Knowledge hub / AFG Music / rhythm
Polymeter vs Polyrhythm, Explained Properly
Music knowledge / Guides and practical tools
About this reference
From the music-learning collection, adapted for Philojain Music Muse. Referenced sources remain credited in the article.
Rhythm & Groove · Meter, Untangled
Two different conflicts get called by the same three or four names, constantly. Here is the precise distinction, the displacement and metric-modulation devices that get lumped in with it, the arithmetic that predicts exactly when two independent cycles next share a downbeat — and a calculator that proves it rather than asserting it.
Every specific “realigns after exactly N repetitions” figure that could not be independently verified was left out — see the method note under the ledger.
01 — The Distinction
One conflict, or a different one.
Polyrhythm is two or more conflicting subdivisions of one shared span of time. Both parts start together and cover the same stretch together — what conflicts is the fine grid inside that shared stretch. The classic case is three evenly-spaced notes against two evenly-spaced notes over the same duration: a 3:2 ratio, formally called a hemiola. Downbeats coincide; the fine grid doesn't.
Polymeter is different. Two or more parts share the same tempo and note value — both ticking at the same speed — but group that shared pulse into different-length bars. Because the group lengths differ, each part's own “beat one” lands somewhere new relative to the other's bar-lines almost every time through, until, after a calculable stretch, they share a downbeat again. The grid stays aligned throughout; the downbeats drift, then periodically resync.
One memory device worth keeping: in polyrhythm the downbeats line up but the grid doesn't; in polymeter the grid lines up but the downbeats don't. A single odd meter used alone — a 7/8 pattern with nothing layered against it — is neither: both terms describe a conflict between two simultaneous layers, and one meter on its own has no second layer to conflict with. The two categories aren't mutually exclusive either — a single passage can legitimately use true polyrhythm in one place and polymeter in another.
Beat and accent displacement is a third, related but distinct idea. The meter itself never changes; what moves is where inside that unchanging meter an accent falls. Academic metrical-dissonance theory calls this “displacement dissonance” — a layer using the same group-size as the surrounding meter but offset by a fixed number of beats, landing accents on normally-weak positions — and distinguishes it from “grouping dissonance,” which is what polyrhythm and polymeter actually are (differently-sized groups sounding together). A displaced accent can be built two ways with the same audible result: subtractively, by deleting hits from a fast, regular grid so the survivors fall in what read as gaps; or additively, by inserting isolated hits into silence. Same relocation, opposite construction, different effect on the surrounding texture.
Metric modulation is a different animal again — it is the one device on this page where the tempo itself genuinely changes. A specific notated note-value is used as a pivot: a value that plays one role at the old tempo is reinterpreted to play a different role at a new tempo, while its real-world duration holds constant across the boundary. Polymeter, polyrhythm and displacement all hold one tempo fixed and vary only the grouping; metric modulation pivots the tempo itself. A practical test: if you can tap one unchanged foot-tap speed through a passage, you're hearing polymeter, polyrhythm or displacement. If the foot-tap itself must change speed partway through — even though the transition feels logical rather than sudden — that's metric modulation.
The device this page's calculator models directly — a straight, unchanging reference pulse held by one part while a second part repeats a figure whose own length doesn't evenly divide the reference bar — is polymeter's concrete, arrangement-level realization. Both parts stay in one actual tempo and can be notated in a single unchanging meter throughout; nothing about the notated time signature needs to change. What's exploited is the gap between what's counted (one steady, uninterrupted meter) and what's heard (an accent that seems to roam to a new position each cycle). The device only reads to a listener if the reference role is unambiguous and constant — if the fixed part is buried or not clearly established early, the payoff becomes inaudible. Independent sources document the same underlying arithmetic — two things of different cycle-length, drifting and then periodically re-coinciding — appearing separately in at least two other, mechanically different musical traditions across history, which is exactly why the arithmetic below is treated as a general principle rather than a one-genre trick.
Related and separately documented: performers commonly internalize an asymmetric bar by feeling it as small cells of twos and threes (five felt as “2+3,” seven as “2+2+3,” and so on) rather than counting linearly. This is a real, named ethnomusicological category — “aksak,” meaning roughly “limping” — and, like a lone 7/8, it is not itself polymeter or polyrhythm unless combined with a second, differently-grouped layer.
02 — The Full Ledger
12 devices, sortable.
