Magic Series

MAGIC:70

One-knob 1970s production character, where the knob increases console and tape interaction rather than noise.

29 39SALE

Free 10-day trial. The download is the whole collection, 22 plug-ins in one installer, so you can try every one; start MAGIC:70's trial from inside the plug-in when you are ready. Downloading does not start it, and trying one plug-in does not start the others.

AU (macOS)
VST3
AAX
macOS Universal
Windows 64-bit
v1.0.0

What it does

Not dirt. Interaction.

Turn one control clockwise and the source progressively sounds as though it were tracked, processed, mixed and printed through a high-quality 1970s analogue studio. At zero the plugin is transparent, not “gentle”, transparent: all forty-five macro destinations sit at exactly their neutral value, for every Style, Era and Tape Condition, and the transparency test asserts the output is bit-identical to the input.

There is a line this product holds explicitly: vinyl is a playback medium, not the sound of 1970s multitrack production. Surface noise, hiss, hum, dropouts and print-through are all available and all off by default, in every factory preset, at every position of the knob. What the knob increases is console and tape interaction.

Each of the eleven Styles is a genuinely different signal-flow recipe, not a preset. They differ in which ambience algorithm runs, whether it sits before or after the tape, where the modulation block is placed, and which dynamics character is used.

Key features

Eleven Styles that reroute the chain

Warm Studio · Golden Soul · Deep Funk · Analog Rock · Arena '79 · Disco Floor · Dub Chamber · Cosmic Electronic · Soft Focus · Midnight Soundtrack · Experimental '79. Deep Funk puts its envelope filter and phaser in front of the preamp so the nonlinearity fills the notches. Golden Soul prints a dark plate to the tape. Dub Chamber feeds a spring reverb into the tape echo.

Three Eras and three Tape Conditions

Early, Mid and Late '70s modify the selected Style without replacing it; Fresh, Used and Worn modify it again. Worn is an intentional special effect, not a degree of realism.

Band protection that stops the chain seeing the kick

DRY BELOW and DRY ABOVE let frequencies outside them bypass the whole chain and sum back at unity. Usually the larger benefit is that the tape compressor and limiter stop seeing that energy, so the chain stops ducking on every hit. Measured at 39 dB of protection on a 55 Hz tone with a 200 Hz split, bands recombining within 0.08 dB.

MAGIC MATCH is deterministic and explainable

Press it, play a few bars, and it measures twenty-three features, estimates the source type with a stated confidence, and proposes offsets. Every rule is a documented threshold, and the sentence you read is generated by the same rules that produced the offsets, so the explanation cannot drift from the behaviour. It refuses rather than guessing: silent, clipped, too-short and inconclusive input each fail by name.

Switches and turns without clicks

Continuous settings glide on a two-pole curve that lands exactly on its target. Discrete choices such as the console, tape speed and formulation, dynamics type and generation count morph the numbers they stand for over 40 ms, and the rotary, flanger and chorus run silently until they have filled, then fade in. Measured above 6 kHz on a held tone across 256 Style, Era and Tape changes, the worst now stands 3.0 dB over the settled sound, where it stood 71.9 dB over before. With nothing moving, the output is bit-for-bit what it was.

Auto-gain measured per Style

One shared curve could not do it: at maximum, the Styles span from +6.3 dB to +12.2 dB. Each Style now carries a curve measured by running a programme corpus through the whole engine.

The interface

The MAGIC:70 advanced drawer open on its Tape page, exposing tape speed, formulation, record level, bias, tape saturation, tape compression, head bump and generation controls. — MAGIC:70 · 1 of 1

Audio

Audio demos for MAGIC:70 are not published yet.

We would rather show you nothing than show you a demo that was not recorded through this plugin.

Measured

Stage 11 and 15 validation, the transition tests and Magic70Calibrate · programme corpus through the full engine

Style × Era × Condition combinations rendering in range: all 99, no NaN
Style pairs separating acoustically at MAGIC 0.45: all 55
Third harmonic of a 1 kHz tone, MAGIC 0 → 85: −82 dB → −14 dB
Style, Era and Tape changes on a held tone, worst of 256, above 6 kHz: +3.0 dB over the settled sound (before the glides: +71.9 dB)
CPU: MAGIC 0: 0.43 % of a core
CPU: 2×: 3.4–4.7 %
CPU: maximum quality: 8.3–9.8 %
Chain at MAGIC 0, both splits off: bit-exact
Band recombination with a 200 Hz split: within 0.08 dB
auval: validates

Figures are produced by this plugin’s own test suite and benchmark, on the hardware and sample rate named above. They are not estimates.

Technical details

Formats, macOS: AU · VST3 · AAX
Formats, Windows: VST3 · AAX
Operating system: macOS 10.15 or later · Windows 10 or later, 64-bit
Architecture, macOS: Universal: Apple Silicon and Intel
Architecture, Windows: 64-bit
Plugin code: Mgc7
Version: 1.0.0
CPU: 0.43 % at MAGIC 0, 3.4–4.7 % at 2×, 8.3–9.8 % at maximum quality
Factory presets: 71

In ten categories: a Reference ladder through the knob's zones, then Mix, Vocals, Drums, Bass, Guitars, Keys, Synths, Ensemble and Effects. A preset stores what the macro was given (the Style, Era, Tape Condition, knob position and the handful of things changed by hand) rather than 114 resolved values. That is the difference between a library that follows the curves as they improve and one stranded on the voicing it was authored against.

Worth knowing

It reproduces the recording and mixing fingerprints of the decade. It cannot change the performance, the composition, the players, the microphone placement or the room, and nothing here markets it as though it does.
Auto-gain accuracy is stated rather than summarised: 0.02 dB where it was fitted, up to 2.16 dB on a single source, and up to 2.09 dB of rise between adjacent tenths of the knob. Both limits are asserted every build.
MAGIC MATCH's source estimator has been validated on seven synthesised items and a handful of constructed edge cases. It has never been run against a library of real recordings.
Every claim on this page is a measurement. No listening evaluation has been performed.