Not So Bit Perfect About DSD vs. PCM? Part IV

Here’s the continuation of the analysis of BitPerfect’s challenge to my assertion that 96 kHz/24-bit audio provides better fidelity than DSD.

BitPerfect: “Let me tell you how DSD Master approaches this in our DSD-to-PCM conversions. We know that, for 24/176.4 PCM conversions for example, we need only concern ourselves in a strict sense with that portion of the ultrasonic noise above 88.2kHz. It needs to be filtered out by at least 144dB or we will get aliasing. However, the steepness of the filter and its phase response are governed by the filter’s cut-off frequency. For the filters we use, the phase response remains pretty much linear up to about 80% of this frequency. Therefore we have some design freedom to push this frequency out as far as we want, and we choose to place it at a high enough frequency that the phase response remains quasi-linear across the entire audio band. Of course, the further we push it out, the more of the ultrasonic noise is allowed to remain in the encoded PCM data.”

John Siau The key phrase is “quasi-linear”. The DSD post filter can be built to achieve reasonable phase response in the audio band, but it is not as accurate as the phase response provided by an oversampled PCM system. Also it is important to note that the DSD post filter is implemented in the consumer’s D/A converter. The quality of this filter will vary significantly from product to product. The D/A converters in SACD and Blu-ray players may not have very good phase response. Unfortunately, the DSD consumer is often forced to use these devices, as there is no standard means of delivering the DSD stream from a disk to an outboard D/A converter. This forced use of internal converters is a major problem when delivering music over DSD.

Dr. AIX In another blog post on Richard’s site (click here), he states that, “…SDM and PCM are mutually incompatible formats. Although both are at their roots nothing more than numbers, you cannot convert losslessly from one format to the other. The conversion process invariably results in the musical data being irrecoverably ‘smeared’ in the time domain.” But in this section of the original post above, BitPerfect is advocating for conversion from DSD to PCM…something he specifically states will “irrecoverably smear the musical data”. So which it is? Do we applaud DSD to PCM conversion or condemn it, Richard?

BitPerfect: “As an aside, you might well ask: If the ultrasonic noise is inaudible, then why do we have to filter it out in the first place? And that would indeed be a good question. According to auditory measurements, it is simple to determine that humans can’t hear anything above 20kHz – or even less as we age. However, more elaborate investigations indicate that we do respond subconsciously to ultrasonic stimuli that we cannot otherwise demonstrate that we hear. So it remains an interesting open question whether the presence of heavy ultrasonic content would actually have an impact on our perception of the sound. On the other hand, a lot of audio equipment is not designed to handle a heavy ultrasonic signal content. We know of one high-end TEAC DAC that could not lock onto a signal that contained even a modest -60dB of ultrasonic content (that problem, once identified, was quickly fixed with a firmware update). Such are probably as good reasons as any to want to filter it out.”

John Siau It is important to note that the ultrasonic noise is intermodulation distortion – it is not harmonically related to the music. This notion that it somehow contributes to the audio is erroneous. The ultrasonic noise contains intermodulation beat frequencies between the 64X sample rate and the audio content. There is absolutely nothing musical about the ultrasonic noise and distortion in a DSD signal. It must be removed.

If the ultrasonic noise is not removed, it can cause distortion in the power amplifier, and in the tweeters. This distortion will fold the ultrasonic noise and distortion into the audible frequency range.

Dr. AIX The 1-bit DSD encoding method requires that noise get shifted above the audio band. There’s simply no way to achieve reasonable dynamic ranges without doing this. The noise in a DSD 64 file starts at around 23 kHz and rises quickly. Other flavors of DSD have the same issue but the noise starts at a higher frequency.

Richard seems to be making the point that perhaps humans “respond subconsciously to ultrasonic stimuli” so having musical capture and reproduction to say 40 kHz (the JAS high-res spec) would be a good thing. His “on the other hand” scenario states what I said previously, that the ultrasonics could create difficulties for some hardware. See next paragraph.

BitPerfect: “So what do we do with the DSD content above 20kHz? In developing DSD Master we take the view that the content of this frequency range contains both the high-frequency content of the original signal (if any), plus the added high frequency noise created by the SDM’s noise-shaping process. We try to maintain any high frequency content within the signal flat up to 30kHz, and then begin our roll-off above that. Consequently, our DSD conversions at high sample rates (88.2kHz and above) do contain a significant ultrasonic peak in the 35-40kHz range. However, that peak is limited to about -80dB, which is way too low to either be audible(!) or to cause instability in anyone’s electronics. Meanwhile, the phase response is quasi-linear up to the point at which the ultrasonic noise rises above the signal level.”

John Siau DSD Master is flat to 30 kHz and it delivers quasi-linear phase. PCM 96 kHz is flat to about 45 kHz and it delivers linear phase. Both are excellent! The important point is that 64x DSD is not better than 96 kHz PCM.

Dr. AIX This gets back to the original point of this whole exchange…whether DSD 64 can outperform 96 kHz/24-bit PCM? There are musical instruments that produce ultrasonic sound. And there are microphones that can capture ultrasonic sound. Regardless of whether you believe these frequencies make a difference to our musical experience, a 96 kHz/24-bit PCM can capture them, while a DSD 64 encode cannot. I realize this becomes a theoretical argument for some but it’s too easy to do for me to simply ignore it.

Tomorrow will finish up this thread. I hope you’ve found it interesting. The fervor over DSD is misplaced…but will undoubtedly continue.


Mark Waldrep, aka Dr. AIX, has been producing and engineering music for over 40 years. He learned electronics as a teenager from his HAM radio father while learning to play the guitar. Mark received the first doctorate in music composition from UCLA in 1986 for a "binaural" electronic music composition. Other advanced degrees include an MS in computer science, an MFA/MA in music, BM in music and a BA in art. As an engineer and producer, Mark has worked on projects for the Rolling Stones, 311, Tool, KISS, Blink 182, Blues Traveler, Britney Spears, the San Francisco Symphony, The Dover Quartet, Willie Nelson, Paul Williams, The Allman Brothers, Bad Company and many more. Dr. Waldrep has been an innovator when it comes to multimedia and music. He created the first enhanced CDs in the 90s, the first DVD-Videos released in the U.S., the first web-connected DVD, the first DVD-Audio title, the first music Blu-ray disc and the first 3D Music Album. Additionally, he launched the first High Definition Music Download site in 2007 called A frequency speaker at audio events, author of numerous articles, Dr. Waldrep is currently writing a book on the production and reproduction of high-end music called, "High-End Audio: A Practical Guide to Production and Playback". The book should be completed in the fall of 2013.

2 thoughts on “Not So Bit Perfect About DSD vs. PCM? Part IV

  • July 5, 2015 at 3:21 pm

    This is an interesting link to a discussion and test of conversion using different encoders/decoders between PCM and DSD.

    • July 6, 2015 at 9:53 am

      This article is very informative…thanks for the link.


Leave a Reply

Your email address will not be published.

10 − 9 =