Why a streamer with a UnitiServ?

Noogle posted:

@Allan

"Bits are bits" is a very scary wormhole that this forum gets sucked into from time to time.  As regards digital audio, all that matters are data and timing .  For example, CDs contain 16-bit samples which were recorded at 44,100 times a second.  As long as the DAC input gets the right data at the right time that's the best you can do in the digital domain.  In practice though there is noise (jitter) in the DAC input clock signal which means the samples aren't 1/44100 seconds apart so the DAC's output isn't true to the original recording

Most modern DACs are asynchronous - they use their own internally-generated very low jitter clock signal which means that the jitter from the source device is irrelevant. 

Many users on this forum have their own theories as to why different digital sources sound different with a particular DAC or streamer.  Some of these are mentioned above and include RFI coupling of jitter from the source device, noise coupling through the mains, the fact that digital data signals aren't true square waves etc. etc.  I think you'll find as many of these theories as you care to look for.  Whether they are based on experiments or stand up to close examination is another matter.

Plus, RF noise - nothing to do with jitter - can wreak havoc to SQ in the DAC, so DACs RF filtering of the input is a must, or prevention of RF pickup all the way from the music store, renderer, cables etc. To be fully effective includes RF isolation of any input: USB, ethernet etc, and as not all DACs have as effective RF isolation as others, this is a very real cause of differences with different sources and different DACs with different cables, and with different ancillary equipment around that may create RFI.

Innocent Bystander posted:

Plus, RF noise - nothing to do with jitter - can wreak havoc to SQ in the DAC, so DACs RF filtering of the input is a must, or prevention of RF pickup all the way from the music store, renderer, cables etc. To be fully effective includes RF isolation of any input: USB, ethernet etc, and as not all DACs have as effective RF isolation as others, this is a very real cause of differences with different sources and different DACs with different cables, and with different ancillary equipment around that may create RFI.

I am tracking AOIP and people over on Head-fi are lauding it as the best thing since sliced bread. Along with Alan in this thread I have been trying to unpick why differing encapsulation of the audio data might effect sound quality, i.e. DLNA/uPNP vs SMB vs Dante vs AES67 etc, and USB vs AES vs Toslink vs coax. This has me thinking whether it is something simpler and less software based, is the galvanic isolation and twisted pair RF rejection simply better at delivering the audio data?

Currently many AOIP users are using Rednet 3 / D16 to handle the data and pass it out via spdif to their DAC. Some DACs are now being produced which can handle data via an RJ45, although I think they rely on proprietory protocols, such as Dante.

I found that the SQ of my Oppo 105D browsing my NAS direct is very good, not as good as the microRendu / Mutec MC-3+ USB ....but then that is comparing a swiss army knife to a scalpel. Perhaps the 105D is leveraging the TP advantage without the extra protocol hop.

This leads me to reflect on Naim building their renderers into their pre-amps / DACs, something that I hate. Perhaps they have found that by getting the files direct into the amplifier they gain a technical advantage over using a form of breakout box and then moving it using an additional protocol. This points at the 272 as a product with a definite technical edge over using a US / HDX / NS01 as a renderer, and of course the streamers if they don't use their internal DAC.

M