Reading music as groove geometry.

Excursion, width and velocity: the three mechanical demands a record places on the cutting stylus, and why they describe completely different physical problems.

Put a record groove under a microscope and you will see something that does not immediately look like music. At the outer edge of a side, the groove walls sometimes show large, slow, almost majestic lateral sweeps. Further in, you may find a section that looks almost still — barely any visible modulation. And somewhere between them, if the programme has bright percussion or vocal sibilance, the walls become ragged, corrugated, moving in extremely tight rapid oscillations that look chaotic beside the confident sweep of the bass groove.

The interesting thing is that the calm-looking section may be mechanically the most demanding of the three.

This apparent paradox is the starting point for understanding groove geometry. To cut a record well — and to prepare audio intelligently for cutting — you have to stop thinking about music purely in terms of dB, frequency response or integrated loudness, and start thinking about what the programme physically asks the cutting stylus to do.

There are three mechanical behaviours that describe this. Excursion: how far the groove moves. Vertical modulation: which direction it moves, and what stereo information does to groove geometry. Velocity: how fast it moves. These are not three ways of measuring the same thing. They are three separate physical dimensions of the problem, and they respond to the music in different ways.

Excursion: how far the groove moves

Groove excursion is displacement — how far the stylus is pushed sideways from the groove centreline. A deeply modulated bass note creates a groove that visibly swerves. A quiet section produces a groove that barely moves.

The relationship between displacement, frequency and velocity for a sinusoidal signal is:

Where v is stylus velocity, f is frequency and x is displacement. The practical consequence: for the same velocity, lower frequencies require progressively greater physical displacement. Halve the frequency and the groove needs to move twice as far to carry the same energy.

This is why bass is physically expensive. A sustained 40 Hz tone at a given level demands far larger groove excursion than the same level at 400 Hz or 4 kHz. A kick drum with strong low-frequency weight pushes the groove walls outward in large, slow cycles. Synthesiser bass with extended sustain can occupy enormous amounts of disc space. Even when these elements are not particularly loud in conventional digital terms, their effect on the physical record is substantial.

Groove spacing determines how much lateral room is available between adjacent turns. If a strongly modulated bass groove is spaced too tightly, the walls of adjacent turns approach each other. In extreme cases the cutter breaches the neighbouring groove entirely: groove collision, producing distortion on playback and potential damage to stylus and record alike.

This is why two recordings with similar integrated loudness or similar peak levels can behave completely differently at the lathe. A programme with extended, sustained sub-bass consumes disc space in a way that a loud but transient kick drum does not. Loudness as measured digitally does not describe this. The frequency content and the low-frequency envelope of the programme are what determine the excursion problem.

A large bass groove is not automatically a bad groove. The real engineering question is whether the modulation fits safely within the chosen spacing and cutting conditions. The problem only arises when the mechanical margin runs out.

Width: which direction the groove moves

A vinyl groove carries a stereo signal using the 45/45 cutting system. Each groove wall is oriented at 45 degrees to the vertical, one carrying the left channel contribution and one the right. The stylus modulates both walls simultaneously.

When both channels are identical — a mono signal, left equal to right — the stylus moves purely laterally, in the plane of the disc surface. The groove sways left and right. No vertical component, no depth change.

When the channels diverge, a vertical component is introduced. The conceptual relationship:

Mono content lives in the lateral dimension. Stereo difference content creates vertical motion. A signal that is strongly out of phase between channels produces predominantly vertical groove modulation.

This matters mechanically. Vertical stylus motion changes groove depth dynamically. It creates geometry that playback cartridges handle less comfortably than lateral motion, particularly at low frequencies. Excessive vertical modulation can cause tracking problems, groove contact difficulties and specific demands on the cutter head that lateral modulation alone would not.

Now combine this with the excursion relationship. Sub-bass is already physically expensive — large displacement for a given velocity. If that sub-bass is also strongly out of phase between channels, it produces large excursion and significant vertical movement simultaneously. The groove geometry becomes complex and the mechanical margin shrinks quickly.

The distinction is frequency-dependent, and this matters. Wide stereo ambience at 8 kHz is mechanically very different from wide stereo content at 40 Hz. At high frequencies, even substantial stereo difference information involves small displacement, so its mechanical impact is limited. It is specifically low-frequency out-of-phase content that combines large displacement with vertical groove complexity in a way that consumes mechanical margin and creates playback difficulties.

Stereo width on vinyl is not simply an aesthetic mastering consideration. At low frequencies, it becomes a question of physical geometry.

Velocity: how fast the groove moves

The third mechanical dimension is groove velocity — not how far the stylus moves, but how fast.

Return to v = 2πfx. At low frequencies, even large displacement involves modest stylus velocity. A slow, deep bass swing occupies physical space but does not ask the stylus to move quickly. At high frequencies, the mathematics inverts. Even a small displacement involves high velocity, because the cycle must complete in a very short time. Increasing frequency at a given displacement increases velocity in direct proportion.

