Native DSD versus DoP


How big of a deal should it be to ensure one has true native DSD capabilities in their digital front end?  Case example: I have a Lumin U2 Mini and a Denafrips Pontus 15th (out to McIntosh C53 and MC312, B&W 702 S3, REL t/7x).  I am thrilled with the SQ and I am really enjoying downloaded DSD playback.  I am just curious - would anyone suggest that I can match or beat this SQ with a non-native-DSD-capable DAC?  I mean, for a comparable price?

I play SACDs on an MCT500 to the DA2 in the C53.  Would anyone suggest I can get better SQ running SPDIF to a separate DAC?

P.S. If you think my speakers are a weak link, that's fine.  Assume I upgrade them, does that change your answer?

mattsca

Just like @mikelavigne I rarely choose DSD from the library unless there’s a specific album I’d like to hear in DSD or when I like a certain recording in high-res and want to compare it to its DSD counterpart. Another case is when Roon shuffles in a DSD track from the library when it spins Roon radio.
As to DACs, I owned a Bricasti M3 that featured Bricasti’s proprietary DSD 1 bit DAC. I could choose between that or DSD PCM using the conventional path with Analog Devices DAC chip. Bricasti accepted native DSD. I preferred its proprietary DSD DAC in that case. 
I now have Meitner MA3i that doesn’t accept native DSD. It’s an FPGA DAC that up-converts everything to DSD regardless of incoming bit rate. Meitner DSD over PCM sounds amazing playing local DSD files. So as I said earlier, don’t sweat it. It’s fine if the DAC supports DoP only. 

I agree with the above, Also 16/44 pcm is plenty fine, who'd ever thought! I remember when the common claim was 16/44 could never provide enough resolution and pcm inferior, how wrong.These days I pay no attention to whether recording 16/44 or upsampled, oversampled, whatever.

The DAC in my main system will not do DSD. I came to wonder if that was a big deal. I have a Ayre DAC on my headphone system so I made a quick survey of how much DSD material was available and compared some DSD recordings. The difference was small, as were the number of recordings. I decided it was of no importance for me. 

I find that the difference in ones system components make enormous differences... say the range of 100. The differences in interconnects and power cords make differences in a range of 10. The difference in recording quality a range of 20. Then improvements because of DSD a range of 0 to 2. Since I am sure no one has optimized your components closer than +/- 10 the over difference in recordings is another 20... the small DSD difference in the big picture is so tiny as to be obscured. 

I'm not sure how to quantify the differences I have heard streaming 24/192 vs. playing an SACD in DSD.  They are subtle.  One difference I noticed was in playing "The Priest" from Joni Mitchell's Ladies of the Canyon, her guitar was set further back from my speaker when playing the SACD vs. streaming the 24/192 file from Qobuz.  Perhaps this difference would not exist with a streamer that's better than my Node 130 with Teddy Pardo LPS.

Comparing other albums, high-res PCM streaming has been as good or better than my SACDs.  But I should note that I'm using a Cambridge CXU universal disc player, not a dedicated SACD player, so maybe that's comparable in SQ to my Node streamer.

fyi, if it helps. I have some similar hardware.  Optimized Ethernet, Volumio Rivo+, McIntosh MCT500, McIntosh D150, McIntosh C2300, two Mc275 amps.  I have ripped all my SACD to DSD - dsf files.  I put them on a microsd card and run them through the Volumio Rivo+ to the USB in the D150.  In addition, I can play the same SACD discs throught the DIN cable to the D150.  Also, I have the same DSD - dsf files on a usb stick that I can play through the MCT500 USB input, which goes through the DIN cable to the D150.  BTW, if you have not tried it, the user interface is easy and fast on the MCT500 using the USB stick.  For the D150, the Rivo+ encapsulates the dsf files in DoP to deliver the DSD to the D150 through USB.  The D150 shows DSD64 on the display when playing it.  My experience is that if the DAC shows DSD64 on the display and not a PCM number, then, you are getting the best quality DSD playback.  I would be hard pressed to tell the difference in a blind test between all these 3 ways to deliver DSD to my Wilson W/P 8 speakers.  Now, if I have the Rivo+ convert the dsf files to PCM and the D150 shows PCM frequency for those files, then, I believe I can tell the difference, although, I have not done blind tests to verify it. The Rivo+ also supports I2s, but, I have not tried that since the D150 does not support it.   So, bottom line, for me, there is a difference between "encapsulating DSD in DoP for delivery through USB" and actually converting DSD to PCM.  The former being the same as "Native DSD" and the latter being somewhat undesirable - but, needs to be verfied in your system.

PS.  One way to test your Pontus and DA2 with DSD is to rip you SACD to dsf files, or just buy/download dsf files from online, and feed them to each DAC for hearing tests.  For the DA2, you can use the MCT500 USB stick interface to send them through the DIN cable.  For the Pontus, if your streamer supports usb stick or microsd, or hard drive,  you can send it through that using the USB connection.