Just a comment, the Core I am sure is a great bit of kit… but there is no ‘reclocking’ of the ripped CDs, the media is simply serialised and clocked so it can be sent via SPDIF… prior to this the data is not ‘clocked’
The point about SPDIF transformer is for galvanic isolation, most if not all quality SPDIF interfaces use galvanic isolation transformers… further the transformer can help maintain the 75 ohm CI of the SPDIF transmission line. Be wary of poor or cheap SPDIF interfaces with no galvanic transformers… you may well have trouble with Earth loops in such circumstances and errors…
Jitter is a random distribution of variation around typically a clocking frequency . Galvanic transformers as used in SPDIF interfaces don’t inherently add jitter. (But can add deterministic clock phase modulation)
Naim as far as I am aware have always used SPDIF galvanic isolation transformers, as that is usually the recommended way of providing a quality and reliable SPDIF termination.
However any jitter in the SPDIF transport frame protocol clocks is not related to any jitter in the sampled media (just as with Ethernet or USB) … other than through electromagnetic coupling electronic noise that can couple from one function into another…
Lots of technical explanations here…but…the first principle should be this: does ‘high res’ sound better/different to 16/44. For me, ‘high res’ does sound better…no amount of technical explanation to the contrary will change the empirical evidence. Just in case anyone should suggest it…no, I am not fooling myself, I am very familiar with the concepts of observer and expectation bias.
HA, the Core runs optimally in UPNP mode, connected to a switch and used as a server/ nas for your Nd555. With a digital cable, your Nd555 is only used as a Dac. You loose all the streaming part of it, which is more advanced vs the streaming part of the Core.
What I see is that you come from nowhere, and seems to know all better than everyone.
On different topics you give some definitive advises and you are new on that forum.
The last months I have helped out re-mastering a big library of recordings. When using professional equipment I have the same experience as you. With high quality equipment there is really no gain going above 44/24 or 48/24. For distribution and listening at even loud domestic levels 44/16 can sound very, very good.
Any DAC will only replay 20-21 bit due limitations in the analog domain.
A 16-bit CD with noise-shaped dither might give you upwards of 20-bits in the midrange. Depends on the noise-shape, I usually select dither by listening.
On the other end of the scale I sometimes master through a 2-track Studer B67. When carefully recapped these reel-2-reels give a lovely dark tone and the analog tape noise mask the low-level digital crap of early F1 mastera.
What you have missed is the definition of the original master… if greater than 44.1/16/2 which many modern masters are, with 48/24 and 96/24 being popular for distribution format reasons such as for video codecs, then clearly 44.1/16/2 will be inferior because of digital decimation and the loss of information. Noise shaping can’t create definition on LPCM … that would be alchemy.
To what extent you notice this inferiority is equipment dependent on course, not least the reconstruction filter technology and converter linearity and resolution. It’s a myth to relate sample word size simply to signal overall dynamic range… … it has a big part in the accuracy of the reconstruction filter too and reconstructed signal accuracy. One has to also think of the replay chain effects… the AES had an interesting paper on this aspect that many seem to forget.
I find these days 44.1/16/2 is somewhat inferior to higher resolution formats, with 48/24 and 96/24 often sounding optimal for me… sure 44.1/16/2 can sound good if it was originally digitally mastered as that , but it can sound so much better if mastered as typically is often the case now at higher definitions since CD is not the ubiquitous media format it once was. I also think the 48 or 96 down sample to 44.1 is not an optimum conversion in terms of reducing digital noise/artefacts.
The high definition option of the Qobuz streaming service is a good example of where you can compare distribution masters and their decimated legacy 44.1/16 cousins, as well as see which new titles are only released as 44.1/16/2 which seems to be declining.
Looking at the 10 recent album releases on Qobuz streaming this morning, 7 of them are released and distributed at greater than 44.1/16.
My experience mirrors Simon’s. On good equipment, 44/16 can sound very good but, the best requires all things to be right and the resolution to be higher and the reconstruction to be superb.