All Collections
High Resolution DSD Listening
Additional reading
The subtle differences between DSD and PCM
The subtle differences between DSD and PCM

This article was written in response to a listener's question

Tom Caulfield avatar
Written by Tom Caulfield
Updated over a week ago

A Question from a Listener

One of the reason to buy from NativeDSD is not the DSD file format. Why? Because nobody has convinced me yet that DSD is superior to PCM. You would be able to do so. It would be great if you could record some sample files in DSD and PCM. There should be no conversion to either format. The recording should be parallel to DSD and PCM. The audio chain should be the same, where applicable, through the whole process. Especially the volume should be the same. Just to show that there is really a difference. My audio chain is pretty simple. I am using a Raspberry PI streaming from a SSD, which has a XFS filesystem, to a Teac UD-301 DAC into my Sennheiser HD-600. I think I would hear a difference if there is one. 

Answer by Native DSD's Tom Caulfield

DSD has several improvements over PCM:

  • DSD has an improved soundstage

  • Better dynamics

  • More detailed without sounding analytical

  • The music had more room to breathe

  • The music is more emotional

  • Less hearing-fatiguing effects

The sound quality difference between DSD and PCM is subtle, and primarily in the very low level information we hear as spaciousness cues. But they are there, and easily demonstrable with a level matched low latency A/B switching test. This unfortunately is difficult to implement without a Digital Audio Workstation. My acoustic memory is about two or three seconds long, as are most peoples, so the time required to switch formats and sources in consumer gear is too long for other than a subjective preference judgment.

It would be great to implement a simultaneous "no conversion" pair of native DSD and native PCM recordings, but that is not possible. There's currently no manufactured non converting native PCM A/D converter available. The last one to be manufactured is the Pacific Microsonics PM-2, about 14 years ago, and the only label I'm aware of currently using it is Reference Recordings Keith Johnson recorded series. He's also the co-inventor of that ADC.  

All current ADC's use a Sigma-Delta Modulator as the digitising front end element, producing a one or multi-bit Pulse Density Modulation (PDM) bit stream, of which DSD is the one bit variant. PCM recordings are produced by an additional "in the ADC box" PDM to PCM converter, one for each channel.

So the quest for comparison tracks of native DSD and native PCM is not achievable. We do however offer free examples of unedited un-post processed DSD session files with additional computer processed PCM (DXD-352.8KHz PCM) conversions in our NativeDSD member's area for just such testing and experimenting.

This is a great interest; the combination of music's beauty and emotional gifts, and the technical quest of ever improving its home enjoyment.

Did this answer your question?