Search a device, filter by family, or sort any column.
| Polymeter | Tempo and note values | Each part's own bar/grouping length, so barlines land differently | Each part's own cycle length in beats; downbeats coincide again after LCM(A, B) | Two different-sized wheels turning together — slow drift, then a clean snap back |
| Polyrhythm (hemiola-type) | The one shared time-span both parts fill together | How that single span is subdivided inside | Express as a ratio (3:2, 4:3, 5:4…) — same span, different subdivision | A brief cross-grain “gearshift,” usually a beat or a bar long |
| Beat / accent displacement | The notated meter and bar length | Where inside that meter an accent falls, by a fixed offset | State the offset directly: “+n” or “−n” beats from the strong beat | Something leans, stumbles, or reads as behind/ahead of the beat |
| Metric modulation | The real-time duration of one chosen note-value | The actual tempo (speed of the beat) | State the pivot equivalence: “value X at the old tempo = value Y at the new” | A tempo change arriving through an exact, notatable pivot, not a jump-cut |
| Static pulse vs. asymmetrical figure | One part holding an unchanging reference pulse throughout | A second part's figure, whose length doesn't divide evenly into the reference bar | Figure length vs. bar length, same unit; realign after LCM(reference, figure) beats | Mounting, machine-like tension toward an audible “click” of arrival |
| Realignment / resolution point | The two stated cycle-lengths, held perfectly fixed | The phase gap between the parts, growing every repetition until it returns to zero | Compute LCM(a,b); divide by each length for repetitions completed | Anticipation cashing out in an audible landing; often a structural arrival point |
| Mixed meter | Only one meter sounds at any instant | The notated meter itself, bar to bar, in sequence | Read time-signature changes directly, bar by bar; no LCM applies | An immediate lurch at each bar-line, not a gradual drift-then-snap |
| Hemiola (3:2) | The total duration of one shared short span | Whether that span groups as 2 units of 3, or 3 units of 2 | Simplest possible ratio, 3:2 — the most common polyrhythm | A quick gearshift or lean, often at a cadence; resolves almost instantly |
| Additive / asymmetric grouping (aksak) | The total beat-count of the bar | How that total is internally chunked into uneven cells | State the cell breakdown explicitly — 7 = 2+2+3, not just “seven” | A limp or uneven walk, felt in cells rather than counted as one number |
| Subtractive vs. additive syncopation | The resulting accent pattern can be identical either way | Whether hits are deleted from a dense grid, or inserted into silence | Name the direction used — it changes the surrounding texture even when the accent matches | Subtraction “trips” forward through gaps; addition feels sparse and punctuated |
| Cross-rhythm (umbrella term) | Nothing specific — a broad label for either mechanism below | Whether the conflict is within one span (polyrhythm) or shared-pulse/different-bar (polymeter) | Always name the specific mechanism underneath; never leave “cross-rhythm” unexplained | n/a — a terminology entry, not a perceptual quality of its own |
| Grid-role assignment | Which part holds the fixed reference pulse, set early and never doubled | Which part is allowed to be the drifting/cycling voice | Not a counting concept — an arrangement decision | The whole device is inaudible if the reference role isn't clear throughout |
Method: every row reflects standard, cross-checked music-theory terminology (see Sources), not a single worked example from any one piece. Any “realigns after exactly N repetitions” figure tied to one specific worked example was left out of this page entirely — the arithmetic itself (the lowest common multiple) is the durable, generic fact, and the calculator below lets you check any pair of lengths yourself rather than trusting a single quoted number.
03 — Try It
Where do they realign?
Pick a pulse length and a figure length, both in beats, and see exactly where the two next share a downbeat. Pure arithmetic — the lowest common multiple of the two lengths — no audio needed and nothing to guess.
The realignment calculator
Inputs are capped 2–16 beats to keep the grid readable — the arithmetic works identically at any length.
Deeper — The West African Bell Pattern
The same structure, a different tradition: the West African bell pattern
Every worked example on this page so far comes from Western notation or a Western/prog-metal academic catalogue — worth naming, because the structure this page defines is not a Western invention. Ethnomusicologist Michael Frishkopf's peer-reviewed account of West African polyrhythm (SHS Web of Conferences, 2021) describes the double bell, or gankogui, as the instrument that “marks the basic ostinato timeline” for an entire ensemble — a fixed twelve-pulse cycle functioning exactly as the “stays constant” reference row in the ledger above defines a polymetric grid.
That same twelve-pulse cycle, per Frishkopf, supports more than one simultaneous reading: it can be “divided into 4 equal beats (three pulses each), 6 equal beats (2 pulses each), or a combination-division into 5 unequal beats, such as 3 beats… plus 2 beats… as in hemiola.” Different instrumental lines in the ensemble lock onto different divisions of that same fixed cycle at once — the identical grouping-dissonance structure Section 01 defines abstractly, independently documented in a tradition with no historical connection to the Western time-signature notation the rest of this page uses to describe it.
