Every feature, in detail

This is the full specification of Discata 1.0.0 for iPhone: what each part of the sound chain does, the ranges of every control, and where the measurements behind the rooms and headphone profiles come from. Everything is processed on your iPhone.

The Tone editor with numbered points on a response curve, and an Auto level switch.

The sound engine

Discata has its own native audio engine. Every effect runs on the iPhone, in a fixed order, and the delay through the chain is the same whichever effects are on.

Signal path32-bit floating-point audio. Equaliser filters are calculated and run with 64-bit precision.
Order of processingTone level → tone EQ → stereo width → space (crossfeed, room or Orbit) → correction level → headphone correction EQ → Sound On / Original mix → Discata volume → true-peak limiter.
Why that orderTone shapes the music before it is placed in a room. Headphone correction runs after the room, so it corrects the headphones without bending the spatial cues.
Processing rateThe sample rate of the current output. Files at another rate are converted by iOS before processing.
Constant delayEvery mode is time-aligned, so switching a mode, or switching to Original, never shifts the music in time. The Output screen shows the delay in milliseconds.
Smooth changesLevel changes ramp over 10 ms, and changes of mode crossfade over 30 ms, so nothing clicks.
Real-time safetyThe audio thread never allocates memory, takes a lock or waits.
BatteryThe engine goes to sleep 10 seconds after you pause, and digital silence is skipped instead of processed.
Engine checkThe About screen can run a test tone through the engine and show the calculated and the measured gain side by side.

Tone equaliser

Tone is for taste. Use the 10-band graphic editor for quick changes, or the parametric editor for full control. The response curve is drawn as you edit, and you can drag and pinch the points on it.

Parametric bandsUp to 32, each with its own on/off switch.
Filter shapesPeaking, low shelf, high shelf, low pass and high pass.
Frequency20 Hz to 20 kHz in the editor.
Gain−18 to +18 dB per band in the editor, in 0.5 dB steps.
Width (Q)0.1 to 20 in the editor.
Imported profilesAccepted from 1 Hz to 20 kHz, −30 to +30 dB and Q 0.05 to 50.
Graphic EQ10 bands at 31.25, 62.5, 125, 250, 500 Hz, 1, 2, 4, 8 and 16 kHz, each −12 to +12 dB in 0.5 dB steps.
Overlap compensationNeighbouring graphic bands overlap, so Discata adjusts them to hit the level you set at each frequency.
Auto levelKeeps the loudness the same when the EQ is on, so you hear the tone change, not a volume change. A Level trim of −6 to +6 dB sits on top.
Manual preampWith Auto level off, −30 to +12 dB.
Filter designAudio EQ Cookbook biquad filters.
PresetsSeven: Reference, More bass, More detail, Clearer voices, Quiet listening, Immersive and Less harsh. Each has a description in plain words.
Sound stylesNine one-tap starting points: Balanced, Reference, Bass, Treble, Immersive, Speakers, Vocals, Chill and Orbit.
ImportAutoEq and Equalizer APO ParametricEQ.txt files, GraphicEQ.txt files (followed with 31 third-octave bands) and Discata JSON.
ShareAny profile as ParametricEQ.txt or as Discata JSON.

In the app: Sound › Tone

The Tone screen with a response curve and a list of presets: Reference, More bass, More detail, Clearer voices and Quiet listening.
Tone presets, each explained in plain words.
The Tone editor with numbered points on a response curve, and an Auto level switch.
Drag a point and hear the change as you make it.

Headphone correction

Correction is about your headphones, not your taste. It is a separate equaliser stage with its own 32 bands, so a full correction profile and a full tone profile can run together.

Profiles8,849, from the open AutoEq project, stored in the app.
SearchBy headphone model, offline.
SuggestionsDiscata suggests profiles from the name of the headphones you connect. Nothing is scanned and nothing is sent anywhere.
Per outputYour choice is remembered for each output, so each pair of headphones gets its own correction.
Own stageUp to 32 bands, separate from the 32 tone bands, with its own Auto level.
Your own filesImport a correction file in the same formats as tone profiles.

