I (inadvertently) made a comparison.
In a secondary room I have a fairly old Bel Canto DAC, with very limited USB capability because in those days USB was just emerging.
I feed it with an even older Meridian MS600, and have to use coax SPDIF.
I tested with the CCA optical, and it was noticeably worse.
But then I tried the CCA with my modern Chord Hugo 2, optical, compared with a MicroRendu over USB and found no differences.
My theory is that the old DAC is more sensitive to jitter than the new one, and the MS600 is better in that respect than the CCA.
Btw: why does jitter matter? When they invented digital music and SPDIF connection, they made the bad engineering decision to let the source control the timing. Probably influenced by turntables, where you invest in fancy mechanical engineering for stability. So people bought expensive CD players with heavy platter mechanisms. You have a digital technology, sensitive to timing at the picosecond level, and you try to solve that with a mechanical device? Stupid. Everything should be asynch, like modern USB and networks, and the DAC is responsible for its own timing. Anyway, in a classical DAC design, timing errors translate directly to amplitude errors (distortion) because they work with ramped signals — I remember reading this some twenty years ago, don’t remember exactly how.
How doesn’t matter. Jitter matters for SPDIF (including optical), more or less depending on the DAC design. Not for USB or network because they have no jitter. They cannot, because they are asynchronous. Wikipedia: jitter is the deviation from true periodicity of a presumably periodic signal. The network is not presumed periodic.
My opinion is, in this age, we can effectively forget about jitter, especially over a 1m interconnect. It’s hardly measurable, let alone auduable. The slight differences are heard in the analogue domain, not digital.
Thanks @anon97951896 for directing me to the thread about the dropouts. It’s clearly a problem with high-res files only, and perhaps improved by adding a UGreen ethernet adapter, although that doesn’t seem to have worked for everybody (see Chromecast Audio dropouts), because a buggy firmware update looks to be the root cause.
I purchased a Chromecast while it was on sale before Christmas. I figured because I was simply plugging it in through my Anthem MRX710 I wouldn’t hear a difference between streamers.
The sound was a bit thin with the chromecast, so as an experiment I purchased a demo/close out bluesoind mode 2 from a local dealer and connected it via coax to the Anthem.
Now, I like to play a game with unsuspecting guests who are not “into” audio. I invite them to sit in the living room, and while I make a drink or get a snack for them I ask them to choose a song they like on my tablet. I wait until the.misoc starts to go to the kitchen. I return with drink and/or snack. After I hand it to them I take the tablet and switch the endpoint from the chromecast to the bluesound.
Without exception and without solicitation they will ask something like, “Whoa - what did you just do?”
I have the Beolab90 speakers from B&O and have tried the following:
Roon through CCA connected with optical
Roon through Bluesound Node 2i with optical
Oppo 105D directly to BL90s with optical
Streaming Tidal (either directly or via Roon). I have to say that the difference between the CCA and Bluesound was marginal. I like to use Carla Bruni for these tests as her voice just shines on the BL90s, and all the beautiful facets sounded 99% the same. I do not have an optimal listening setup (speaker placement) these days, but really - it was odd to see this little thing feeding a set of $80K speakers so brilliantly.
The Oppo105 has become my main source of music, and to me it is superior to the two options above. If only the UI/UX of the Oppo was a bit better … their app is such a pain.