Deeper — Naming The Other Tradition
Naming the “other tradition” Section 01 only gestured at
Section 01 already notes that the same drift-and-realign arithmetic “appears separately in at least two other, mechanically different musical traditions across history” without naming them beyond isorhythm. This is one, concretely: a fixed bell-pattern cycle read simultaneously as groups of three and groups of two is structurally the same 3-against-2 relationship as the hemiola this page's own ledger already lists — arrived at independently, centuries apart, with no shared notation system connecting the two.
The ledger's “static pulse” row, played by a full ensemble
The ledger above already lists “static pulse vs. asymmetrical figure” as its own device: one part holding a fixed reference while a second part's differently-sized figure drifts against it. Frishkopf's account gives that row a live, ensemble-scale example: the lead drum line, he writes, “often sings longer phrases” and “introduces cross-rhythms that contrast sharply with the regular beat… as well as the underlying instrumental ostinati” the bell and supporting parts keep. The bell holds the row's “stays constant” column; the lead drum is its “what shifts” column, played by two different musicians rather than one performer's two hands.
04 — Common Questions
Asked while counting bars.
What's the real difference between polymeter and polyrhythm?
Polyrhythm is two conflicting subdivisions of one shared span of time — both parts start together and cover the same stretch together, and what conflicts is the fine grid inside that shared stretch (the classic case is three evenly-spaced notes against two evenly-spaced notes over the same duration, a 3:2 ratio called a hemiola). Polymeter is different: two parts share the same tempo and note value, ticking at the same speed, but group that shared pulse into different-length bars. Because the group lengths differ, each part's “beat one” lands somewhere new relative to the other's bar-lines almost every time through, until — after a calculable stretch — they share a downbeat again. Short version: in polyrhythm the downbeats line up but the grid doesn't; in polymeter the grid lines up but the downbeats don't.
Is a single odd time signature like 7/8 a polymeter?
No. A single odd meter used on its own, with nothing conflicting layered against it, is neither polymeter nor polyrhythm — both terms describe a conflict between two simultaneous layers. An unaccompanied 7/8 pattern is just an asymmetric meter; it only becomes polymeter or polyrhythm once a second part is added that groups or subdivides that same span of time differently.
Why does the lowest common multiple matter here?
If a reference part repeats every P beats and a cycling figure repeats every F beats, the two share a downbeat again for the first time after exactly the lowest common multiple of P and F — standard, uncontroversial arithmetic. The catch: LCM(P, F) equals P times F only when P and F share no common factor. When they do share a factor, the true realignment point is smaller than the simple product and arrives sooner, so “just multiply the two numbers” gives a wrong, too-large answer whenever the lengths aren't coprime — try 4 and 6 in the calculator above to see it happen.
What's the difference between polymeter and metric modulation?
Polymeter, polyrhythm and displacement all hold one tempo constant and vary only the grouping on top of it — if you can tap one unchanged foot-tap speed through the passage, that's what you're hearing. Metric modulation is the opposite: the tempo itself genuinely changes, using a shared note-value as an exact pivot between the old speed and the new one. If the foot-tap itself has to change speed partway through, even though the transition feels logical rather than sudden, that's metric modulation.
What is “aksak” and how does it relate to polymeter?
Aksak (a documented ethnomusicological term meaning roughly “limping”) describes an asymmetric meter built from small cells of two and three — 5 felt as “2+3,” 7 as “2+2+3,” and so on — rather than counted as one long number. On its own, a single aksak meter is not polymeter or polyrhythm, for the same reason 7/8 alone isn't: there's no second conflicting layer. It becomes relevant to this page's other devices only when an aksak-grouped part is combined with a differently-grouped part.
05 — Sources
Where these facts come from
- Encyclopaedia Britannica — “Polyrhythm” (formal definition)britannica.com · accessed 28 Jul 2026
- Soundbrenner — “Polymeter” explainer, including worked realignment arithmeticsoundbrenner.com · accessed 28 Jul 2026
- Open Music Theory — “Metrical Dissonance” (displacement dissonance vs. grouping dissonance)viva.pressbooks.pub · accessed 28 Jul 2026
- Baylor University Libraries — open-access rhythm and meter textbook, chapter on metric modulationopenbooks.library.baylor.edu · accessed 28 Jul 2026
- Encyclopaedia Britannica — “Aksak” (additive/asymmetric meter)britannica.com · accessed 28 Jul 2026
- Encyclopaedia Britannica — “Isorhythm” (independent historical tradition of two different-length cycles overlapping and periodically re-coinciding)britannica.com · accessed 28 Jul 2026
- Music Theory Spectrum (Oxford University Press, for the Society for Music Theory) — peer-reviewed analysis of large-scale odd time signatures and metric superimposition in a technical/progressive-metal recorded catalogue, Vol. 29/2, Fall 2007academic.oup.com · accessed 28 Jul 2026
- Music Theory Spectrum (Oxford University Press) — follow-up peer-reviewed rhythmic analysis extending the same territory, Vol. 40/1, Spring 2018academic.oup.com · accessed 28 Jul 2026
- University of Puget Sound — “Music Theory for the 21st-Century Classroom,” §35.2 “Phase Shifting”musictheory.pugetsound.edu · accessed 28 Jul 2026
- Michael Frishkopf (University of Alberta), “West African Polyrhythm: culture, theory, and representation,” SHS Web of Conferences, vol. 102, art. 05001, 2021shs-conferences.org · accessed 11 Aug 2026
About this page: the distinction, the terminology and the arithmetic above are all cross-checked against the sources listed here. Every specific numeric claim tied to one particular worked example that could not be independently verified was omitted rather than repeated — the calculator lets you verify the underlying arithmetic yourself for any pair of lengths.
