If USB audio transfer also like data transfer has error correction then it will be the same 1:1 DoP & DSD Native. Unfortunately in USB specification transferring data mode (USB flash drive) & transferring audio mode (USB DAC) isn't the same.
Data mode has error correction to assure all the packets transferred are correct, if any packet has CRC error device (USB flash drive) will request for resend.
Audio is under streaming mode. All data transferred to devices (DAC) cannot do error checking if request resend sound will pause, so under streaming mode will play whatever data received. Because of this reason the cable quality is very important if you want to listen to very high quality audio sound. Another word that doesn't want to miss any bit of sound from the source using an audio grade USB cable is the must.
There are tons of audio USB cables in the market. From affordable pricing to crazy expensive cables but I am very lucky I found a cable that meets my requirements without too many attempts of trying different cables.
My preferred DSD native is because let us assume my current USB cable is still not able to stream the audio data without loss then I only lost the sound quality of missing data in streaming. If using DoP any bit lost might cause the entire encapsulated packet or decapsulate to noise.
Of course for dsd 1x and CD 44.1k 16bits low bit rate might not be easy to notice the difference but when upsample to 705.6k 32bits (or even higher if JRiver added later) then there will be high chances of causing bit loss in transfer. Just a 5cm of USB extension cable also can notice the sound quality dropped in my current setup.
I would like to give a compliment to the JRiver engineers who have changed the audio processing following the upsampling rate not limited to 192k last time. These changes make the sound much better now. I just want to let you know that all the hard work has also had good results in return so keep it up.
Lastly, I have some ideas that can give JRiver a niche and others that are hard to compete with. If JRiver can put it off in the highest quality standard it might be able to revolutionize the audiophile market. For the meantime must improve the audio processing (DSP) modules using current available common hardware like SSE2, SSE4, AVX, AVX2, AVX512, even GPU and upcoming NUP if needed. The first step is to create the Delta Sigma Demodulator (dDSD) DSP.
The multi-process model is better than multithreading model to utilize the processor power like chrome browser (can refer to here for some ideals
https://www.igvita.com/posa/high-performance-networking-in-google-chrome/)
PS: Give me some time to draft it out. I'm currently a bit tired of looking at screens for a long time since after twice brain tumor removal surgeries in Jan & July.