How it works

The sound, by stage

A and B each pass the music through the stages a real system does, in the same order, and each stage does what the real product was measured doing.

1. Source
2. Amplifier
3. Speakers
4. Room
5. Your ears

1. Source

   Digital plays the recording as it is, and MP3 as an MP3 at 320, 128 or 64 kbps. Vinyl adds what a record, turntable, cartridge and phono stage each do, from their bench tests: the cartridge's tone and channel separation, the groove's distortion, surface noise and clicks, treble lost near the label, speed wobble and rumble, and the phono stage's accuracy, hiss and overload.

   Vinyl, step by step

   • Cutting the master. Bass is narrowed toward mono when the master disc (the lacquer) is cut, and the treble is held within what a record can carry: cutting systems limit how fast the cutting stylus accelerates. Ours turns the band above 4 kHz down, briefly, whenever it would pass 830 g, the setting of Neumann's 1974 acceleration limiter; real records measured by Jim Lesurf (Hi-Fi News, 2007) stay within about 1,000 g. Very bright recordings lose a little top treble, as they do on vinyl; most of ours are untouched. The release time (50 ms) is our choice.
   • The cartridge: its frequency response (how evenly it plays each pitch) and its crosstalk (left and right leaking into each other).
   • Tracing distortion. The stylus tip can't follow the groove's exact shape. The distortion is made in the groove and calibrated to Shure's measurements by stylus shape.
   • Surface noise, clicks and scratches, by the record's condition.
   • Treble lost near the label, where the groove is slower: Miller's model (as restated by CBS Laboratories, 1962) with each cartridge's tracking force. Conical tips use their own radius; for other shapes the radius is fitted to Hi-Fi World's measured inner-groove losses (our choice).
   • Mistracking. On the loudest passages the stylus can leave the groove wall. Each cartridge's limit is its measured tracking ability (µm of groove swing at 300 Hz; maker's or typical figures where none was published), checked against Hi-Fi News's test-record results; the limit at higher frequencies scales from Shure's V15 IV measurements (our choice). Most music never reaches it.
   • Speed wobble (wow and flutter), played as the speed swings each turntable's published figures imply: Hi-Fi News's weighted peak readings with the weighting taken back out, and older reviews' unweighted RMS readings as the peaks of a steady wobble.
   • Rumble and record warp, both shaped by the tonearm resonance. Each cartridge plays on an arm that suits it: the turntable's own arm when the resonance lands in the usual 8–12 Hz window, otherwise the effective mass that aims for 10 Hz, kept within our measured arms' 7–15 g (our choice, as an owner would change the arm or add headshell weight). You can switch to the stock arm to hear the mismatch.
   • Left out after testing: tracking angles. A pivoted arm reads the groove at a slight angle to the cutter's path, and a cartridge's stylus can sit at a different vertical angle from the one the record was cut at; both add a little 2nd-order distortion. With each turntable's arm at its maker's geometry, the change stayed under half a grade (our threshold for audible) on every test clip. A vertical angle 10° off the cut stayed under it too; 15° off just crossed it on one clip's inner grooves, further off than any of our cartridges measures, except one 2007 sample of the 2M Red (33°) and perhaps the Grado, whose sibling measured over 30°. So neither is modeled (the figures and the test are in data/vinyl/README.md).
   • The phono stage, from published bench tests: its RIAA error (how far it strays from the curve that undoes the record's treble boost and bass cut), its hiss, its input overload, and its 2nd harmonic where it matters (the EMT 128: 0.3% at its nominal input, growing with level). Hiss and headroom are set against each cartridge's output, through the moving-coil input for cartridges under 1.5 mV and the moving-magnet input otherwise (our choice). Only stages with that input are offered, as an owner would choose: the Fosi has no moving-coil input and the EMT 128 no moving-magnet one. The hiss is white (our choice, close to the measured spectra). Overload clips the cartridge's signal, with the clicks and surface noise, at the measured point at 1 kHz; where headroom falls toward 20 kHz, a shelf above 5 kHz (our choice) lowers it to the 20 kHz figure. The Pro-Ject RS2 is the default.