Ask the engine
Any music topic, at the depth you choose
The Concierge
Still not sure which is which?
Describe the two parts you're hearing and ask the concierge whether it's polymeter, polyrhythm, or something else — trained on AFG Music's real pages.
The three engines, and how they combine
Polymeter — The Length Engine
Same pulse, different bar lengths. The grid lines align and the downbeats don't — which is the mechanism behind almost every riff you like.
The definition, and the conflation problem
- Polymeter: "the grid lines are aligned, but the downbeats aren't." One tempo, one pulse, but two or more layers grouping that pulse into different lengths — a 7-beat riff under a 4-beat cymbal pattern. They realign at the least common multiple, and that realignment is the payoff.
- Polyrhythm is the opposite arrangement: "the downbeats are aligned, but the grid lines aren't." Same starting point and cycle length, different subdivision rates.
- These get conflated constantly, and the person who put it best notes we "could have saved a lot of confusion by calling this 'polytempo' or 'polysubdivision' rather than 'polyrhythm.'" Odd meters — 5/4, 7/4 — are also frequently mistaken for both, despite being neither.
- Krebs's formal name for it is grouping dissonance, notated G x/y, where x and y are the two layers' cardinalities and x/y is not a whole number. A 7-against-4 riff is G7/4.
- Two sub-types worth knowing. Explicit polymeter notates both meters — Bartók's "Harvest Song" (1931). Implicit polymeter notates one and implies the other through accent and phrasing — Ravel's String Quartet, second movement (1902), written in 6/8 with a melody implying 3/4. Almost all metal polymeter is implicit: the chart says 4/4, the riff says otherwise.
Meshuggah — the numbers, because they're better than the description
- "Bleed": the riff is built from sixteenth-note groupings of 7-7-5-3-5 = 27 sub-beats, cycling against a steady 4/4 drum pulse. Twenty-seven against sixteen — they realign every 432 sixteenths, which is why it feels endless.
- "New Millennium Cyanide Christ": transcribed by Pieslak as a 23/16 figure.
- "Dancers to a Discordant System": a 20-beat phrase repeated three times — 60 beats, 16 bars.
- Flag: those figures come through a synthesis paper citing Pieslak and Capuzzo rather than from the primary texts directly, so treat them as one layer removed. A fan analysis of "Ligature Marks" gives 37 eighth notes in the intro, alternating 20/21 in the verse and 22 quarter notes in the outro — enthusiast-sourced, plausible, not peer-reviewed.
- And the band deflates the mathematics themselves. Tomas Haake: "There's less mathematics involved than anyone would think. It's more organic than it might look from the outside." On the method: "The songs they'd written before I joined had an odd time signature aspect to them, but there was no real repetition going on — you'd just let the riff roll over the bar line. That's what we still do, although we've taken it to the extreme now."
- "Let the riff roll over the bar line" is the entire technique in six words, and it's a better instruction than any theoretical term.
The constant that makes it groove instead of collapse
- Across every documented Meshuggah example, the cymbals hold a steady quarter-note pulse. The riff cycles at an odd length; the drummer's right hand does not. That is the load-bearing decision — without a stable reference layer, polymeter is just arrhythmia.
- Car Bomb's Greg Kubacki describes the perceptual effect from the player's side: "I've always been fascinated with rhythms that sound like they're speeding up and slowing down, but are in a particular time. You get this perception of a certain tempo while we're thinking in another one." That is a guitarist independently arriving at Krebs's premise.
- Battles do it with a machine instead of a band. John Stanier: "There's this master loop on every single one of our songs... In a normal band situation the drums is what's in the driver seat, but in Battles the loop is the boss." The misalignment is against a fixed pre-recorded layer rather than negotiated live — mechanically different from Dillinger's all-human version, and it sounds different.
- And the famous cymbal has a verified explanation. Stanier's crash sits about six feet up: "Originally it was a joke, but it's also because in the very beginning I would rarely hit it, sometimes only once in a song, so when I did hit it I wanted it to look like a really big deal." Elsewhere: "I use it as a marker. It's like a master reset button." A reset marker is exactly what a polymetric arrangement needs — no broken-stand story, contrary to the usual telling.
