The Commentary: If My System Is Better, Why Does His Sound Better?

When the needle dropped on a worn copy of Kind of Blue, the room filled with something improbably alive. The trumpet didn’t just sound brassy. It sounded present — as if it had taken a train and arrived slightly irritated.

The problem was this: the same album on my brand-new digital system sounded cleaner, quieter, and technically superior — and faintly like it was being demonstrated in a showroom by someone called Martin.

How can reproduction technology improve so dramatically while the experience of being moved by music remains stubbornly resistant to progress?

“I don’t get it,” I said.

“You don’t get what?”

“Why this sounds better.”

Tom’s speakers were old enough to vote. The carpet looked historically damp. And yet when the trumpet entered, the room changed temperature.

At home I have a new DAC, an immaculate amplifier, and cables thick enough to restrain livestock. The same record sounds pristine. Detailed. Respectable.

This sounded alive.

“Maybe your system’s too perfect,” Tom said.

Deeply unhelpful, given what I’d spent ensuring it was.

Progress Is Real (Annoyingly)

Modern drivers use carbon fibre and beryllium. Lower mass. Greater stiffness. Fewer breakup modes. Amplifiers are quieter than ever. Digital conversion is astonishingly clean. Distortion figures that once required heroics now require microscopes.

On paper, we’ve solved almost everything engineers in the 1950s worried about.

“Exactly,” I said. “We’ve fixed the problems.”

“Have we?”

The question stopped being whether progress is real and became whether we’ve been polishing the right parts of it.

The Mechanical Inconvenience of Reality

Loudspeakers still move air.

And moving air inside a rectangular room filled with reflective surfaces and a coffee table the size of a small ferry is not precision engineering. Cones flex. Cabinets store energy. Crossovers pass frequencies like slightly distracted relay runners.

You can flatten frequency response until it resembles a Scandinavian skyline.

But the coffee table is still reflecting midrange into your shins.

“Your room’s terrible,” I said.

“Thank you.”

And yet Miles sounded present.

Which was awkward. Because technically speaking, my room is better. I have measurements. I once spent a weekend moving a rug three inches and calling it research.

Timbre, or Why We’re All Slightly Primitive

Flat frequency response is not the same thing as believable timbre.

Timbre — the difference between a trumpet feeling metallic and a piano feeling wooden — lives in harmonic relationships, in transient behaviour, in how drivers integrate. It’s about how energy starts and stops, and whether it behaves like one instrument or a small committee.

A speaker can measure beautifully and still make a piano sound faintly laminated.

Human hearing is extraordinarily sensitive to timbral cues. We evolved to recognise voices in the dark — to detect tension, warmth, threat. We don’t analyze those signals.

We respond.

Not all distortions are equal. Some low-order harmonics are perceived as warmth rather than error. Meanwhile, tiny timing inconsistencies — barely impressive on a spec sheet — can subtly erode realism.

“Are you saying vinyl is more accurate?”

“I’m saying it’s more convincing.”

Different claim. Much more irritating.

Diminishing Returns

There comes a point where improvement becomes refinement.

In audio, once distortion falls below perceptual thresholds and frequency response is broadly neutral, further reductions don’t scale emotionally. A tenfold reduction in distortion does not produce a tenfold increase in goosebumps. It produces better forum arguments.

Psychoacoustics suggests we’re not infinitely sensitive. Once errors drop below certain levels, they’re masked — by music, by the room, by the fact that you’re thinking about dinner.

At some stage, we are polishing decimals.

Decimals do not necessarily make Miles Davis appear in your living room.

The System Is Larger Than the Boxes

We talk about “the system” as if it ends at the speaker terminals.

It doesn’t.

The real system is:

Hardware + acoustics + psychology.

Hardware produces the signal.
The room reshapes it.
The brain interprets it — filtered through memory, expectation, and attention.

Measurements capture only part of that loop.

They describe electrical behaviour with extraordinary precision. They model acoustic output with increasing sophistication. But they cannot quantify nostalgia. Or expectation. Or the subtle integration cues that make a performance feel coherent rather than assembled.

Listening isn’t a lab test. It’s a collaboration.

When those elements align — even imperfectly — the result feels real. When they don’t, perfection can sound faintly like homework.

The record ended. The room went quiet in that particular vinyl way — charged, slightly smug.

“So what are you saying?” I asked.

“Maybe the goal isn’t perfect,” Tom said. “Maybe it’s convincing.”

Modern systems are objectively better. More accurate. Less distorted. The numbers are triumphant.

And yet the experience of being moved by music hasn’t kept pace with the specification sheets.

The big revolutions happened decades ago: stereo imaging that works, noise floors that don’t hiss, amplifiers that don’t combust. What we’re doing now is refinement.

Refinement is admirable.

But refinement isn’t revelation.

Maybe the ceiling isn’t technological.

Maybe it’s perceptual.

Next weekend, I’m bringing my amplifier to Tom’s place. Purely for scientific reasons.

After all, you can eliminate distortion — but you can’t spreadsheet goosebumps.