02 - Catalogue · 2026 / 製品 / Two products

Two products, built without compromise.

Analog Realism keeps a deliberately small catalogue. Each release is engineered to be useful for years, not seasons. Sphinx 101 for the master bus, and Sphinx Channel for every track.

products / sphinx-101 / v1.6.9 · 2026

Sphinx 101

Master bus processor. Component-accurate analog modeling with TrueRail Technology. Three main circuits - SLL, Nevy, Amok - with thirty-four analog modeling mechanisms tuned to the harmonic and dynamic signatures of the modeled console classes. Added with known hardware circuits - Pultey, Nevy, SLL, Amok and Maney - for EQ, Filter and all Dynamic Modules.

Get Sphinx 101 69€ + tax Get Both ··· + tax Try Sphinx 101 · instant delivery · 2 seats · one-time purchase
Format AU · VST3
Platform macOS 11+ · Windows 10+
UsageMaster Bus Processor
CircuitsSLL · Nevy · Amok
TransformersM1166 · C9049 · K1166 · L1544
Mechanisms 34 modeled
Seats 2 per serial
Telemetry never
Code language JUCE 8 and C++
01 / Screens

Two skins. Different vibe.

  • 01 - Nevy skin / full panel
  • 02 - Amok skin / full panel
  • 03 - Signal analysis
02 / Feature highlights

Thirty-four mechanisms. Acting as one circuit.

  • Transformer physics
  • 01

    Jiles-Atherton flux-integrated hysteresis

    The core is driven by integrated flux (V/f law), producing frequency-dependent saturation that matches real iron.

  • 02

    Four core material models

    Distinct saturation curves, permeability, coercivity, and remanence per material class.

  • 03

    Winding asymmetry

    Even-order harmonics whose magnitude varies with transformer model.

  • 04

    Bias-DC flux walk

    Program-dependent DC magnetisation offset producing session-responsive even harmonics.

  • 05

    Remanence

    The core retains a fraction of its magnetisation during silence, starting each new phrase from a non-zero magnetic state.

  • 06

    Bertotti dynamic core loss

    State-domain eddy current and excess loss terms that resist rapid flux changes like physical friction in the iron.

  • 07

    Copper winding loss

    Signal-dependent, with thermal DCR rise over the session.

  • 08

    Leakage LC resonance

    From leakage inductance and winding capacitance, per model, entering the measurable band at high sample rates.

  • 09

    Gated demagnetisation

    Signal-presence-gated demagnetisation with remanence target.

  • 10

    Inter-transformer coupling

    Inter-transformer magnetic coupling through chassis proximity.

  • Power supply
  • 11

    Shared rail sag

    Asymmetric attack and release, soft-clamp asymptotes on seven consumer sites.

  • 12

    Reservoir ripple

    Conduction-gated, with rectifier harmonics and mains frequency selection (50/60 Hz).

  • 13

    Differential per-channel sag

    Differential per-channel rail sag under asymmetric stereo.

  • 14

    Series-pass regulator

    Series-pass voltage regulator model with per-topology PSRR.

  • 15

    Ground-return injection

    Signed injection carrying the signal fundamental through per-topology chassis impedance.

  • Active electronics
  • 16

    Ebers-Moll BJT

    Class-AB crossover emphasis and thermal junction bias shift.

  • 17

    Koren vacuum tube

    Grid-current blocking at millisecond charge rate and hundred-millisecond discharge.

  • 18

    Child-Langmuir diode detector

    Child-Langmuir 3/2 power-law diode detector.

  • 19

    VCA feedthrough and crossover

    Control-voltage feedthrough and class-AB low-level crossover distortion per manufacturing grade.

  • 20

    EQ inductor hysteresis

    Jiles-Atherton inductor hysteresis with genuine path-dependent inductance modulation.

  • 21

    Capacitor dielectric absorption

    Multi-time-constant charge re-release per dielectric type.

  • 22

    Slew-rate limiting

    Per-topology amplifier slew-rate limiting.

  • 23

    AC-coupled output path

    Gain-cell DC offsets never reach the output, whether the output transformer is engaged or not, matching hardware output coupling.