Polyrhythm — The Subdivision Engine
Same span, different divisions. The downbeats align and the grid lines don't — the older, more danceable, and far more misused of the two.
The definition, and the genuinely old vocabulary
- Polyrhythm: two different subdivision rates sharing the same starting point and cycle length. Three against four, three against two, five against four. Both layers begin together and end together; what differs is how many pieces each cuts the span into.
- Hemiola is the 3:2 case and it is very old — from Greek hemiolios, "half again," adapted into rhythmic usage from fifteenth-century mensural notation onward. Structurally a hemiola is a Krebs G3/2 grouping dissonance; the term just predates the taxonomy by five centuries.
- Cross-rhythm is a related but stricter term, defined in The New Harvard Dictionary of Music as "a rhythm in which the regular pattern of accents of the prevailing meter is contradicted by a conflicting pattern," and introduced analytically by A.M. Jones in 1934. The distinction that matters: cross-rhythm is the case where the conflict is the generative principle of the entire piece — "the meter is in a permanent state of contradiction" — rather than an incidental simultaneity.
- Contrametric is Kofi Agawu's term for accents that contradict the underlying meter, as against metrical accents that reinforce it. Agawu argues the contrametric layer is often the more perceptually fundamental one in African musics — which is a useful inversion if you're used to thinking of the drums as ground truth.
- And 3-against-4 / 3-against-2 is described as "the single most commonly used rhythmic device in Black American music." Meanwhile 7-against-4 is essentially absent from pop. If you want something that grooves rather than merely puzzles, the ratio matters enormously.
Where the funk actually comes from
- The timeline pattern — clave in Cuban practice, "the standard pattern" in West African scholarship — is a series of attack points, not a set of durations. The classic seven-stroke pattern in 12/8 lands on 1, 1a, 2&, 2a, 3&, 4, 4a. Scholars including Gerhard Kubik, Kofi Agawu and David Locke treat these as the structural core.
- And there's a documented analytical error worth knowing, because it is the error most listeners make too: A.M. Jones identified the pattern's importance but, per later critics, "mistook its accents as indicators of meter rather than cross-rhythmic phenomena." Reading the accent pattern as the meter is exactly the mistake a listener makes when the reference layer is not stated.
- Additive versus divisive rhythm, coined by Curt Sachs in Rhythm and Tempo (1953): divisive divides a span into equal units; additive concatenates unequal units end to end — 9/8 as 4+2+3 in Bartók's Fifth Quartet, 3+3+2 in Stravinsky's Octet. Aksak (Turkish, "limping") names the Balkan and Middle Eastern additive tradition. Flag: Justin London has criticised the additive/divisive dichotomy for conflating notation with perception — an active debate, not settled consensus.
- Funk-metal's slap vocabulary sits on top of this. Nuno Bettencourt is described for "syncopated riffs, funk-inspired rhythms, dexterity, technical ability, and speed" — one writer calls him "the funkiest dude in the shredder family tree," and Brian May reportedly called the "Get the Funk Out" solo "a landmark in rock history." Living Colour's Time's Up (1990) is documented as blending "jazz fusion, punk rock, Delta blues, hip hop, funk, thrash metal, jive, and hints of electronica" — the eclectic template, twenty years early.
One term to avoid, and one debate to respect
- Do not use "isorhythm" for this. It is a real term — coined by Friedrich Ludwig in 1904 for a thirteenth-to-fifteenth-century French motet technique where a fixed rhythmic pattern (talea) runs against a melodic pattern (color) of different length so the two periodically realign. That is structurally almost exactly what Meshuggah do, which makes the borrowing tempting — but I found no scholarly usage applying it to rock or metal, and using it will read as imprecise rather than erudite.
- "Systemic versus figural" could not be verified at all as a term-pair in this literature, despite targeted searching. If you've seen it, it isn't from Krebs.
- And the honest caveat about the whole enterprise: Stephen Hudson's "How Much Math Is in Math Rock?" (2020) argues against over-mathematising metal rhythm, proposing embodied "motional conceptual models" instead. Its actual analytical objects are Meshuggah, Metallica, Dream Theater and Led Zeppelin rather than the math-rock canon. Not every scholar thinks counting is the right lens — which matches Haake's own "more organic than it might look."
Metric modulation — The Pulse Engine
The beat itself relocates. A subdivision of the old tempo becomes the new pulse — and the trivial version of this is what a straight half-time drop actually is.
The definition, precisely, with numbers
- The term was coined by critic Richard Franko Goldman in 1951, reviewing Elliott Carter's Cello Sonata (1948). Carter himself preferred "tempo modulation." So the usual "Carter coined it" line is half right — he pioneered the technique, not the name.
