Personal Preference vs. Measured Performance: Is Better Really Better?
One of the most interesting, and often contentious, subjects in high-end audio is the relationship between measured performance and personal listening preference. An entire online community has emerged around the premise that maximizing SINAD (signal-to-noise and distortion ratio) should be the overriding consideration when evaluating audio equipment.
But if one audio component measures objectively better than another, does that necessarily mean you will prefer the way it sounds? Unfortunately, it's not that simple. And if you prefer the sound of a component that doesn't measure as well, does that invalidate the measurements? Absolutely not.
These are two fundamentally different concepts that are too often treated as the same.
Measured performance is objective. Personal preference is subjective. Both are important, but they answer fundamentally different questions.
Measurements tell us how accurately a component reproduces an electrical signal. Personal preference tells us how much we enjoy listening to it.
There are several excellent examples of this distinction in the world of high-end audio.
VINYL VS. DIGITAL AUDIO
Perhaps the most compelling example of personal preference versus measured performance is the enduring popularity of vinyl records.
From a purely technical perspective, digital audio offers substantial advantages over vinyl playback. Even the now decades-old 16-bit/44.1kHz CD format offers greater dynamic range, lower noise, lower distortion, superior channel separation, and freedom from the speed variations inherent in mechanical playback systems.
A turntable introduces numerous variables into the signal reproduction process, including cartridge design, tonearm resonance, tracking distortion, surface noise, mechanical vibration, and variations in rotational speed. Even exceptionally well-engineered turntables cannot eliminate all of these limitations.
Yet millions of music lovers continue to embrace vinyl, and some enthusiastically prefer its sound to that of digital audio.
Some of that preference may be attributable to differences in mastering, the sonic characteristics of the cartridge and phono preamplifier, or the inherent characteristics of vinyl playback itself. There is also something to be said for the physical experience of selecting a record, placing it on a turntable, lowering the stylus, and actively engaging with an album.
Does the continued popularity of vinyl invalidate the technical advantages of digital audio? Of course not. Nor do those technical advantages invalidate someone's preference for vinyl.
If measured performance were the sole determinant of listening enjoyment, the turntable would have largely disappeared decades ago. Instead, it remains one of the most celebrated components in high-end audio.
DELTA-SIGMA VS. R2R DACS
Modern delta-sigma DACs are capable of extraordinary measured performance, offering exceptionally low distortion and noise, excellent linearity, and tremendous dynamic range. Achieving comparable measured performance with traditional R2R ladder DACs is considerably more challenging. Overcoming some of the inherent limitations of conventional multibit conversion was a significant motivation behind the development and widespread adoption of delta-sigma technology.
Yet many audio enthusiasts prefer R2R ladder DACs, often describing their presentation as more natural, organic, or musical, even when the particular DAC they prefer doesn't measure as well.
Does that make R2R a superior conversion technology? Not necessarily. Does the superior measured performance of a particular delta-sigma DAC make someone's preference for an R2R DAC any less valid? Of course not.
It's also important to recognize that the conversion architecture is only one part of a DAC's design. Digital filtering, power supplies, and analog output stages all influence the performance of the finished product.
VACUUM TUBES VS. SOLID STATE
Perhaps the most familiar example in amplification is the longstanding preference among some audio enthusiasts for vacuum tubes.
A well-designed solid-state amplifier will typically exhibit lower distortion, lower noise, and lower output impedance than a traditional tube amplifier. Yet many listeners are drawn to the warmth, richness, and dimensionality they associate with tube amplification.
In some tube amplifier designs, the harmonic distortion characteristics and the interaction between the amplifier's output impedance and the loudspeaker contribute to audible differences in the presentation of music.
From a strictly engineering perspective, these changes represent departures from accurate signal reproduction. From a listening perspective, they may be entirely desirable.
Distortion is a measurable departure from accuracy, but not all distortion is audible, and not all audible distortion is necessarily objectionable.
LINEAR VS. SWITCHING AMPLIFIERS
For decades, Class A and Class AB amplifiers were widely regarded as the preferred choices for high-end audio reproduction.
Early Class D switching amplifiers reinforced that perception, as many had measurable performance limitations compared with their linear counterparts.
But amplifier technology has evolved considerably. Today, some of the finest Class D amplifiers available exhibit exceptionally low distortion and noise, remarkably low output impedance, and outstanding linearity, while delivering significant efficiency improvements.
Despite this, some listeners continue to prefer the presentation of traditional Class A or Class AB amplifiers, while others prefer modern Class D designs.
There is nothing inherently wrong with either preference. There are exceptionally well-engineered products employing both linear and switching amplifier topologies.
GaN FETs VS. SILICON MOSFETs
Even within the world of Class D amplification, there are different technologies and listening preferences.