   The MP3s are one of our recordings, "When I Cross The River", encoded with LAME at a constant bit rate and decoded again ahead of time, so each lines up with the original sample for sample.

2. Amplifier

   Each amp is calibrated to its bench tests: the overtones it adds, how and where it clips, its hum and hiss, and its output impedance. The last matters with tube amps: a high output impedance lets the speaker's electrical load reshape the tone.

   Amplifier details

   • 2nd and 3rd harmonic distortion (added overtones), falling with level as in real amps.
   • Clipping (flattened peaks when the amp runs out of power) at the measured power, soft or hard, and lopsided where the real amp clips that way. Into speakers with a low impedance it clips sooner where the amp was measured running out of current before voltage (its clipping power into 4 and 2 Ω, at the speaker's typical load).
   • Hum and hiss at a fixed voltage, so they're audible through very efficient speakers (ones that play loud from little power).
   • The amp's output impedance (its internal resistance) against the speaker's measured impedance (its electrical load at each pitch), rising in the treble as measured at 20 kHz. With tube amps, the speaker reshapes the tone and distortion rises where its impedance dips. The budget Class D amp's output filter, fitted to its measured response into 8, 4 and 2 Ω, makes its top octave depend on the speaker too.

3. Speakers

   Each speaker's measured sound straight ahead and at every angle, how it behaves when played loud, the load it puts on the amp, and how its crossover times its drivers. The sound it sends to the sides counts too: in a room, that's what the reflections are made of.

   Speaker details

   • Measured frequency response straight ahead (on-axis) and at angles (off-axis), smoothed to 1/6 octave.
   • How the sound spreads: the off-axis sound drives the room's reflections and reverberation.
   • Impedance, the speaker's electrical load at each pitch, read from published curves; the amp's output impedance acts against it.
   • Loud-level behavior in four frequency bands: compression (output falls short on loud peaks as the cones near their limits, acting on each drum hit as it happens), distortion that grows with level, and the limiter of active speakers (ones with built-in amps). The distortion comes from published tests; in the deep bass it is restated against the signal driving the speaker, since below a speaker's bass roll-off a published percentage is a share of a much weaker output and can read several times too high (how).
   • Crossover timing: how far apart in time the sound from each driver (woofer, tweeter) arrives. An ordinary crossover delays the lower driver slightly; first-order and linear-phase designs don't.

   Some comparisons edit a speaker (a bump, dip or bass roll-off, another speaker's off-axis sound, or an ordinary crossover in place of a linear-phase one); the card says which.

4. Room

   You and the speakers sit in an equilateral triangle in one of six rooms, from a treated listening room and a large living room to a small bedroom. An on-wall speaker hangs on the wall behind it instead, its front as far out as its cabinet is deep, while you stay in the same seat; it was measured standing free, so the room adds what the wall does to it. Each reflection is shaped by the speaker's sound in that direction and by the surface it bounces off, the bass resonances are worked out at each ear, and the lingering sound follows the surfaces' measured absorption.

   Room details

   • The speakers are angled in toward you (toe-in).
   • Direct sound and early reflections, each shaped by how the speaker spreads its sound at that angle (its dispersion) and by the wall it bounces off.
   • Bass, where rooms boom: below 140 Hz the room plays as its resonances, blending into separate reflections by 280 Hz. Each resonance dies away at its own rate, set by the walls it bounces between (Kuttruff, Room Acoustics), and is worked out at each ear's position, so the bass differs slightly between your ears, as it does in a real room. How much of each speaker's bass goes backward follows its measurements: at each frequency its measured spread is matched to the nearest simple pattern, from equal in every direction through a cardioid's, silent straight behind (our fit), so a cardioid stirs up less of the room's bass and less of the wall behind it. Dipoles are modeled as a front and a back of opposite sign.
   • Reverberation (the sound that lingers after each note) for each octave band, set by the total sound the speaker sends into the room.
   • Air absorption, and surface materials from a published absorption database (listed in Part 2).

