Why digital amp sims feel flat.

The number printed on the back of your amp is a label. It is not a measurement.

Most guitarists treat speaker impedance the way they treat a USB plug: there is one correct hole, and any other hole is wrong. Sixteen-ohm cab, sixteen-ohm jack. Eight-ohm cab, eight-ohm jack. The label is the rule and the rule is the label.

The label is a fiction.

A speaker is not a fixed resistor. It is a coil of wire wrapped around a magnet, suspended in a cone, sitting in a chamber of moving air. Its impedance is a curve. The curve climbs at the cone's resonance, dips through the midrange, and rises again at higher frequencies. Eight ohms nominal means the curve sits around eight ohms across most of the audio band — but at the resonant peak it might hit thirty, and at the cabinet's low-frequency dip it might drop close to five.

That curve is what your amp's output stage actually sees.

And the curve does not even hold still. The cone's motion generates a reverse current that flows back into the amp and changes the load the amp sees from one instant to the next. The conversation between an amp and a speaker is non-stationary. It changes with what you are playing.

The rest of this essay is about what that means.

Three ways to wire multiple speakers, three different loads for the amp to drive.

Parallel. Connect each speaker positive-to-positive and negative-to-negative. The voltage across each is the same — they all see the full amplifier swing — but the current divides. The total impedance drops below the smallest individual speaker. Two eight-ohm speakers in parallel land at four ohms. Four sixteen-ohm speakers in parallel land at four ohms also. Add speakers and the amp works harder; the load gets harder to drive.

Series. Connect them in a chain — positive of one to negative of the next, only the first and last touching the amp. The current is the same through every speaker; the voltage divides between them. The total impedance is the sum. Two eight-ohm speakers in series land at sixteen. Add speakers and the load gets easier to drive, but each speaker sees a smaller share of the voltage and therefore less power.

Series-parallel. Two pairs in series, then those pairs in parallel — or two pairs in parallel, then those in series. Net load is the same as one speaker. Four sixteen-ohm Celestion Greenbacks in a Marshall four-twelve, wired series-parallel, present a sixteen-ohm load. Four eight-ohm speakers in the same cab present eight. Every classic four-twelve runs this scheme.

The amp's tolerance for being given the wrong total load is real and not zero. Anything within roughly half to twice the nominal tap of a tube amp is broadly survivable; below half is where amps die. That is the safety floor. The calculator below covers it.

The interesting question is everything above that floor: given two loads that are both safe, what does the amp actually do with each one?

Plan your wiring below. The calculator handles the arithmetic for any combination of two to eight speakers in parallel, series, or series-parallel, and flags the total against the amp impedance you tell it about. It will keep you above the safety floor — which is the part most articles on this subject stop at. The rest of this piece is the part the calculator does not cover.

The calculator is the floor. Two loads that both pass it are still two different loads. What follows is what that difference does to your sound.

Pick any guitar speaker that ships in multiple impedance variants — Celestion Greenback, Eminence Swamp Thang, WGS ET-65, Jensen C12N — and the catalogue presents them as a spec choice. Eight ohms or sixteen, your call, depending on what your amp wants. The cone is the same, the magnet is the same, the basket is the same. Only the voice coil is different.

Only the voice coil. That is what is meant to sound the same.

Run the measurements and they are not the same speaker.

The Eminence Swamp Thang in eight ohms has a resonant frequency around ninety-seven hertz. The same speaker in sixteen ohms resonates at one hundred and thirteen — sixteen hertz higher. Its voice-coil inductance is sixty per cent greater. Its motor strength — the magnetic force per ampere that grips the cone — is fifty per cent greater. The mechanical mass is the same. The cone is the same. The voice coil is what changed, and the voice coil is what couples the amp to the air.

The same speaker model at four, eight, and sixteen ohms is not the same speaker.

The mechanism is not the impedance number. The number is the label. The mechanism is winding architecture. To deliver a higher rated impedance you wind the voice coil with smaller wire and more turns. Smaller wire raises DC resistance, more turns raise inductance, and the two together change how the speaker responds at high frequencies — inductance rises with frequency, and inductance in series with a tube amp's output transformer creates a voltage divider that lifts the top end. More turns in the same gap also increase the motor strength, which is what determines how firmly the cone is gripped by the magnet at any given current. Motor grip is where compression character, sag, and breakup live.