  • Dynamics
  • 24

    Variable-Mu feedback compressor

    Gain-reduction-coupled harmonic generation.

  • 25

    HF soft clipping

    Program-responsive, in Tube and DPX modes.

  • Environment
  • 26

    Cross-channel crosstalk

    Cross-channel crosstalk with frequency-dependent shaping.

  • 27

    Manufacturing tolerance

    Per-instance manufacturing tolerance variation from seed.

  • 28

    1/f thermal drift

    Filtered-noise 1/f thermal drift, aperiodic and per-instance.

  • 29

    Thermal self-heat

    Program-responsive, with fanout to transformer permeability, transistor bias, tube transconductance, and core loop parameters (k, a, c).

  • 30

    Component warm-up

    Physical component warm-up on instance creation.

  • 31

    Current-scaled shot noise

    On five Class-A device stages.

  • 32

    1/f noise spectral shaping

    Per-topology, across the full audio bandwidth.

  • 33

    Channel-independent noise floors

    Each channel draws its own noise stream while the stereo pair shares matched component tolerances, so the floor sums to mono and nulls the way two real devices do.

  • 34

    True Gaussian noise statistics

    The modeled thermal floor carries the statistics of real thermal noise rather than a uniform digital approximation.

03 / Downloads

Install Sphinx 101.

macOS Apple Silicon Intel · v1.6.9

Sphinx_101_v1.6.9_AU_Universal.dmg

~24 MB · AU · Universal binary (Apple Silicon + Intel) Download · .dmg
macOS Apple Silicon Intel · v1.6.9

Sphinx_101_v1.6.9_VST3_Universal.dmg

~24 MB · VST3 · Universal binary (Apple Silicon + Intel) Download · .dmg
Windows x64 · v1.6.9

Sphinx_101_v1.6.9_Setup.exe

~19 MB · VST3 · Windows 10 or later Download · .exe
Single-architecture builds are on the support page. After install, insert Sphinx 101 on your master bus or any stereo bus - and play. One-time web activation; the plugin never calls home.
04 / Technical specifications

Specifications.

Format & platform

AU and VST3 on macOS 11 (Big Sur) or later - universal installers carry Apple Silicon (arm64) and Intel (x86_64) in one package, with single-architecture builds also available. VST3 on Windows 10 or later (x64). 4 GB RAM, approximately 50 MB disk.

Audio Sample rates 44.1-192 kHz (verified identical character). Buffer 64-2048 (bit-identical output). Internal precision 64-bit. Oversampling 2× / 4× / 8× / 16× applied globally - one minimum-phase polyphase IIR half-band wraps the entire signal chain, every module runs at the elevated rate. The minimum-phase design eliminates pre-ringing; phase rotates monotonically with frequency, matching analog hardware. Noise floor produced by the component noise model; exact figures for every personality and topology combination are documented in the shipped Noise Verification test report. Per-element alias measurements documented in the shipped HF Alias Regression test report; full-chain Plugin-path alias measurements in the shipped Plugin-Path Alias Regression test report.

Operating point

−6 dBFS = 0 VU. Bypass transparency bit-exact in steady state (zero residue; TrueRail chassis character only while enabled). Polarity non-inverting at all frequencies. Sample-accurate DAW latency compensation.

Architecture 79 parameters, all automation-compatible. 34 TrueRail mechanisms. CPU: 3-10 % single-core at 2× on Apple Silicon. Authorization one-time, 2 seats per serial.

products / sphinx-channel / v1.0.1 · 2026

Sphinx Channel

Analog channel summing. A component-modeled input chain meant for every track - input transformer, discrete drive circuit, TrueRail power-supply grades, and a stereo width stage. Every instance connects to a shared virtual console frame: supply loading, mains-ripple phase, channel bleed, signal ground, and ambient heat are shared across the whole session, the way strips share one frame in a real console. A single instance always sounds like one channel; a full session gains the common electrical character of one desk.

FormatAU · VST3
Platform macOS 11+ · Windows 10+
UsageChannel Summing
CircuitsSLL · Nevy · Amok
TransformersM1166 · C9049 · K1166 · L1544
Mechanisms32 modeled · incl. shared console
Seats2 per serial
Telemetrynever
Code languageJUCE 8 and C++
01 / Screens

Two skins. Different vibe.