- The mechanism: a note value in the old tempo is made durationally equal to a differently-subdivided note value in the new tempo, and that shared duration is the audible pivot linking the two.
- Worked example — Carter's Eight Etudes and a Fantasy (1950): starting at quarter = 84 in 4/4, two half notes at the old tempo are made equal to three half notes at the new. The old quarter at 84 becomes a dotted quarter at 84; 84 × 1.5 = quarter = 126.
- Second example — "Canaries" from Eight Pieces for Four Timpani (1949): triplets in a 3/4 bar pivot into 3/8 with the eighth-note value held constant, then reverse to return.
- The essential requirement is a non-duple pivot. A ratio like 3:2 or 4:3 is a real modulation because the new pulse was derived from a subdivision that wasn't already the beat.
Which means the half-time drop is something else — and that's useful
- Half-time doubles note lengths while "the actual time value remains the same." The click never changes; only the backbeat placement is reinterpreted — the snare moving from 2 and 4 to one hit per two bars. In dubstep terms: "Half time of 140 is 70, that's just math" — the drums run at the halved rate while the bassline runs at full tempo so the track still feels like 140.
- So 180 BPM thrash to a 90 BPM drop is half-time, not metric modulation. The 90 was already there, embedded in the 180 as every other beat. Nothing was derived; nothing pivoted.
- And by Krebs's own definition it's not even dissonance. Metrical consonance is "one layer is an integer-multiple of the other and both layers have coincident phases." A clean 2:1 halving with aligned phase is formally consonant — which is a slightly funny thing to say about a breakdown designed to level a room, but it is rigorously true, and it explains why half-time drops feel massive rather than unstable.
- Here is the useful part. If you want the drop to feel like the floor genuinely moved rather than merely widened, pivot on a triplet instead: take the triplet-quarter of 180 and make it the new beat, which gives you 120 BPM, a 3:2 relationship, and a genuine modulation. 180 → 90 is heavier. 180 → 120 is stranger. They are different tools and you should ask for them by name.
Phasing — the continuous version, and the accident that produced it
- Steve Reich's It's Gonna Rain (1965) came from equipment failure.* He played the same tape loop of a preacher's voice on two Wollensak recorders intending them to align; "minute differences in the machines, the length of the spliced tape loops, and playback speed" made them drift. He kept it: "all possible repeated harmonies are explored before the two loops eventually get back in sync." Come Out*** (1966) applied the same technique to "come out to show them."
- Piano Phase (1967) is the live version: two pianists start a repeating figure in unison, one gradually accelerates until exactly one note ahead, then locks. Three sections of 12-, 8- and 4-note patterns; the final phase cycle repeats eight to sixty times per the score.
- Phasing is a process; Krebs's displacement dissonance is a state. Phasing passes continuously through every possible offset. D x+n describes a snapshot — a fixed offset held. For a riff-based band the fixed offset is far more usable, but a slow phase across a long section is a genuinely underused move in heavy music.
Metal meets electronics — the real chronology, with one correction
- "Thrashstep" is not an established genre term. It exists as a Last.fm community tag; I found no usage by any major music publication. Treat it as your own coinage, which is fine — but don't expect anyone else to recognise it.
- Korn's The Path of Totality (2011) is the canonical metal-dubstep record — released 2 December 2011 in Europe, 6 December in the US, with Skrillex, Noisia, Excision, Downlink, 12th Planet, Feed Me, Kill the Noise, Datsik and Flinch. #10 US Billboard 200, 55,000 first week, #1 US Dance Albums, ~270,000 US copies by January 2013. Metacritic 58; AllMusic 4/5; The A.V. Club gave it an F.
- And Korn rejected the framing outright. Jonathan Davis: "We didn't make a dubstep album. We made a Korn album." His own preferred label was "future metal." On the concept: "We're mixing metal and electro music, and you're not supposed to do that."
- Enter Shikari is the earlier and cleaner example — formed 1999, breakthrough with Take to the Skies (2007), blending post-hardcore with drum and bass, dubstep, techno and trance. Critics called them "the first to release a metalcore album that could also pass as a respectable techno effort." Four years before Korn.
- The Skrillex connection is real and properly sourced. Sonny Moore fronted From First to Last from 2004, recording Dear Diary, My Teen Angst Has a Bodycount (2004) and Heroine (2006), leaving in 2007 after vocal surgery. And critic Simon Reynolds observed that "the high-energy, distorted bass riffs of brostep functioned more similarly to metal guitar than subsonic bass" — a named critic drawing the line, not just internet folklore.
