Tag Archives: PCM

How dense are you?

Yesterday I surprised a few of you when we learned that the output of just about every modern DAC with 24 bit resolution is actually DSD, or Pulse Density Modulation, also commonly known as 1-bit. We understand that classic ladder type DACS reach a brick wall at about 16 bits because the accuracy of parts needed to extended beyond that range simply do not exist.

I would ask if any of you remember back to the beginning of this series when I told you that PCM (which is what CD’s work with) is a code decipherable only to a machine that understands that code – and DSD is not a code at all and your analog stereo system can play it back directly. That’s a big difference.

We learned that PCM measures the musical signal with a series of snapshots called samples and memorializes each sample as a multi-bit word. We reconstruct the samples by reversing the process and creating a stepped output which is identical to the input once we smooth out the little steps.

1-bit audio is very, very different. Today we’re going to simply give you an overview of how it works, which is a simple process, and then we’ll get into the nitty gritty of the process which can be a little disturbing to say the least. But let’s save the disturbing stuff for later once we’re all on the same page.

1-bit audio is simple to understand in concept. There are no samples, there are no words, there is no code. Instead there is a continuous streaming “train” of single identical bits that are either on or off. The more bits that are on, the higher the eventual output voltage becomes. The more bits that are off, the lower the eventual output voltage. We refer to this type of scheme as Pulse Density Modulation because when you have a greater number of on bits it appears as more densely populated. Here’s a picture that will help you visualize a 1-bit system.

Pulse density modulation 2 periods How dense are you?

Note the blue areas are on and the white areas are off. Also note the periodicity between single bits is identical. The red sine wave overlaid on this image shows the results of more bits or fewer bits. Where there are no on bits (all white) the sine wave is at its lowest point – lots of on bits and it’s at its highest point.

The speed of the bits is 64 times the sample rate of a CD and some DSD schemes run at 128 times faster than a CD.

Here’s the interesting part of this: if you take a DSD stream and run it through a simple analog lowpass filter to smooth out the on/off transitions, you get music! This is amazing considering that if you do the same with PCM you get only noise.

DSD is a lot closer to analog than PCM ever thought to be.

Paul McGowan – PS Audio, Intl.


Today’s post marks the start of yet another one of our “discovery” series where we try and get a little learning under our belts. These series seem to be ok with most people and they are fun for me to write, so what they heck. Let’s go for it.

I think there’s a lot of talk today about DSD and at PS Audio we get a lot of questions about it – especially after TAS Editor Robert Harley made a comment in his RMAF show report that DSD was one of the most interesting trends in DAC’s today – this despite the fact there’s little DSD files available to play on such a DAC.

What is DSD and how does it differ from a standard CD? Is DSD available only in Sony’s version called SACD? What is the fate of SACD? And, did you know that nearly every modern DAC today has a DSD output?

I think these are great questions and ones we’ll set out to tackle.

As always we’ll start our series with a bit of history. The two types of digital audio we’ll be discussing are called PCM and PDM (DSD is the popularized name for PDM).

Pulse Code Modulation (PCM) and Pulse Density Modulation (PDM) have been around as ways of encoding analog information into a digital equivalent far longer than our introduction to digital audio through the CD in 1982. In fact, PCM dates back to the 1920′s, PDM to the 1950′s.

Both are means of representing an analog signal in digital terms that can be understood by a non-anlog device such as a computer. In the same way that an ancient Abacus uses 1′s and 0′s to represent complex number models, a digital audio signal breaks what we understand as a continuous analog movement of electricity, into small individual discrete “quanta” or units that represent the analog signal.


One of the major differences between analog and digital music is that the shape and form of one is critical; and mostly unimportant in the other.

To imagine the differences let’s use a simple visual model. Picture a microphone connected to a loudspeaker (we’ll ignore the needed amplification for this example). When you speak into the microphone an electrical signal is generated that is an exact replica of the moving element in the microphone. The signal then moves the loudspeaker cone back and fourth pressurizing the air in the room and you hear sound. If you change anything in the path between the microphone and the speaker, such as the shape of the signal, what comes out of the loudspeaker will be different than what went into the microphone. That’s an analog model of sound.

Using a similar example, let’s now imagine a microphone with a digital output and a loudspeaker with a digital input. The digital bits streaming between the microphone and the speaker change content according to what is spoken in the microphone – but their shape remains the same.

Digital audio has but two states, on or off, and the shape and size of those “bits” will not affect the outcome of the loudspeaker. The number and position of those bits will change the outcome but not the actual shape.

The advantage of digital over analog, with respect to recording the information, should be immediately obvious. In one case the quality of the recording doesn’t matter and in the second case it matters greatly. Certainly an over simplification but that’s ok, we’ve a long wau to go.

Tomorrow we dig deeper.

Paul Mcgowan – PS Audio, Inc.