   The test-room standard. When audio researchers run listening tests to find small differences, they use rooms built to an international standard, ITU-R BS.1116. Such a room is mid-sized (a floor of 20 to 60 m²), its proportions fall within the standard's limits, and sound dies away quickly and evenly. In our treated room (5 × 4 m, 2.6 m high) the standard asks for a decay of about 0.2 s through the midrange, give or take 0.05 s, and up to 0.3 s longer in the deepest bass. Reflections arriving in the first 15 ms must be at least 10 dB quieter than the sound straight from the speakers. The point is a room that adds little of its own, so differences in the equipment are easier to hear. That makes it a reference, not the only good sound: many people prefer a livelier room at home. On the site, the treated listening room is built to this standard, and the Room sound section compares any room with it.

5. Your ears

   A simple model treats your head as a sphere and places each sound by its direction: the small delay between your ears and the shadow your head casts on the far ear. It is generic, not fitted to your head, so some people hear the speakers close or inside the head.

Extras in Build your own

Build your own adds what owners add to a system, each modeled on how it's commonly used:

• Moving the speakers or the seat, from the Room setup tile.
• Subwoofers take over below 80 Hz through a steep (4th-order) crossover. Each is sealed and −3 dB at 20 Hz (our choice of a typical one), and set at the main seat to be as loud from 40 to 80 Hz as a speaker in the main speakers' places. Its phase is set there too, as makers tell owners to: for the most bass around the crossover, with a 0–180° phase control and a polarity switch (our choice of a typical subwoofer's controls). Without that, a speaker whose own bass is already rolling off at 80 Hz arrives out of step with the subwoofer, and the two partly cancel. The speakers then no longer play that deep bass, so their own distortion and compression there drop out, and an active speaker's limiter reacts only to what it still plays (when it acts, it turns the subwoofers down too, our approximation); the subwoofers themselves play clean, since there are no measurements to model one.
• Room correction evens out the bass toward the neutral target, modeled on how Dirac Live is commonly used.
  How it's set

  It works below 300 Hz, as many owners of good speakers set it, and not below where the system's bass runs out (the main speakers' −6 dB point, or 25 Hz with subwoofers). It cuts peaks and boosts dips by up to 6 dB, but leaves dips deeper than 12 dB alone: they come from sound cancelling at the seat, which boosting can't fill. Those limits are our choices. It models only level, not Dirac's timing correction, and lowers bass peaks without much shortening their ringing, which only absorption does.

• Tone controls are the taste test's bass, presence, treble and air filters. They act after the amp and speakers, so a bass boost doesn't make them work harder, as it would in a real system played loud. "Use my taste result" sets them from your latest taste test.
• Added distortion puts one of four kinds ahead of the amp, like a preamp or effects box: a smooth 2nd or 3rd harmonic, crossover distortion or clipping, each at levels matched to measured gear (tube preamps, a 300B amp, analog tape, class-B amps, clipped masters; the levels and their sources).
• One speaker or two. On headphones you can play one speaker in the center instead of a pair: in loudspeaker researcher Floyd Toole's tests, listeners heard speaker differences more easily with one than with two. Some comparisons switch to one speaker themselves.

Headphones or your own speakers

Use the toggle at the top to say what you're listening on. On headphones (recommended), the site simulates the whole system, with two virtual speakers in the chosen room. On speakers, your own speakers and room play the result, so the site applies only the amplifier and the tonal difference between the chosen speaker and a reference (KEF LS50 Meta). Room choices are off, so the questions about rooms, placement, room correction, subwoofers and off-axis sound need headphones.

On headphones, Pick headphones beside the toggle corrects the tone of the pair you're wearing, so you hear the simulated system rather than your headphones. Each correction brings that model's measured response to a diffuse-field target, the ear's response averaged over all directions, since the simulation already plays the room. A and B get the same correction, and switching it on or off keeps the loudness the same. The picker charts your pick's measured response against the target: the correction closes the gap between them, with boosts limited to +6 dB (AutoEq's default, not a research figure). The measurements are oratory1990's and RTINGS', via AutoEq; permission to use them is pending.