- Correction: Bring Me The Horizon's That's the Spirit (2015) is not a dubstep turn. It's an alt-rock, pop and synth turn — Metacritic 88, described as an "evolution from artsy metalcore to cinematic pop rock." Don't cite it as metal-dubstep.
- Rusko on brostep, in his own words: "Brostep is sort of my fault, I took it there and now everybody else has taken it too far." And riddim is the properly documented subgenre — 140–150 BPM, "repetitive and minimalist sub-bass and triplet percussion arrangements," emerging in Croydon in the early 2010s, named from the Patois pronunciation of "rhythm." Note the triplet percussion: that's the non-duple pivot, already sitting in the genre.
3. THE DISPLACEMENT INDEX
Polymeter ▸ Metric modulation — Polymeter has no exit. Modulation is the door.
The variable that moves: The relocating pulse moves into the disagreeing lengths — and gives you a way to end the argument.
Hinge (what already agrees): Both are ways of making the listener lose the downbeat, and both are documented in the same tradition. Krebs's displacement dissonance D x+n describes a fixed offset; Reich's phasing describes the continuous version, arrived at by accident on two Wollensak tape recorders in 1965. Polymeter creates displacement by geometry; modulation resolves it by decree.
Friction (what fights): But polymeter's realignment is arithmetic and modulation's is compositional. A 27-against-16 riff realigns when the maths says so, whether or not that's a musical moment. Metric modulation happens exactly when you decide it does. So the two disagree about who is in charge — the least common multiple, or the writer.
Resolution: Use modulation to cut the polymeter short. Run the displaced riff long enough that the listener has genuinely lost the downbeat and stopped expecting a resolution — then, instead of waiting for the arithmetic realignment, pivot on a triplet and relocate the pulse so the displaced layer becomes the new beat. The riff that was fighting the meter becomes the meter. That is an exit polymeter alone does not have, and it is far more satisfying than the long-division arrival, because the resolution arrives early and from an unexpected direction. This merge is the single most useful structural idea in the guide.
Recipe
- Chassis: polymeter — 100–110 BPM, riff in 7 against steady 4/4 cymbals
- Run it long: at least sixteen bars, until the listener has stopped counting and stopped expecting resolution
- The pivot: take the triplet-quarter of the current tempo as the new beat. At 108 that gives 162 — a real 3:2 modulation, not a halving
- The reveal: on the new pulse, the previously displaced riff sits square. The thing that was wrong becomes the thing that's right
- Drums switch reference: they were the anchor, now they follow the riff. The roles invert at the pivot
- One bar of total silence before the pivot — this is the cheapest and most effective way to make the relocation legible
- Optionally reverse it later to return to the original tempo, the way Carter reverses the pivot in "Canaries"
- Avoid: halving or doubling as the pivot. A 2:1 relationship is metrical consonance and the ear hears it as the same tempo, which throws away the entire effect.
PRECISION USED TO CREATE THE SENSATION OF CHAOS
All three at once — named after your own best sentence, because it is the correct thesis.
Three mechanisms running simultaneously with one rule holding them together. Polymeter owns the phrase length: the riff cycles at seven against a drum layer that never deviates from four. Polyrhythm owns the internal division: inside that seven, the notes group in threes, and the bass plays attack points rather than subdivisions. Metric modulation owns the exit: when the displacement has run long enough that the listener has stopped counting, the pulse relocates on a triplet pivot and the riff that was wrong becomes right. And the rule that stops all of it collapsing is the one constant across every documented example in this guide — exactly one layer stays rigid. The cymbals hold quarter notes. Everything else is free to disagree, because there is something to disagree with. Your own formulation is the best statement of the goal I've read: make the instrumentation feel as if every musician has independently discovered a different groove, yet the band is so technically precise that all of those grooves eventually converge into one enormous groove. The only thing I'd add is that the convergence is more powerful if it relocates rather than merely arrives.
Recipe
- The rigid layer: cymbals on unwavering quarter notes at 104 BPM, snare on a fixed beat. This never changes and nothing is permitted to disturb it
- Polymeter (guitars): riff phrase of 7, or an unequal chain like 7-7-5-3-5 sixteenths, rolling over the bar line and landing differently every cycle
- Polyrhythm (internal): notes inside the riff grouped in threes, so length and division disagree simultaneously
- Bass as a third opinion: slapped and popped attack points contradicting both the drums and the guitars, never losing the downbeat
- Convergence points, computed and placed: full-band unison hits at the genuine realignment, not scattered near-misses
- The exit: after a long displaced section, one bar of silence, then a triplet pivot relocating the pulse so the riff sits square
- Vocals across three registers: shouted punk delivery, guttural screams, and rhythmic spoken phrases that follow the guitars rather than the drums
- Guitars gated tight so every note is discrete — displacement this dense smears if notes ring
- Dynamics: abrupt full-band stops, one clean spacious section with wide ringing chords, and total silence before each structural event
- The rule: never let two reference layers move at once. One thing is always countable, or the whole thing reads as sloppy rather than deliberate
Two more crossings
4. THE FLOOR MOVES AND THE RIFF DOESN'T
Metric modulation ▸ Polymeter — Modulate the beat while a displaced riff is mid-cycle, and the arrival point is one nobody can predict.
