Unfortunately it is a bit more complicated than that... and even the ways in which it can be complicated can be complicated.
If you're doing it as a file share, where the player is simply accessing the file on a server, the answer is pretty simple.
The sound quality is entirely up to the player and about the only effect the upstream part could have would be if there were actual dropouts due to things like network traffic.
(And it would be up to the device at the player end to use enough buffering to rule that out.)
I don't use Roon... mostly, to be honest, because I just don't care about most of the cool features it has, and I don't especially favor "comprehensive music managers".
But, from what I've heard, it does an extremely good job of showing you what it's doing...
So, for example, you can see the original file format, the final format, and any conversions that it's decided to do.
(And, by implication, this means that RAAT is quite capable of delivering a proper bit-perfect audio signal.)
(From what I've heard, if I actually had a need to distribute music to multiple rooms, I would probably use Roon... even if I didn't use most of its features.)
I don't know an awful lot about the internal workings of DLNA... but that's mostly because I know enough to know that I don't trust it.
From what I've seen the main priority with DLNA is "delivering a nice solid audio signal over the widest possible variety of connections and devices".
The catch seems to be that, in order to do this, it is willing to resort to lossy formats, or lower bit rates, if necessary to achieve that goal... and probably not warn you when it does.
(Much as, with Youtube, the priority is that you get to watch your video without interruptions, and the quality you get is further down the list.)
From what little I know, you CAN control some of this, and SOME DLNA software and endpoints allow you to do so...
However, in the end, it always seems to be a bit vague about what you're actually getting...
Different DLNA servers and endpoints seem to have very different capabilities and options...
Which strongly suggests that things could get a lot more complicated with different combinations of them...
And there seems to be little or no reporting that you can use to confirm exactly what's going on at any given moment.
(Again that seems to vary rather widely.)
As you know my background is in computers...
And, when I copy a file over the network, or onto a USB stick or hard drive, I don't have to worry about whether it's going to be the same file at the other end.
And, to be quite blunt, I expect that same assurance from digital audio.
(And, with DLNA, I seem to get something more like "trust us... it's playing".)
To be quite honest, there MAY be SPECIFIC DLNA implementations that don't have these weaknesses, but to me it isn't worth the bother to try and ferret them out.
I haven't personally used Dante, but it is widely used in pro audio, and has a reputation for being both robust and trustworthy.
However, it is also complex, tends to require somewhat expensive hardware, and in the end it is just a transport mechanism.
It's a way to connect things like audio sources, controls, and powered speakers together over a network...
And, from what I can tell, its options for controlling signal distribution are fantastic...
But "it's something that other things run over"...
It is not, in and of itself, a turn-key solution for playing music from your collection in different rooms... and it doesn't seem to be geared to "home users".
Although there is at least one low cost Dante player that I've seen, and Dante is a sort of standard, to me it "feels" more like Control4 than it "feels" like Roon.
(I'm not seeing too many music players that offer "Dante" as an output device choice.)
I went on (at great length) about your last question in another post (before I read this post)... however...
With USB,
MOST DACs have an asynchronous USB input, so the signal is being clocked by the DAC.
This makes the quality of the signal being provided by the source must less important.
(You absolutely want the source to be providing bit-perfect data but the DAC should be taking care of most of the other details.)
Toslink is essentially a S/PDIF signal... so the DAC is using the clocking from the source... after it makes its way through the cable... and most DACs do
NOT re-clock their S/PDIF input.
Toslink transmitters, receivers, and cables, all have a reputation for having very high levels of jitter... although I've rarely seen specs there.
I suspect that some may actually perform just fine, but who knows, and, since they aren't being re-clocked by most DACs, that results in a huge unknown.
And you've also got to assume that this will vary depending on the length of the cable, and even the
TYPE of cable, and maybe even how well it's plugged in, or whether the window on the end is clean.
(You can spend a few dollars more to get a glass Toslink fiber rather than plastic... they're rated for longer distances and probably have less jitter... but, again, who knows.)
Toslink has one really great benefit: it is electrically totally isolated from the source (so ZERO ground or power voltage noise and zero potential for ground loops).
From my experience, I've heard cases where the Toslink and Coax inputs sounded different, and where they didn't (I can't remember any where the Toslink input sounded different and
better though).
(It's also worth mentioning that the optical connection on most devices runs through the S/PDIF transmitter and receiver... with the optical conversion circuitry an extra step at both ends.)
Now, compared to Toslink, Coax S/PDIF has some benefits, and some drawbacks.
The benefit is that it is known for lower jitter... (again largely anecdotal, but it is "a more direct connection".)
However, while Coax cables are also more "trustworthy" in general than Toslink, the signal is also more sensitive to cable length.
Therefore, it is always recommended to keep Coax connections as short as possible... and lengths over a few feet can be problematic.
(You need to worry more about using good cables, with good connectors, and the possibility of "compatibility issues".
In this case that means that, with longer cables, and with higher capacitance cables, some sources have trouble delivering enough signal to please some receivers.)
The other thing to consider is that very few streamers
OFFER a "USB DAC output" while many offer Coax and Toslink.
(So your choices are often limited.)
Here's my personal order of preference...
1. If the USB option is available I prefer it for several reasons (which can be summed up as "I'm trusting the DAC more and the source less").
2. If there is no USB option my next choice would be Coax (by reputation and experience I still prefer it to Toslink).
3. If there is no Coax option, or if I'm going to be forced to run a long cable, I'll use Toslink (and I'll definitely spend the extra $5 or $10 for a glass Toslink cable).
(Note that the glass Toslink cables also tend to be thinner and more flexible... but are also claimed to be easier to damage.)
Thanks
KeithL -
So, I'll need a streamer that will convert Ethernet input packets to my choice of USB / TOSLINK / other outputs. Now for an interesting question - Does it matter whether my streamer uses DLNA, RAAT, or Dante provided that my source can output one that the streamer is accepting on its input?
I wouldn't think so, but I could be wrong. I'm currently using Roon to feed a (DLNA?) Apple TV as the endpoint. I'm virtually certain that Apple doesn't bother to support RAAT, yet the system seems to work fine. To phrase my question in another way, is there any audible benefit to having a Roon-RAAT capable endpoint?
And yet another follow-up question, please: Should I expect an audible improvement from a streamer that will feed the DAC a USB feed rather than an optical one? There may be a USB standard, but (assuming that the bit depth and frequency are not outside either the streamer or the DAC's capabilities) is USB technically superior to TOSLINK optical?
Thanks - Glenn
