Are Modern DACs Actually Worth It? This Audio Expert Says ‘No’

Monty claims that high-res audio is a gimmick, but it’s up to you to believe it (or not).

Monty Montgomery of Xiph.Org Foundation busts some common beliefs about digital signal and high-resolution audio in a recently-resurfaced video from 2012. While it may seem dated, the topics he covers are still been rounding the audiophile forums to date.

Here, he suggests that high-end DACs that claim to improve sound quality may not actually make that much of a difference.

To prove this point, he showed his experiment using a both analog and digital equipment including:

• An HP 3325 signal generator
• A Tektronix 2246 oscilloscope
• An HP 3585 spectrum analyzer
• An eMagic USB 1 audio interface

This resulted in what he claims is a validation that even a standard 16-bit/44.1kHz digital audio (like on CDs) can accurately capture and reproduce all frequencies humans can hear without any quality loss. These findings go against the myths aboutthe limits of digital audio spread in some audiophile communities.

Debunking the “Stair-Step” Myth

One of the most widespread beliefs is that digital audio represents waveforms as “stair-steps” due to the discrete sampling process. However, Montgomery’s tests prove otherwise.

In his video, he converts a 1kHz sine wave to 16-bit 44.1kHz digital audio and then back to analog. Despite the digitization, the output waveform on the oscilloscope stays a perfectly smooth sine curve with no jagged edges at all.

He even when increased the input all the way up to 20kHz – the highest frequency within the range of human hearing and the theoretical Nyquist limit for the 44.1kHz sample rate. But, the reconstructed analog sine wave still showed no degradation or distortion.

As Montgomery explains, these stair steps don’t actually exist in the output of a properly functioning DAC. He says the confusing link between digital signals and stair steps is likely due to:

• Zero-Order Hold: This method extends each digital value until the next one, visually creating a stair-step pattern. It’s commonly illustrated in explanations of how DACs work but is not representative of the final analog output, which is smoother.
• Simplified Illustrations: Engineers and educators often use simplified diagrams to explain complex concepts. So, the stair steps are used for visual clarity, although this is technically inaccurate. This method is similar to representing pixels as squares in digital images, where both are point-like and dimensionless.

Bit Depth’s Impact on Noise, Not Waveforms

Another common misconception is that higher bit depths result in smoother waveform reconstruction and better audio quality. But Montgomery’s demo shows this isn’t true.

Yes, higher bit depths can lower noise floors and increase dynamic range, as seen in his experiment. Yet, the audible benefits of going beyond 16-bit may be minor for most listeners.

He also notes that at 16-bits, digital audio boasts a theoretical dynamic range of 96dB. 24-bit extends this to an unnecessary 144dB – “greater than the difference between a mosquito and a jackhammer one foot away.”

Montgomery also touches on dither in his video. His demonstration shows several key aspects of how dither affects digital audio processing:

• Improves Audio Quality: By adding a low level of noise before reducing the bit depth, dither makes the resulting audio sound smoother and less harsh.
• Effective at Low Bit Depths: Even at low bit depths, such as 8 bits, dither can maintain a smooth sine wave on the output. This means dithering can enhance the quality of digital audio, even with significant reductions in data.
• Controls Noise Characteristics: Dither noise is more consistent and less harsh than quantization noise. This means the characteristics of the noise floor can be controlled and optimized.
• Reduces Distortion: By spreading out quantization noise into a uniform noise floor, dither prevents noticeable distortions.
• Minimal Impact at Higher Resolutions: At 16 bits, which is standard for CDs, dither’s effect is almost negligible. This points to the idea that the complexities of dither and its impact are minor.

The Truth About Digital Sampling

The video also addresses the belief that digital audio can’t accurately capture signals between samples. Many audiophiles claim that this results in timing errors or losing transients.

But, Montgomery slams this using a band-limited square wave input, which contains an infinite sum of odd harmonics. Here, he shows that the original analog signal waveform is perfectly reconstructed by the 44.1kHz sample rate regardless of where the samples are positioned.

This works because anti-aliasing filters band-limit input signals before sampling per the Nyquist theorem. Any frequencies above the Nyquist limit (half the sample rate) are blocked to avoid aliasing distortion.

Speaking of aliasing, he also clarifies the belief that they cause unnatural “ringing” in reconstructed waveforms, known as the Gibbs effect.

Montgomery shows that passing the same band-limited square wave through the filter repeatedly doesn’t amplify or change the existing ringing artifacts.

What This Means on Audio Consumer Choices

Montgomery’s analysis questions the real need for high-end DACs for most listeners. It suggests that premium features marketed as essential for high-fidelity playback may not provide enough audible benefits to justify their costs.

For example, the demo shows no harsh digital-to-analog transitions. So, buying a high-end DAC to eliminate non-existent problems (stair steps) is based on a fundamental misunderstanding. Inexpensive DACs can also produce artifact-free analog outputs.

For bit depth, Montgomery notes 16-bit already exceeds most listeners’ hearing limits. This questions the value of ultra-high bit-depth DACs for everyday use.

Similarly, he shows dither’s effects are minor, and basic dithering methods are adequate for high quality audio. As such, advanced dithering features in DACs are likely more relevant to audio professionals than to general consumers.

Community Reactions

The reaction to Montgomery’s claims was largely positive, with many praising the clear and effective way the ideas were presented.

However, the video also reignited discussions among audio professionals and enthusiasts. Some led to a reevaluation of long-standing beliefs about digital audio.

Aaron Vockley, a sound engineer, expressed a revelation upon watching Montgomery’s demonstrations.

Meanwhile, some audiophiles showed a preference for personal experience over technical specs.

However, there was also skepticism from those sticking to traditional views despite the presented evidence.