This is why a vocal sibilant can be mechanically more demanding than a loud bass note, despite looking far smaller in the groove and measuring unremarkably on a level meter. The cutting head must follow those rapid oscillations accurately and cut clean groove walls at high speed. The cutter coil’s thermal and mechanical limits are tested by accumulated high-frequency energy even when the programme appears modest in conventional terms.

The same applies to cymbals, hi-hats, bright synthesiser transients, distorted guitars with complex upper harmonics, sharp electronic percussion. Material that appears as brief bursts on screen, or that measures at conservative levels, can produce groove velocity demands that the cutting system finds genuinely difficult. When that threshold is crossed the result tends to appear as coarse or broken groove walls, sibilance that smears and buzzes under a stylus, and high-frequency content that tracks roughly on playback.

Managing velocity, however, requires restraint. Suppressing high-frequency energy broadly reduces velocity but removes what that energy was carrying: cymbal presence, vocal air and definition, transient sharpness, the texture of reverb tails, the character of acoustic instruments. A record that cuts cleanly but sounds dull is not a success.

The goal is precision. Identify the events that are genuinely creating problematic velocity — not those that simply measure high, but those that demonstrably exceed what the cutting conditions can manage. Address those events as specifically as possible. Leave healthy material alone.

Three dimensions, not three versions of loudness

Excursion, vertical modulation and velocity are not three different ways of measuring the same thing. They describe separate mechanical axes, and the programme places demands on all three simultaneously.

Low frequencies ask the stylus to travel. High frequencies ask it to move fast. Stereo difference information asks it to move in a different direction entirely. These demands do not add up in any simple way, and ordinary digital loudness measurements do not describe the combination.

Consider three examples.

A loud mono 40 Hz bass tone creates large excursion — the groove swerves widely — but has almost no vertical component and modest velocity, because the signal is identical in both channels and the cycle is slow. Mechanically this programme asks for groove space and careful pitch. The vertical and velocity problems are relatively contained.

The same tone, strongly out of phase between channels, creates large excursion and substantial vertical demand simultaneously. Now the groove is fighting on two fronts: lateral space and vertical complexity, both driven by the same low-frequency energy.

A sharp 10 kHz cymbal transient presents almost the inverse situation. The displacement is small — nearly invisible under the microscope beside the bass groove. But the stylus must complete many thousands of cycles per second, cleanly and accurately. This is where the cutting head’s velocity ceiling and thermal margin are actually tested, not by the programme that looked dramatic on screen.

A moderately loud midrange passage — a clean chord, a fairly even vocal — may be mechanically the most straightforward of all. Moderate displacement, no extreme vertical demand, velocity within comfortable limits. Programme that may need very little attention.

LUFS, peak level and waveform appearance are not complete descriptions of the mechanical problem. Programme that looks alarming in digital analysis may cut straightforwardly. Programme that looks innocuous may carry bass, width or velocity demands that only emerge at the lathe.

The cutting engineer’s perspective

Managing these three dimensions in practice means combining different kinds of information. Digital analysis can identify where the largest excursion events occur, where low-frequency stereo difference content is strongest and where high-frequency velocity accumulates. This is genuinely useful. It identifies the regions of a side worth examining most carefully, and the programme characteristics worth understanding before the first test cut.

But analysis operates on the audio signal, not the groove. The relationship between signal and groove depends on cutting level, groove pitch, the cutter head and its current alignment, and the physical properties of the blank. A metric that appears concerning may produce a perfectly clean groove in practice. A metric that appears acceptable may reveal a problem under the microscope that the analysis did not predict. Investigating excursion, vertical modulation and velocity as separate measurements — identifying the strongest events in a side and validating them physically — is more useful than treating a single combined number as an answer.

The physical test cut remains the authority. If a passage cuts cleanly, has healthy geometry under the microscope and tracks well on playback, it should be left alone regardless of what the numbers say. Processing should solve a demonstrated or strongly justified mechanical problem. It should not chase arbitrary targets, and it should not remove from the music something the groove can carry without difficulty.

The goal is the closest possible mechanical translation of the programme to the format. That sometimes means careful intervention. More often it means restraint, and knowing precisely which events warrant attention and which do not.

Reading the groove

Once you understand excursion, vertical modulation and velocity as distinct properties of the groove, the way you look at a record changes.

A slow, wide bass sweep in the outer groove is not simply a loud moment. It is the programme asking for physical space at a specific frequency, with a specific stereo geometry, and the groove recording exactly how much it asked for.

A passage of tight, rapid corrugations that looks nearly featureless beside the bass is not a quiet moment. It is the programme demanding fast, precise stylus motion over many thousands of cycles, in a region where the groove walls are tested in a completely different way.

A midrange passage where the walls look simple and regular is telling you something about the programme’s mechanical character — not just its level.

The groove is a physical record of what the music required the cutting stylus to do. Once you can read it in those terms, it stops looking like a waveform scratched into plastic and starts looking like exactly what it is: a precise mechanical trace of the demands the programme placed on the cutter, one frequency at a time.