As @plaudiofeel said. The subject was not about integrals vs. separates.
It was about system synergy the industry fails on:
2 main areas:
- Desire to avoid double functions, ie. one DAC or one phono stage in the system is enough
- Lack of standard interconnects
I fully understand @stevensegal. I like high quality integrated units for which I apply the above criteria equally as others who prefer separates.
At some point we need to connect a source to an integrated pre amp or / digital hub.
For wired analog the balanced XLR connections are undoubtedly superior to single ended.
For digital it is a problem. Today’s high bit rates and high frequencies render Toslink, coax and EAS EBU inferior to connections that transport the data separately from the clock and other auxiliary data, or interfaces that offer a bandwidth that offers sufficient headroom to ensure that every bit received well and perfectly matched with the clock.
Toslink: regardless it’s limited bandwidth, a keeper as it allows me to connect less critical sources like TV’s, set top boxes and game consoles to my stereo while they remain completely galvanically separated. Perhaps one day replaced by a better optical standard, but considering its purpose more than good enough.
Coax, was rendered obsolete 15 years ago for digital signals.
USB-2: Does not provide sufficient bandwidth to allow proper handshakes and checksums of the arrived high bit rate audio data packages like any computer data in order to play it fluently.
Ethernet: Provides enough bandwidth to do what USB-2 is not capable of, there are standard galvanically separated interfaces available. A serious candidate to replace USB-2.
HDMI: Fast enough to transport video data for even 4k televisions which is a multitude of high bit rate high frequency audio data, offers enough physical links to separate the clock and auxiliary data from the audio data which saves time consuming steps to at the source to combine it all in a timely matter and at the receiver side to separate it all in a timely matter. Something that could also be achieved over Ethernet cables including handshakes and checksums.
USB-C, the current standard for data transmission does everything Ethernet and HDMI can do together, but perhaps a bit to short to place the sources away from the hub (whether an integral (pre) amplifier or separate DAC).
Therefore either I2S over HDMI or Ethernet are the superior single cable interlinks when it comes to high res audio sound quality.
Wireless:
To my opinion not developed to where it should be.
At the moment WISA is the state of the art standard as it offers 24bit 96kHz PCM only.
Whilst I think that it allows very good sound quality, it does not offer other audio formats to be transmitted natively YET. So DSD or higher resolution formats need processing which defeats the purpose of those formats. I am in PS Audio’s camp in that the conversion from PCM to DSD can be done with less loss in SQ than converting DSD to PCM when converting to analog.
If and when wireless becomes audio file format independent it becomes the most serious alternative to wired digital connections. With one catch, if I don’t want all that radiated data, it can not replace the analog XLR connection. Never.
Does it really make a difference? Experts are still fighting over that. It is a subject beyond my knowledge. Practically however, I have and will keep mobile devices in the living room in which I want to live and listen to music. So there is always some level of radiation, even radiated by the high end equipment itself.