In the app: Sound › Headphones

The Headphones screen: a headphone correction switch, a correction curve and a search field with AirPods Pro results.
Correction for the pair you are wearing.

Space: 8D rooms and crossfeed

Space moves the music out of your head. Choose one mode at a time: Off, one of three crossfeed strengths, Room or Orbit. These effects are made for headphones.

8D audio: six measured rooms

In Discata, 8D audio means music that comes from around you, and it is built on rooms. The rooms are not synthetic reverb. Each one is a set of impulse responses measured with a dummy-head microphone in the broadcast studios of WDR in Cologne, for 72 directions of the head, one every 5°.

RoomMeasured inSound sourceLength
Control roomWDR Control Room 1Studio monitors, close by0.3 s
Mix roomWDR Control Room 7Studio monitors in a larger control room0.4 s
Studio hallWDR Small Broadcast StudioPA speakers with a centre speaker0.85 s
ChamberWDR Small Broadcast StudioOmnidirectional source on stage0.9 s
Concert hallWDR Large Broadcast Studio (Klaus von Bismarck Saal)PA speakers with a centre speaker1.8 s
Orchestra hallWDR Large Broadcast Studio (Klaus von Bismarck Saal)Omnidirectional sources on stage1.9 s

Length is how much of each measured response Discata uses.

Room controls

Room reflections0 to 100 %. 100 % is the room as measured; lower keeps the speakers but makes the room drier.
Speaker spread−15° to +15° in 1° steps, from the room’s measured directions.
Clean bassBass below a crossover skips the room’s reflections, so it stays tight. The crossover is a 4th-order Linkwitz–Riley filter, 60 to 200 Hz in 10 Hz steps, 120 Hz by default.
Head trackingWith AirPods Pro, AirPods Max and some Beats, the speakers stay put when you turn your head. Recentre sets a new straight ahead.
Virtual surround0 to 100 %. Steers the voice to a centre speaker and the ambience to speakers beside and behind you. Nothing is invented: it is all in the recording.
Height speakersPart of the ambience comes from above you, at 45° elevation.
ExperiencesFive one-tap setups: Studio, Front row, Club, Concert hall and Late night. Everything stays adjustable afterwards.
Loud mastersA slow headroom control keeps loud, dense masters from leaning on the limiter.
Your own roomImport a HeSuVi impulse response file. Spread, head tracking and virtual surround work with the bundled rooms.
DiagramThe Space screen draws your settings as speakers around you. It is a picture of your settings, not a measurement.
ConvolutionPartitioned FFT convolution, using Apple’s Accelerate framework.

Crossfeed

StrengthSettingCut-offLevel of the crossfed signal
GentleJan Meier650 Hz9.5 dB below the direct signal
MediumChu Moy700 Hz6 dB below the direct signal
Strongbs2b default700 Hz4.5 dB below the direct signal

Orbit, a motion effect

MotionThe whole mix circles around your head. One circle takes 2 to 30 seconds; 8 seconds by default.
Stereo is keptLeft and right travel together as two sources 60° apart, so the stereo picture survives while it moves.
How it is madeLevel differences between the ears, a time difference of up to 0.66 ms, and filtering for the shadow of the head and for sounds behind you.
What it isA creative effect, not a room.

Stereo width

Stereo width runs from 0 to 200 %. 100 % is the original, 0 % is mono, and above 100 % widens the image.

In the app: Sound › Space

The Space screen: a diagram of speakers around a listener, experiences such as Front row and Club, modes including crossfeed, Room and Orbit, and the six rooms.
Speakers around you in a measured room.
Space controls: speaker spread, a Clean bass switch with a 120 Hz crossover, a Head tracking switch and a Virtual surround slider.
Clean bass, head tracking and virtual surround.

