Sneaking in the back door

Join Our Community Subscribe to Paul's Posts

With the growing interest in DSD, which is the main format of SACD, I thought it might be helpful to explain DoP. Yeah, I know, too many TLA’s! (Three Letter Acronyms). DoP stands for DSD over PCM and was invented by a group of high end audio manufacturers doing their best to get great sound to all the rest of us. PCM is the standard format for digital audio like that found in CD’s.

So here’s the idea. Because the PCM format is everywhere, a format understood by all digital audio players, computers and DACS, it is the logical candidate for a format. As you can imagine, getting all the computer manufacturers and operating system manufacturers together on something as little used as DSD would be a huge undertaking and one that probably would never get off the ground. But if it were possible to leverage an existing format, like PCM, then it’d be a much easier task. And that’s just what these folks did.

The PCM format uses a series of frames, or snapshots, for the data. Each frame is a separate grouping of audio containing a left and right sample. In a CD we get samples of the music and each sample is taken 44,000 times per second, once for the left channel and again for the right channel (so the process happens twice as fast in order to get two 44K samples in the same time). Now, mentally take those two samples (left and right) and put them into a package. That’s the frame we are talking about in PCM. Each frame contains one left and one right sample along with some extra information that explains to the equipment what to expect.

The DSD format, on the other hand, does not have any frames; instead it is a steady stream of bits with more or less density according the the music.

So here’s the clever thing the designers of DoP did. They simply took the DSD stream apart into discrete 16 bit groups; every 16 bit block was separated from the stream. They then combined a 16 bit right group together with a 16 bit left group, added the identifier information and dressed it up so computers and DACS would think it’s PCM. As far as any computers or USB streams are concerned, it is indistinguishable from PCM so it is passed onto the DAC via USB, S/PDIF or AES/EBU. Very clever indeed.

To match the speed of single and double DSD, they use 176.4kHz and 352.8kHz PCM rates respectively.

Once it gets to the DAC, it is then converted back to its original stream by simply connecting all the bits back together. There’s no data manipulation, the bits are unscathed and, if your DAC can play DSD you’re good to go. If the DAC does not support DSD, you get silence.