Fair comparisons

Small differences are easy to imagine and easy to miss. These rules keep each comparison about the sound:

• Equal loudness. A louder system almost always sounds better, so we match both at the end of the chain with broadcasting's standard loudness measure (ITU-R BS.1770), measured on a second pass of the music, so the whole chain has settled, as when you listen.
• Instant switching. Both systems play all the time; A and B only choose which you hear, so no gap gives the switch away.
• One change at a time. Each comparison changes one part and keeps everything else the same: a speaker comparison doesn't change the room. Where one part brings others with it (a vinyl source needs a cartridge and turntable; an active speaker has its own amps), the card says so.
• A quiet setting. Comparisons of amps, sources, vinyl gear, response bumps and crossover timing play in the treated listening room (the test-room standard), where the room adds the least of its own. Comparisons of speakers, off-axis sound, bass extension, rooms, placement and subwoofers play in the home rooms, because how a speaker meets a room is part of what they show.
• Blind by default. In the questions, which system is A, which one the question asks about, and the order of names are random each visit.
• Questions about what you hear. Where a comparison has a clear direction, it asks about the sound ("Which one has deeper bass?"), not which system is which.
• Blind test. In Build your own, "Blind test this comparison" plays X, secretly A or B, for 10 or 16 rounds; the result shows how often guessing would do as well.
• Music you know. Differences are easier to judge in songs you know, so the questions invite you to upload some, with our recordings as the backup. A few comparisons need their own recording, for a test tone or very deep bass.
• Original recording. This button plays the music with no system or room, at the same loudness. Both systems in a comparison still share their other parts (an amp comparison keeps the same speakers and room), so neither will sound like the original; compare A with B.
• Playing level is separate from your volume. "How loud will it be played?" sets how hard the system works (70–105 dB at the listening seat), bringing out compression, clipping and distortion. Your device volume sets how loud you hear it.

Will you hear it?

Every comparison comes with an expected difference: Likely inaudible, Subtle, Clear or Obvious. It rests on one idea. For each kind of difference, published listening tests found the smallest change people notice. We call that a step and measure the gap between A and B in steps: under one step is likely inaudible, one to two subtle, two to four clear, and four or more obvious. Eight kinds of difference are measured, and the verdict is the highest rating any of them gives. The steps come from research; how they map to the four words is our choice.

It says why. Each finding names its cause where one is known: the amps' output impedance, a room's resonances, the speakers' off-axis sound, a record's surface noise, clipping, crosstalk.

And in listening terms. A tone difference also gets the words reviewers use for that part of the range, from J. Gordon Holt's Audio Glossary (Stereophile, 1993), one for the louder system and one for the quieter: "boomy" and "cool" for the midbass (40–80 Hz), "bright" and "dull" for 4–8 kHz, "tizzy" and "closed-in" above 10 kHz, and so on. It is the one published glossary we found that ties its words to frequencies; standards and research vocabularies (the Sound Wheel, EBU Tech 3286, Harman's listener training) define theirs without them. It is a reviewer's vocabulary rather than a tested one, other glossaries put some words elsewhere (Bruce Bartlett's "boomy" sits near 125 Hz), and giving each band the nearest of Holt's ranges is our choice. The words say which way the two differ, not which one is right; the charts show which is nearer neutral.

Where each step comes from

Thresholds vary a lot between listeners and studies: later tests put the decay threshold anywhere from about 3% to over 20%, and a 2023 study found it depends on each listener's hearing and experience. The anonymous tally of your guesses will show how well the ratings hold up.

What's different, and why

Under every comparison, the "What's different, and why" panel shows the charts and numbers behind what you hear.