The variable that moves: Disagreeing lengths move into the relocating pulse — and make the pivot land somewhere unexpected.
Hinge (what already agrees): Both are about the listener's mental downbeat being revised. Metric modulation revises it deliberately; polymeter erodes it gradually. And the scholarly bridge is real — Krebs's subliminal dissonance describes conflict with a meter that is implied but not sounding, which is exactly the state a listener is in halfway through a long displaced riff.
Friction (what fights): But a modulation needs a clean pivot to be legible, and a displaced riff has no clean moments. The pivot works because the ear grabs a shared duration and holds it across the change. If the riff is mid-cycle and the listener has already lost the downbeat, there is nothing to hold — the modulation reads as a mistake rather than a move.
Resolution: Put the pivot in the drums and let the guitars ignore it. The drums state the triplet clearly and unambiguously for two bars — a legible pivot the ear can grab — while the guitars continue their displaced cycle straight through the tempo change without acknowledging it. The floor moves and the riff doesn't. So the modulation is perfectly readable because one layer executes it cleanly, and the resulting alignment between riff and beat is somewhere neither the listener nor the arithmetic predicted. Then let them converge a few bars later. This is Krebs's direct dissonance and subliminal dissonance operating simultaneously, and it is the strangest sound in this guide.
Recipe
- Chassis: metric modulation — a clear, audible tempo pivot on a triplet, 3:2 ratio, executed once and unambiguously
- The pivot lives in the drums only: two bars of clearly stated triplets, so the ear has something to hold across the change
- Guitars run their displaced cycle straight through the pivot without adjusting, acknowledging or resetting
- The consequence is the point: after the modulation the riff sits against the beat in a relationship that was not planned and cannot be predicted
- Converge three or four bars later on a unison hit, so the accident resolves and reads as intentional
- Bass chooses a side — put it with the drums so the guitars are alone in their disagreement, two against one the other way from merge 1
- Keep the guitar tone identical across the pivot so it's audibly the same riff, unchanged, in a new context
- Avoid: letting the guitars re-sync at the pivot. If they follow the drums through it, you have a tempo change and nothing more.
6. THE ARGUMENT AND THE AGREEMENT
Polyrhythm ▸ Metric modulation — Cross the grids, then take the wrong exit on purpose.
The variable that moves: The relocating pulse moves into the crossing grids — and turns a round trip into a one-way journey.
Hinge (what already agrees): A polyrhythm already contains its own modulation and simply declines to use it. Every 3-against-4 offers a choice at the resolution point: land on the shared downbeat, or accept the crossing layer as the new pulse. Polyrhythm always takes the first option. The second option is a metric modulation and it's sitting there in every hemiola ever played.
Friction (what fights): But polyrhythm is danceable precisely because it returns. The physical entrainment comes from the promise that the grids will resolve — that's why 3-against-4 is ubiquitous in Black American music and 7-against-4 isn't. Take the exit and you break the promise, and you lose the groove that made the section work.
Resolution: Keep the promise for most of the song and break it exactly once. Run genuine 3-against-4 through the verses and choruses, resolving reliably on the shared downbeat every time, so the body learns the pattern and trusts it. Then, at the final chorus, take the other exit — let the crossing three become the pulse, land at a new tempo, and never return. The groove that has been resolving all song suddenly doesn't, and the piece ends somewhere it never announced it was going. One broken promise, placed correctly, is worth more than a whole song of unpredictability — and it is the closest thing in this guide to your own line about grooves converging into one enormous groove, except the convergence relocates instead of arriving.
Recipe
- Chassis: polyrhythm — 96 BPM, deep funk pocket, live loose drums, genuine 3-against-4 between guitar and drums
- Resolve reliably through every verse and chorus. The body has to learn to trust the return before the break means anything
- Slapped funk bass on attack points, contradicting the accents without losing the downbeat
- Guitars percussive and funk-inflected: muted scratches, wah lines, sharp stabs, one real bluesy improvised solo
- The break, once, at the final chorus: the crossing three becomes the pulse. 96 becomes 144 on a 3:2 pivot
- Never return. The piece ends on the new tempo — no reversal, no resolution back to the original
- Mark it: the whole band lands together on the first beat of the new tempo, so the relocation reads as arrival rather than error
- Avoid: foreshadowing it. If you modulate early the listener stops trusting the returns and the final break costs nothing.
Keep exploring