Yep, so if you do the math, you find that half the bits in any DoP protocol are devoted to administration, leaving the other half for music. So you need double the bit rate required for native DSD. Thus:
705.6k PCM will support DSD512 for native DSD. However, for purposes of DoP only 352.8 PCM can be used for music, and so only DSD256 is supported.
Similarly, for any given DSD resolution, you need double the PCM resolution for DoP as for native DSD.
My audio system does not go through DoP but uses DSD native output by ASIO,this DAC supports native DSD1024 decoding, but audirvana only supports native output up to DSD256.
What you are saying seems only congruent with the alternative DoP 1.1 protocol for transmission schemes that don’t support a compatible PCM carrier sample-rate of the primary DoP 1.1 encoding scheme. … Otherwise it is erroneous:
Keeping it in the DoP 1.1 context and not raw native DSDxxx transmission… 705.6kHz only seems congruent with the alternative DoP 1.1 encoding scheme where the DAC does not support the required PCM carrier sample-rate…
My DAC will play 5.6Hz DSD (native or via DoP 1.1)… It will not play DSD256 via DoP 1.1… Your statements regarding the DoP PCM carrier sample-rate do not follow the DoP 1.1 Standard.
5.6MHz bit rate, you mean. That’s DSD128. And if you switched to Windows your DAC would immediately be able to play 11.2MHz bit rate or DSD256. That’s because DoP devotes half the bits (5.6MHz in your case) to things other than music. It says so in the initial reference you cited.
My DAC, which supports double your PCM rate (705.6/768k), does DSD256 over DoP on MacOS (double your DSD rate), and DSD512 with native DSD on Windows or Linux (doubled again).
The fact that your DAC does DSD128 over DoP while being able to do DSD1024 with native DSD using ASIO suggests to me that there may be something custom about the ASIO Windows driver.
If you have a direct connection to a Windows computer, see whether there is a manufacturer firmware or driver update available. If you are using UPnP/DLNA from an endpoint rather than a direct connection, then your maximum rate is controlled by that endpoint (likely Linux/ALSA) and it makes perfect sense that your DAC would be limited to double the DoP rate in that situation.
It is also the case that you will not necessarily get better sound quality (lower distortion) at DSD1024 than at DSD512 or DSD256. Unless you have seen measurements at each rate for specific filters and modulators, I would not automatically assume DSD1024 is better.
@Jud …Yes… This follows the DoP 1.1 protocol, and it is what I have been describing in relation to the DoP 1.1 Standard… However you made a confusing statement (maybe a translation problem) here below, if you were stating this in the context of DoP 1.1…
I am presuming what you meant in this statement “…will allow DSD512 on Windows and Linux,…” pertains to, raw DSD signals… not DoP 1.1 signals being delivered by Windows and Linux on a PC platform… I already understand that Windows facilitates the transfer of raw DSD data-streams.
This is not true… If Audirvana supports up to DSD256 via DoP 1.1 then this is the limitation, theoretically, the macOS could support DoP 1.1 sample-rates to 2822.4kHz (DSD1024) however, the real-world application of DSD1024 does not warrant the computational overhead… nor does DSD512 warrant the computational overhead.
DSD over PCM (DoP) requires a supporting CODEC to extract the DSDxxx signal from the PCM container/carrier… If the DAC designers determine that DoP support for DSD128 is sufficient, this is their prerogative and justified by whatever criteria they are using to make this decision.