• Live curves show what each system and its room do to the music at the listening seat, over the last second. Each compares the sound at the listening seat with the original recording, lined up on the midrange, so the music's own shape cancels out. Where the curves part, A and B sound different. A faint dashed line, also on the seat chart, marks a neutral target: flat below about 300 Hz (the recording's own bass, which room correction and subwoofers aim for), then sloping gently down, as a neutral speaker does at a seat, because rooms and ears soften the treble. Listeners prefer that slope to a curve forced flat. It comes from the Revel F328Be simulated at the same seat, in the room both systems use (the treated room when they differ).
• Room sound measures the simulated room as an acoustician would a real one, with the space measures above and how audible each gap is. Decay time uses the ISO 3382 method, against the target in the test-room standard (ITU-R BS.1116). Decay and consistency are measured at nine spots: the three seats of a sofa (0.6 m apart), and 0.3 m in front of and behind each, our choice of a typical listening area. Early reflections list the strongest in the first 20 ms and check the standard's rule that those in the first 15 ms be at least 10 dB below the direct sound from 1 to 8 kHz.
• Harmonics describes distortion as listeners describe it. It plays test tones (100 Hz and 1 kHz, at the music's loud peaks, about 12 dB above its average, our choice) through the source and amp, and the speakers too when they differ (identical speakers would add the same distortion to both and hide the difference), and shows the harmonics each system adds, plus two standard intermodulation tests (SMPTE: 60 Hz with 7 kHz; twin-tone: 19 with 20 kHz). These describe the kind of distortion; the expected difference rates how audible it is on the music.

Find your taste

Each round starts from a neutral reference and changes one thing: bass, bass tightness, treble, presence (upper midrange), air (the top octave), room sound, how widely the speakers spread their sound, and optionally tubes, vinyl and punch at loud levels.

• Four lengths. Quick taste is 7 rounds (two votes each on bass, treble and room sound, plus one identical pair) and gives leanings, not a tested profile. Standard adds presence, then spends up to 16 rounds in all on the traits you keep choosing the same way: each can get up to five votes, and five out of five is what "clear" takes. Sound only and Sound and gear run the full test. Each length can carry on into the next, keeping your votes.
• Your own music. Each song in your playlist plays in the rounds it suits, and our recordings fill the rest; ★ in the music menu marks the one we chose.
• Saved as you go. Answers are kept in this browser after every round, so you can Save and exit and resume later.
• Your sound, and how to get it. The results open with one sentence that separates clear preferences from slight ones, then show each trait between its two ends, your votes, and how often random picks would split that way. Under each preference are the EQ setting you chose (the test's own setting, which Build your own's tone controls can apply), the speakers here that measure furthest that way, and for room sound and bass tightness the room changes that move toward it.
• Share and compare. A link carries your profile, and only your profile. A friend who opens it sees it beside their own, with the traits where you differ.
• Listen longer. An optional six-minute step plays your sound and its opposite, three minutes each without switching, then asks which you'd rather keep.

To keep the result honest:

• Sides are random. With our recordings, each trait is heard on different music; with your playlist, your songs take turns.
• You can answer only after both A and B have played for at least a second (our choice), so every vote compares the two.
• Some pairs repeat with the sides swapped, to measure consistency; a few are identical, to catch hearing differences that aren't there.
• A trait counts as a preference only when your votes pass a statistical test; traits that lean one way without passing get extra rounds.
• The room rounds change only reverberation and reflections. Livelier rooms also change tone (about 7 dB more bass around 63 Hz and up to 3.4 dB less treble at the listening seat than the treated room), so these rounds match the rooms' bass with room correction (within 2 dB, though below about 40 Hz the livelier rooms stay lower) and their treble with a gentle lift (within 0.1 dB). That way, liking more bass or less treble isn't counted as liking a bigger room.
• A final check plays your predicted favorite against alternatives.

Recommendations match your profile against traits computed from each speaker's and room's measurements, never hand-labeled. The same traits label the options in Build your own, relative to the others here, not to an absolute standard.