Room measurements: “A Spatial Audio Impulse Response Compilation Captured at the WDR Broadcast Studios” by Philipp Stade, Benjamin Bernschütz and Maximilian Rühl, Technische Hochschule Köln (Zenodo record 3820044, CC BY-SA). Height speakers: SADIE II database, University of York (Apache 2.0).

Loudness and output

Discata measures how loud your music is and can even it out without squashing it. A limiter at the very end catches the peaks that EQ boosts, rooms and crossfades can cause.

NormalisationOff, Track or Album. Track evens out shuffles and playlists; Album keeps the loud and quiet songs of an album as the artist intended.
MeasurementOn the iPhone after import, to ITU-R BS.1770-4. ReplayGain and R128 tags in your files are read too.
Target level−14, −16 or −18 LUFS, with a target adjustment of −12 to +12 dB in 0.5 dB steps.
Prevent clippingHolds back the gain for songs whose peaks would otherwise distort.
LimiterLook-ahead true-peak limiter. Peaks are detected on a 4× oversampled signal, so peaks between samples are caught too.
Limiter timing5 ms look-ahead, with a fast release (50 ms) for single peaks and a slow one (400 ms) for sustained loud passages.
Ceiling0 to −12 dB in 0.5 dB steps, −1 dB by default. The limiter can be switched off.
Limiter read-outThe Output screen tells you when the limiter has recently reduced peaks, and by how much.
Discata volumeA separate volume for Discata, applied before the limiter.
Output factsThe Output screen shows the output’s name and type, the processing rate, the processing delay, the output delay and the number of dropouts this session.

In the app: Sound › Playback and Sound › Output

Sound On / Original

One switch compares everything you have set up with the file as it is.

What Original bypassesTone, headphone correction, space, width and the limiter.
Level-matchedOriginal is matched to the loudness of the processed sound, by up to 12 dB either way, so you judge the sound, not the volume.
Time-alignedBoth paths have the same delay, and the switch crossfades in 30 ms, so the music does not jump.
WhereAt the top of the Sound tab and in the player.
Sound setupsSave tone, correction, space, width, volume and limiter together under a name, and switch between setups in one tap.

Playback

Discata plays the music files you own, offline.

FormatsFLAC, ALAC, WAV and AIFF (lossless); MP3 and AAC/M4A (lossy).
DecodingBy the decoders built into iOS.
ChannelsStereo and mono. Files with more than two channels are mixed down to stereo.
GaplessWith crossfade off, songs play back to back with no gap.
Crossfade0.5 to 12 seconds in 0.5 second steps, equal power. “Keep albums gapless” leaves consecutive songs of the same album untouched.
Skip silenceRemoves long silent stretches at the start and end of songs. Threshold −80 to −30 dB (−50 dB by default), for silences of at least 0.5 to 5 seconds (1 second by default). Fades and short pauses are kept.
QueuePlay Next, Add to Queue and remove from the queue.
Smart ShuffleLearns on your iPhone from what you finish and skip. It avoids the same artist twice within 3 songs and the same album twice within 5, and brings back songs you have not heard for 30 days. It can be switched off, and the history can be cleared.
Sleep timer15, 30, 45, 60 or 90 minutes, or the end of the current song.
In the backgroundBackground playback, with controls on the lock screen and in Control Centre.
Quality badgeCodec, bit depth and sample rate, measured from the file itself, never guessed from its name.

Library

Your library lives on your iPhone and needs no account.

