooks are ripped using dbPowerAmp [and I suppose that begs the question how one is ripping and verifying rips . . ..].

I did similar comparisons in my speaker system when it contained the PS Audio DirectStream and Direct Memory Player. That system also employs several network isolation/noise reduction techniques. I now have a Vivaldi DAC/Upsampler/Clock in that system, and while I have no plans to add the Transport, I may see if I can demo one in my how for just this comparison purpose.

I am not here to argue "bits is bits"; everything upstream can make a difference, audible or not, detectable or not. But I am saying that depending on one's network architecture and steps one might take to optimize it, depending on one's CD transport and how it interfaces with the DAC, depending on how one rips the files, there is no reason that local network delivery cannot sound the same as the physical CD. In fact, it should, if it is delivering the same data into your DAC's buffer. If anything, it should sound slightly better because it doesn't have the disc spin/read mechanism to muck up the works. I would suggest that, if ripped files always sound inferior to physical disks, something in the network playback system is not performing up to snuff. Of course, if the CD transport is meh and/or it is connected to the DAC via an inferior method, then it is more likely network playback will be superior. Just my $.02; YMMV.

 

[mep]

Digital gives people the ability to take a network that should be fairly simple and turn it into a complex kludge of digital dingleberries. It’s beginning to take on the feeling of a competition to see who has the most complex network measured by how many unnecessary digital dingleberries you have in your network.

Maybe we could set some rules for judging the contestants in the Rube Goldberg version of a network used for digital audio:

1. How many conversions do you perform from Ethernet to your DAC? Ethernet to fiber, fiber to coax, etc.

2. How many routers and switches do you have?

3. How many LPS do you have in your network?

 

[drunkenspyder]

Well, “should be” is an interesting standard. How simple should a network be in a large home, with a 5-person family, with multiple listening and watching locations? With multiple computers and printers? I mean, you can make fun of folks with unnecessarily complicated rigs all you want, but until you know what their use case is, it seems a tad presumptuous. Of course one could make a simple audio-only network with one router and just enough ports for a server/streamer and a DAC. Or even forget the network and attach a drive. Yuck.

In my case, except for the fiber between switches, and the GigaFoil at the DAC, I have added no network complexities because of audio. All my network complexities are otherwise a function of the demands of our home. The fine-tuning of the audio portion of the network is all about “protecting it” from all the other stuff in our home.

 

[mep]

Well, “should be” is an interesting standard. How simple should a network be in a large home, with a 5-person family, with multiple listening and watching locations? With multiple computers and printers? I mean, you can make fun of folks with unnecessarily complicated rigs all you want, but until you know what their use case is, it seems a tad presumptuous. Of course one could make a simple audio-only network with one router and just enough ports for a server/streamer and a DAC. Or even forget the network and attach a drive. Yuck.

In my case, except for the fiber between switches, and the GigaFoil at the DAC, I have added no network complexities because of audio. All my network complexities are otherwise a function of the demands of our home. The fine-tuning of the audio portion of the network is all about “protecting it” from all the other stuff in our home.

I’m only addressing the portion of the network used for audio. For me, that starts with the Ethernet jack on my wall next to my digital rack. I don’t care what goes on with the network downstairs.

 

[Boogieman]

Digital gives people the ability to take a network that should be fairly simple and turn it into a complex kludge of digital dingleberries. It’s beginning to take on the feeling of a competition to see who has the most complex network measured by how many unnecessary digital dingleberries you have in your network.

Maybe we could set some rules for judging the contestants in the Rube Goldberg version of a network used for digital audio:

1. How many conversions do you perform from Ethernet to your DAC? Ethernet to fiber, fiber to coax, etc.

2. How many routers and switches do you have?

3. How many LPS do you have in your network?

You are also forgetting that some guys own multiple DACs. And all setups with all the tweaks sound different.

And one of my buddies is really into vinyl and owns 6 tables (including 2 reference level tables), 4 phono stages (1 reference level tube and 1 ref SS), bunch of arms ( I didn't count), and about a dozen cartridges. It gives him an opportunity to tweak and get all different takes on the music.

I guess tweaking is part of the fun

 

[Boogieman]

That leaves out streaming though. That being said, streaming (Tidal, etc.) is like the buffet at The Golden Corral. Plenty to eat, but I wouldn’t call it good. Your ripped CD’s will sound better and a CD playing through the same DAC, better yet again. I would say:

CD = 10/10
Ripped CD = 8-9/10
Streaming = 7/10

A friend owns both MBL reference DAC / transport and MSB. With reference level transports, the gap is wider even more toward spinning disks. CD via transport is a 10, ripped CD is a 7.5, streaming is more like 6.5. But then you are in super rich territory, with some of these transports being like $30K or so.

 

[Puma Cat]

A friend owns both MBL reference DAC / transport and MSB. With reference level transports, the gap is wider even more toward spinning disks. CD via transport is a 10, ripped CD is a 7.5, streaming is more like 6.5. But then you are in super rich territory, with some of these transports being like $30K or so.

If one is spending $30K on a CD player or transport, it damn well better sound good.


Some "systems engineering" basics: virtually every reproduction methodology is system, and as such, is subject to a number of inputs, controls factors, and noise factors influencing the functional response. There is an input, in this case, a digital music file, there are noise factors that impact the quality of the music file transfer and D/A conversion, and there control factors whose role is to provide the functionality and mitigate the impact of noise factors on the response. Shown here is a Parameter Diagram (p-diagram) illustrating this basic principle.

Managing the noise factors by selecting and optimizing the interactions of the control factors can mitigate the impact of noise on a system.

I don't want to go too deeply into a systems engineering rabbit hole there, but the point of this is there are multiple ways to get a digital file to a DAC and output to an amplification component, and as such are no "absolutes" with respect to music reproduction by electronic componentry.

So, in my view, these "comparisons" using different methodologes (CD players, ripped files, streamed files, etc.), is inaccurate and misleading. Moreover, assigning numbers (CD=10/10, ripped 8-9/10, ripped CD a 7.5, etc., etc.) to one methodology is not accurate either because 1) listening to musical reproduction is a subjective experience and 2) because there is almost always more than one way to provide a functional response.

In fact, understanding this principle is the basis for TRIZ. https://en.wikipedia.org/wiki/TRIZ

I have a much simpler set of operating principles: if the system can create a beguiling and engaging listening experience, I'm good. Job done.

 

[drunkenspyder]

I have a much simpler set of operating principles: if the system can create a beguiling and engaging listening experience, I'm good. Job done.

Well put.

 

[Willco]

It's not clear what he's using as a server for his ripped cd's or how his streaming is set up? A comparison only tells us something if you know what's compared.

Fwiw, I've also read comparisons of the Taiko Extreme server PCIE files out performing the highest end Kalista transport.

 

[Alkyogre]

There’s plenty you can read online that I need not add anything. If you’re looking for someone to argue with, trying it yourself in your system should be the prerequisite.

half a year later, I have to admit I made here a mistake, did what you said, listened to it and obtained 2 switches, the Angel Silent Bonn d8, 2 hooked up in serial, powered by the Forester. That power supply is a brand new one and had the first one the shop was able to obtain. According to the vendor it is even better to put 2 switches next to each other, on the second one the network cable to the streamer.

I am sorry, for my reply half a year ago, these 2 new switches did give a big step forward in sound quality, the sound is far better and I am really happy with these.