Where the numbers come from

One rule runs through the whole lab: every number comes from a measurement, a published standard or a published study. Where a model needs a value nobody has published, such as a band edge or a cutoff, we pick one and say "our choice", so you can tell evidence from judgement. The references list 112 sources: 89 we opened and used, and 23 published methods we cite but haven't yet re-checked (marked in the list).

Speakers

27 of the 32 were measured on a Klippel near-field scanner, a robotic rig that measures a speaker up close from many angles and removes the room's effect: 12 by Audio Science Review, 14 by Erin's Audio Corner, and the Cadentia 3 by its designer. Princeton's 3D3A Lab measured the Quad ESL-57 in an anechoic (echo-free) chamber. The DeVore O/96, Klipsch La Scala AL5 and Pi Speakers' four π have no measurement files anywhere, so their curves are traced from published graphs: Stereophile's and Hi-Fi News's for the O/96, Stereophile's for the La Scala, which stop at 60° off axis, and the maker's own for the four π, which stop at 50° (beyond those, each follows a similar measured horn speaker, our choice). KEF measured the Q4 Meta in its own anechoic chamber but published only the standard summary curves, not the sound at each angle, so its angles follow the measured LS50 Meta, which shares its Uni-Q driver design, scaled at each frequency to match KEF's sound power (our choice); its early reflections then match KEF's within 0.3 dB on average. Loud-level compression and distortion, impedance and crossover timing are read from published charts and reviews. The gear page also gives each speaker Harman's predicted preference rating (Olive, 2004), worked out from the measurements at full resolution and checked against the figures spinorama.org publishes. Compare shows each score with the model's typical error (the residual error of 0.8 points Olive reports) and brackets scores less than 1.1 apart as too close to call (the two errors combined, our arithmetic; calling that a tie is our rule). Olive's 70 test speakers included horns, waveguides and active monitors, nearly all forward-facing, but only one dipole, and he warns the model may not cover unusual directivities such as omnidirectional speakers, so dipoles and speakers that fire sound around the room are marked as rough. In its Compare charts, any speaker can be shown with the simulation's subwoofer added: its tone combined with the speaker's own response the way the simulation plays them, and its loud-level charts without the deep bass it no longer plays. Notes with every source: loud levels, impedance, crossover timing, traced graphs.

Amplifiers

Seven from John Atkinson's bench tests in Stereophile (we use the numbers; the graphs stay on their site), the Sansui 7070, a 1976 receiver, from Paul Miller's lab report in Hi-Fi News, and the First Watt F2 from its maker's specs, as we found no independent test. Every amp number with its source page.

Vinyl

Bench tests from reviews in Hi-Fi News, Stereophile, Stereo Review and HiFi/Stereo Review (Hirsch-Houck Laboratories), High Fidelity (CBS Laboratories and United States Testing Co.), The Absolute Sound, LowBeats and HomeTheaterHifi, and Audio Science Review for one phono stage; Shure's research on tracking and tracing; makers' specs where no test was found. Every cartridge, turntable and record number with its source page.

Rooms

Each surface's absorption comes from the open pyroomacoustics database (table below). The treated room is designed to meet ITU-R BS.1116; the home rooms' background noise is the average Bradley measured in 602 Canadian living rooms.

Hearing

What people can hear comes from listening research and standards: Toole and Olive on tone, ITU-R BS.1387 (PEAQ) on distortion and MP3, ISO 226 on the quietest sounds, Larsen, Okano and ISO 3382 on room sound, and more. Will you hear it? gives each.

The label on every part

Every speaker, amp, cartridge, turntable, record, phono stage and room carries a label saying where its numbers come from. Across the 71 parts:

Presets closely approximate the named products but aren't exact replicas: a review usually measures one sample, and we smooth the curves (see What it leaves out).