Adding musicImport files from Files or iCloud Drive; Discata keeps its own copy. Or link a folder and keep it in sync.
ViewsSongs, Albums, Artists and Playlists.
SearchOne field searches titles, artists and albums.
FiltersFavourites, Lossless, Hi-Res, Compressed, added in the last 30 days, and by artist or album, with a choice of sort order.
Hi-ResShown for lossless files above 16-bit or above 48 kHz.
Multi-selectPlay, queue or add many songs to a playlist at once.
PlaylistsReorder, pin, search within a playlist, import files straight into one, and give each its own artwork from Photos or Files.
ArtworkFrom the file, from an image in the same folder, or from an opt-in online look-up.
Your ListeningTop artists, albums and songs for a period you choose, with a text summary you can share. The history stays on your iPhone.
Demo songA 41-second public-domain Beethoven excerpt, so you can try the sound before importing anything.
The Discata library: a song list with artwork and quality badges, a search field, and a player bar above the Library, Sound and Appearance tabs.
Songs, albums, artists and playlists, with quality badges.

Lyrics

Lyrics follow the song, line by line and word by word.

SourcesLyrics inside the file, a .lrc or text file you import, or an opt-in online look-up where you choose from the results.
Line timingThe current line is highlighted from the times in the lyrics.
Word timingExact when the lyrics carry word times (enhanced LRC); otherwise estimated within the line.
InterludesLong instrumental gaps are shown, so you know the lyrics have not stalled.
Bluetooth delayCompensated automatically, so lyrics stay in step on wireless headphones.
Untimed lyricsShown as plain text.
Now Playing showing synced lyrics with the current line highlighted, and the Original switch at the top.
Synced lyrics that follow the song.

Interface and customisation

Discata is designed to be read at a glance and shaped to your taste. Almost everything you see can be changed, and nothing you change touches the sound.

ThemeAuto, Paper (light), Ink (dark) or OLED (true black).
Colour presetsFive: Listening room, Midnight, Vinyl, Studio and High contrast, each in light and dark.
Your coloursPick an accent from swatches or enter any hex colour, and add a background tint.
Readable by designIf a colour you choose would be hard to read, Discata adjusts it to keep text contrast and tells you what it changed.
Artwork coloursThe player can take its colours from the song’s cover, and the whole app can take a gentle tint from it.
GlassA slider sets how translucent the glass surfaces are.
FontsSF Pro, Inter or Roboto.
Library layoutCompact, Regular or Roomy rows; 2, 3 or 4 grid columns; a choice of what the second line shows; quality badge and song length on or off.
Player layoutLarge or medium artwork, and each button (lyrics, queue, sleep timer, output, favourite) can be shown or hidden.
Motion and hapticsMotion like iOS, full or reduced; haptics on or off.
Saved looksSave up to 12 complete looks and switch between them.
AccessibilityFollows the iOS Reduce Transparency, Increase Contrast and Reduce Motion settings. Controls carry VoiceOver labels, including the points on the EQ curve.
First runA setup you can skip, with a demo song and one-tap sound styles. You can run it again from About.
HelpA Listening Guide inside the app, and Report a Problem, which opens an email you see before sending.

In the app: Appearance

Now Playing in Discata: album artwork, the Sound On switch, a seek bar, playback controls and a FLAC 16-bit 44.1 kHz quality badge.
Now Playing: artwork first, with Sound On one tap away.

Privacy and price

AccountNone. There is nothing to sign in to.
On your deviceAll sound processing, your library, playlists, settings and listening history.
Online look-upsOff until you allow them. They send only a song’s title, artist, album and length to MusicBrainz, the Cover Art Archive and LRCLIB.
PriceFree. The free version shows one short ad card after about every 7 minutes of listening, only between songs. You can close it after 5 seconds, and the music keeps playing.
Ad-free foreverA one-time purchase of £9.99 in the UK, not a subscription. It works on all your devices signed in to the same Apple Account.
RequiresiPhone with iOS 17 or later.

See the Privacy Policy for how ads and your privacy choices work.

What Discata does not do

For the thinking behind each setting, read the listening guide, or see the pages on the equaliser, 8D audio and rooms, lossless files and the offline player.

Hear it with your own music

Download Discata free from the App Store and bring in the music you already own.

Download on theApp Store

Free on iPhone. Requires iOS 17 or later. No account.