Premier Guitar's standard answer to can the same model of speaker in different impedances sound different is yes, but probably not substantially. That answer is incomplete. The differences are not subtle once you know what you are listening for: in a tube amp without negative feedback, the sixteen-ohm version has more chime and air at the top, and the eight-ohm version shoves harder in the midrange. The amp is the same. The cone is the same. The voice coil moved everything.

Cabinet construction matters more than the voice coil. Speaker model choice matters more than the voice coil. The voice coil is not the biggest variable. It is, however, the variable the catalogue lies about.

Stop thinking of impedance as a single number and the picture clarifies.

A tube amp's output transformer has a particular characteristic impedance on the primary side — the load the output tubes are designed to drive. The secondary side connects to the speaker, and the impedance the tubes see is the speaker's impedance multiplied by the square of the turns ratio. Higher secondary load, higher primary load. Lower, lower. So far, the standard story.

Now play a note.

The cone moves. As it moves, the voice coil cuts through the magnet's field and generates a voltage of its own — back-EMF — that flows back into the amp as a reverse current. The amp's output stage now sees not just the speaker's impedance curve but that curve plus the back-EMF the cone is generating right now, which depends on how fast the cone is moving, which depends on what you played a millisecond ago. The load is no longer a curve. It is a curve being modulated, instant by instant, by the cone's own motion.

The amp responds. Its output voltage follows what the load is doing. The transformer's behaviour — leakage inductance, winding capacitance, core saturation — interacts with all of this differently depending on signal level. None of it sits still. The whole electrical system between your pick and the air is in conversation with itself, and the conversation is non-stationary.

Most of this is too technical to be useful to a player. What is useful is the consequence: the amp's tone changes with what you are playing, because the load the amp is driving changes with what you are playing. That is not a metaphor. That is what is happening in the wire.

Guitarists describe what they hear from real amps as having a 3D quality — depth, dimension, life on either side of the note. What they are describing is the sound of a system in conversation with itself.

An impulse response is a snapshot. You play a single click into a microphone in front of a speaker that is being driven by an amp at one volume, capture the response, and the resulting file describes how that system shapes audio. Convolve any signal with the IR and you get what the signal would sound like through that system at that operating point.

The mathematics requires the system to be linear and time-invariant. An IR cannot describe a system whose load shifts with the signal driving it — by construction, it is a snapshot of a system that does not sit still.

The amp-speaker conversation does not sit still. It is modulated by back-EMF, by transformer saturation, by the cone's motion. The simple-IR approach to cab simulation is, mathematically, the wrong tool for the job. The math is right. The problem is the wrong shape.

The high-end products know this. Suhr's Reactive Load is built explicitly to handle the reactive part of the load separately from the IR that handles the spectral shape; the marketing line is it feels like you are still connected to a cabinet. Universal Audio markets the OX as Dynamic Speaker Modelling, claiming to capture speaker breakup, drive and cone cry. Fractal's Dyna-Cab includes a speaker-impedance-curve value and a reacting transformer model. These are real attempts at the actual problem.

Two Notes uses the word dynamic in their DynIR product, but their dynamic is mic-position interpolation between many fixed IRs — not back-EMF, not load reactance. The product is good at what it does. What it does is not what the word suggests.

None of the products fully closes the gap. The 3D quality is hard to capture because the conversation is genuinely complex. The flat feeling guitarists report when comparing modellers to tube amps is not snobbery and it is not nostalgia. It is the sound of a snapshot played back where a conversation should be.

Treat impedance as a tonal lever, not a safety match. Pick the cab impedance that gives you the breakup character and frequency tilt you want for the amp you are playing. The calculator above keeps you above the floor; the rest is a musical decision.

Try the same speaker in two impedance variants if you have the chance. The eight-ohm version is not the sixteen-ohm version. Your ears will tell you which one your amp wants. The catalogue will not.

And when a digital sim sounds flat to you next to a real amp, do not assume your ears are wrong. They are hearing the gap between a snapshot and a conversation. The 3D quality guitarists hear in real amps is the sound of a system in conversation with itself — and a snapshot of that system, however good, will always be a snapshot.