How we check it

Checks we run on the code test the model against its own sources:

• The data can't drift. Every copied measurement file is checked against the fingerprint (SHA-256) recorded when it was copied, and the room materials against the original database.
• The treated room meets its standard. Its size, proportions, decay at every octave and nine seats, and early reflections are checked against the BS.1116 rules.
• Speakers spread their sound close to as measured. The simulation estimates how much sound each speaker sends into the room from its horizontal and vertical measurements. For the nine speakers whose scanner files include the measured total, ours always reads a little more directional: within 1 dB for about two-thirds of the octave bands from 250 Hz to 8 kHz, and at most 3.3 dB (the Magnepan at 8 kHz).
• Speakers compress as measured. The compression reproduces the measured loss at the measured level, acts within a single drum hit, and adds no overtones of its own.
• The distortion grade behaves. An unchanged signal grades imperceptible, more noise or deeper clipping grades steadily worse, and a smooth curve grades better than a step at zero with far less THD, as Lee and Geddes's listeners found (more above).
• The verdict follows its sources. The tone, noise, wobble and treble-room rules give the figures their sources imply, and the verdict names the right cause for each kind of comparison.
• The taste test finds real preferences, not imagined ones. Simulated listeners with known preferences check that the scoring recovers them and doesn't invent them from random answers.

What it leaves out

A simulation is only as good as what it includes. What this one doesn't do, in one place:

• Exact products. Each preset is built from published measurements, usually of one sample. Speaker measurements are smoothed to 1/6 octave, which hides very narrow features.
• Your head. The head model is generic, so some people hear the speakers close or inside the head, and the width measure is approximate.
• Everyone's hearing. The steps are averages from published tests; individual listeners vary a lot, and most room-sound thresholds come from concert halls rather than living rooms.
• Every kind of distortion. The distortion grade may be too lenient on crossover distortion, and it can't judge any effect of distortion on the soundstage.
• The deepest bass in rooms. The absorption data start at 125 Hz, so real rooms with thin absorbers may ring longer in the deepest bass. Each resonance's damping uses each wall's average absorption, so a bass trap counts the same mid-wall as in a corner, where the resonances are strongest and it works better.
• The exact amount of sound sent into the room. It is estimated from horizontal and vertical measurements, and reads up to 3.3 dB more directional than the scanner's own total (How we check it), so the lingering sound may be slightly too quiet.
• Treble in the reflections. It is scored only near the one reference level and band where it has been tested.
• Not modeled at all:
  • speakers heating up over minutes of loud playing (where it has been measured it was small);
  • how each turntable drive type (belt, idler, direct) holds speed against the stylus's drag: we found no published measurement to base it on;
  • a phono stage's loading of the cartridge, and step-up transformers;
  • room correction's timing correction;
  • diffusers: reflections are mirror-like, and after the second bounce they join the decaying tail;
  • the drivers' own timing, when two different speakers are compared;
  • the strain a tone-control bass boost puts on the amp and speakers.

Every part and source

What each part is based on

Every cartridge, turntable and record number with its source page: data/vinyl/README.md.

Room materials

Absorption coefficients per octave band (0 = fully reflective, 1 = fully absorbing). In grey, where the database stops at 4 kHz: the model repeats the 4 kHz value.

Speaker measurement licenses

Audio Science Review LLC holds the copyright in its data, which we use under CC BY-NC-SA 4.0 (non-commercial). We smoothed it to 1/6 octave, converted it to relative responses, and share the result under the same license. The Quad ESL-57's measurements are the 3D3A Lab's at Princeton University, under CC BY 4.0. The Erin's Audio Corner data (13 speakers, plus most of the distortion and compression figures) and Audio First Designs' data (the Cadentia 3) have no published license; permission is pending, and we don't republish their measurement files.

References

Privacy

Your own music never leaves your browser. It is kept here between visits until you choose "Forget my songs" or untick "Keep my songs in this browser". Your guesses, blind-test scores and taste results, including Listen longer, go into an anonymous tally of answers, systems compared and listening conditions, with no personal data. For the taste test that includes how long each answer took, how often you switched between A and B, and how long each side played: we're collecting these to learn whether quick answers are the more consistent ones, and they don't change your result; your songs appear only as "own", or, in the blind test, as one of our recommended songs when your file matches one. Your progress and latest taste result are kept in this browser.