  • 01 - Nevy skin / default
  • 02 - Amok skin / light metal
  • 03 - Signal analysis
02 / Feature highlights

One strip per track. One console per session.

  • The shared console - Sphinx Channel exclusive
  • 01

    Shared supply loading

    Every instance's current draw sags the common rail that every other instance reads - a busy session breathes as one power supply.

  • 02

    Console-coherent mains ripple

    All instances phase-lock their supply ripple to one shared console clock, summing the way strips in a single frame hum together.

  • 03

    Console channel bleed

    Each strip contributes a calibrated share of adjacent-channel bleed, held at "one console" level no matter how many instances the session runs.

  • 04

    Shared ground-return coherence

    The signed ground currents of all strips sum, with real cancellation, into each strip's ground-lift path.

  • 05

    Console ambient heat

    The frame warms with the whole session's work, and every strip's thermal state feels it.

  • Channel strip
  • 06

    Input transformer

    Four core materials - M1166, C9049, K1166, L1544 - with full Jiles-Atherton hysteresis, remanence, and Bertotti dynamic core loss.

  • 07

    Three drive circuits

    SLL push-pull BJT, Nevy single-ended Class-A, Amok 5670 triode - each changes the whole chain, not just the drive stage.

  • 08

    TrueRail grades A / B / C

    Component-tolerance spread from factory spec to vintage wide. Same instance seed across tiers - switching widens or narrows the same variation pattern.

  • 09

    Width stage

    -100 to +100 percent stereo width on a modeled tube line stage. Stays off on mono tracks.

  • 10

    ALL remote control

    One button broadcasts knob changes, preset loads, window size, and skin to every Sphinx Channel in the session.

  • 11

    Per-instance tolerances & warm-up

    Every instance draws unique component values from its seed and physically warms up in its first moments, like real hardware.

03 / Downloads

Install Sphinx Channel.

macOS Apple Silicon Intel · v1.0.1

Sphinx_Channel_v1.0.1_AU_Universal.dmg

~24 MB · AU · Universal binary (Apple Silicon + Intel) Download · .dmg
macOS Apple Silicon Intel · v1.0.1

Sphinx_Channel_v1.0.1_VST3_Universal.dmg

~24 MB · VST3 · Universal binary (Apple Silicon + Intel) Download · .dmg
Windows x64 · v1.0.1

Sphinx_Channel_v1.0.1_Setup.exe

~19 MB · VST3 · Windows 10 or later Download · .exe
Single-architecture builds are on the support page. After install, insert Sphinx Channel on all tracks and/or groups - and play. Use the ALL button to control every instance from one plugin. One-time web activation; the plugin never calls home.
04 / Technical specifications

Specifications.

Format & platform

AU and VST3 on macOS 11 (Big Sur) or later - universal installers carry Apple Silicon (arm64) and Intel (x86_64) in one package, with single-architecture builds also available. VST3 on Windows 10 or later (x64). 4 GB RAM, approximately 30-50 MB disk.

Audio Sample rates 44.1-192 kHz (verified identical character). Buffer 64-2048 (bit-identical output). Internal precision 64-bit. Oversampling 2× / 4× / 8× / 16× applied globally - one minimum-phase polyphase IIR half-band wraps the entire signal chain, so every nonlinearity is anti-aliased by the same filter pair with no per-module latency mismatches.

Operating point

−6 dBFS peak, where the nonlinear components produce their most musical harmonics; −18 dBFS RMS sits on the printed 0 VU mark. Bypass crossfades in ~12 ms, bit-exact in steady state. Polarity non-inverting at all frequencies. Sample-accurate DAW latency compensation.

Architecture All parameters automation-compatible except the oversampling selector. Window size remembered per instance, outside the parameter system. Signal flow: Input Gain → Input Transformer → Drive Stage → Width → Output, over the always-active TrueRail shared infrastructure. Authorization one-time, 2 seats per serial. Demo mode: 2-second silence approximately every 2 minutes, synchronized across unactivated instances.