Gallium nitride (GaN) FETs offer several inherent advantages over conventional silicon MOSFETs in switching applications, including the potential for faster switching transitions, shorter dead times, and lower switching losses.
These characteristics give amplifier designers opportunities to reduce certain sources of distortion at their origin, improve efficiency, and rethink the feedback strategies traditionally employed in silicon MOSFET-based designs.
However, the use of GaN FETs does not automatically guarantee that an amplifier will outperform a well-designed silicon MOSFET-based amplifier.
The modulation scheme, feedback topology, power supply, output filter, and overall circuit implementation remain critically important.
And even when one amplifier exhibits demonstrably superior measured performance, an individual listener may prefer the presentation of the other.
This is a distinction we live with every day at Peachtree Audio. We design, manufacture, and enthusiastically support amplifiers employing both GaN FETs and silicon MOSFETs. We absolutely love what both technologies can deliver, even when one particular implementation measures better than another.
How well a technology is implemented is often more important than the choice of technology itself.
DO BETTER MEASUREMENTS MEAN BETTER SOUND?
That depends on how we define better.
If an amplifier introduces less noise and distortion, maintains a more accurate frequency response, and reproduces the input signal more faithfully, we can objectively describe it as more accurate under those operating conditions. It is, in that respect, more transparent to the signal.
Whether someone prefers listening to it is an entirely different question.
We must also recognize that not every measurable improvement is audible. Reducing an already inaudible level of distortion by another order of magnitude is an impressive engineering achievement, but it doesn't necessarily translate into an audible improvement.
Conversely, a single measurement, whether it's SINAD or total harmonic distortion, doesn't tell us everything about how a component performs. The frequency, amplitude, and distribution of distortion products, frequency response, output impedance, behavior under different loads, and numerous other characteristics deserve consideration.
There is also an increasingly important distinction between measurable differences and meaningful differences. As modern audio electronics achieve extraordinarily low levels of noise and distortion, the pursuit of increasingly impressive specifications does not necessarily yield proportionate improvements in the listening experience.
And when attempting to establish whether two components actually sound different, rather than simply whether we believe they do, properly conducted, level-matched, blind listening comparisons can be particularly revealing — and humbling.
ENGINEERING EXCELLENCE AND MUSICAL ENJOYMENT
As manufacturers, we should continually strive to improve the measured performance of our products. Measurements are indispensable engineering tools that allow us to identify problems, validate designs, and quantify improvements. They are fundamental to what we do and should never be dismissed simply because they don't align with our expectations or preferences.
As audio enthusiasts, however, we should also be comfortable acknowledging that our personal preferences may not always align with the measurements.
There is nothing wrong with preferring vinyl to digital audio, an R2R DAC to a delta-sigma DAC, a vacuum tube amplifier to a solid-state amplifier, or one Class D implementation over another.
What becomes problematic is when we attempt to establish that personal preference as proof of the technical superiority of an entire technology. Equally problematic is dismissing someone's listening preference simply because the product they enjoy doesn't measure as well as another.
Personal preference is not proof of superior engineering, just as superior measured performance is not a guarantee of greater listening enjoyment.
A FINAL THOUGHT
Perhaps the high-end audio community would benefit from spending a little less time trying to convince others that their preferences are wrong and a little more time appreciating the remarkable variety of technologies available to us today.
We live in an extraordinary era for audio reproduction. The level of measured performance available at relatively modest price points would have been almost unimaginable a few decades ago. At the same time, technologies that are nearly a century old continue to bring tremendous enjoyment to music lovers around the world.
These two realities can coexist. They are not contradictory.
At Peachtree Audio, we believe in the pursuit of engineering excellence, the importance of objective measurements, and the value of critical listening. We also recognize that the loudspeakers, the listening environment, the quality of the recording, and the preferences of the individual listener all contribute to the overall experience.
We have no interest in telling someone that they shouldn't enjoy their favorite audio system because another system achieves a higher SINAD. Nor should anyone be made to feel that their preference for a highly transparent, exceptionally well-measuring system somehow makes them less passionate about music.
Good engineering deserves recognition. Personal enjoyment deserves respect. Neither needs to come at the expense of the other.
Ultimately, the goal of high-end audio is not to win an argument, validate a purchasing decision, defend a particular technology, or achieve the highest position on a measurement chart. It is to create an experience that makes us want to listen to more music.
If your audio system inspires you to rediscover your favorite recordings, explore unfamiliar artists, stay up a little later than you intended, or listen to just one more song, then it is accomplishing something far more meaningful than any specification alone could ever communicate.
Measurements help us understand how well our equipment reproduces music. They do not determine how deeply we connect with it.
And that connection is, after all, why we fell in love with high-end audio in the first place.