DEVPROAUDIO REVERSIDE [WiN-MAC]

• Product: REVERSIDE
• Developer: DEVPROAUDIO
• Version: 1.0.0
• Format: VST3, AU, CLAP
• Requirements: Windows 10 or later, macOS 11.0 or later
• Source: devproaudio.com

REVERSIDE is a dual-engine algorithmic reverb plugin that splits Early and Late Reflections into two independently editable engines. Its core feature is geometry-driven 3D spatialization — you place a virtual source and listener inside a modeled room, and the room’s shape and size steer how both the early reflections and the tail behave. Fourteen diffusion models, a non-exponential decay envelope, and pitch-preserving modulation give it a detailed, wide sound. It’s built less for one reverb send and more for placing several instruments inside one coherent space.

Key Takeaway

REVERSIDE’s real strength is putting multiple instruments in one believable 3D space, not adding ambience to a single track — that’s what the dual-engine and geometry system are actually for. It costs more CPU than a simple plate reverb, so it’s overkill for a quick vocal slap-back. The payoff shows up when a mix needs several instruments to sound like they’re standing in the same room.

Shape Control Runs From Box to Chaos

A percussion bed built from four or five close-mic’d sources can pull up to 200 individual reflections from the Early engine, or hand the count to AUTO mode, which derives it from the room geometry instead of a manual slider. Room size can be set in absolute terms for a fixed physical space or relative terms that scale with the source, and a Shape control pushes the room from a perfect box toward chaotic, scattered surfaces.

Wall Absorption and Fill model how much of each reflection survives contact with a modeled surface, while Distance-induced Decay and the frequency-dependent MAT filter roll off high content the way carpet or drapery would in a real room. Bounce type switches between Soft and Hard phase modes, changing whether reflections stack cleanly or interfere with slight phase cancellation — audible as a subtle comb-filtering edge on Hard settings with dense reflection counts.

Pre-delay runs in free, tempo-synced, or relative modes, so a reflection pattern can lock to song tempo for rhythmic percussion or stay fixed in milliseconds for dialogue work. Azimuth and Elevation Angle Displays show the directional spread of the current reflection set at a glance, speeding up the match between a reflection pattern and a specific room shape without soloing the engine repeatedly. Producers building layered percussion beds get the most out of the reflection density here; a single overdubbed acoustic guitar rarely needs 200 discrete reflections to sit right.

Moving the Listener Changes the Reflections

Dragging the listener point across the 3D display recalculates the directional character of the early field and the tail together, rather than applying a stereo pan after the fact. Direct path spatialization keeps the dry signal’s apparent origin consistent with where the source sits in the room, which is what makes a placed instrument externalize instead of sitting glued to the speakers.

3D ER and LR spatialization each carry their own quality mode, so a mix bus can run a lighter setting for a background pad while a featured vocal gets the highest-resolution mode the CPU budget allows. Position, rotation, and distance are all automation-ready, so a source can travel through the modeled room over the course of a phrase instead of holding one fixed spot for the whole track.

Two sources placed at nearly identical coordinates can blur into the same directional cluster rather than reading as separate points, since the spatialization responds to actual geometric distance rather than an arbitrary channel index. Sound designers building multi-source scenes for picture get the clearest return from this system; a single mono guitar double doesn’t need independently automated 3D placement to read as centered.

An Envelope That Isn’t Exponential

Fourteen diffusion models sit behind the Late engine, each producing a different texture as the tail builds density — some closer to a smooth plate, others rougher and more grain-like. The EXT envelope breaks from a standard exponential decay curve, so attack and decay get shaped independently instead of riding one fixed curve tied to a single decay-time knob.

Density moves the tail from sparse, individually audible echoes to a thick, continuous cloud, and Shape and Spread reshape the contour from a tight, focused tail to something wide and diffuse. Per-channel decay shaping treats left, right, mid, and side independently, and WIDE mode pushes the stereo or M/S field further apart for tails that need to fill more of the speaker width.

Dialing in a specific tail character this way takes longer than turning one decay-time knob, since size, density, shape, spread, and per-channel balance all interact. Sound designers building evolving ambient beds or scoring cues get the most from that depth; someone needing a fast, generic room tail on a drum bus in thirty seconds will spend more time here than the task calls for.

Door Slam or Curtain Draw

REVERSIDE effectively runs four processing stages rather than two: the Early and Late engines each carry a dedicated mini diffuser layered on top of the main reflection and diffusion models. The ER diffuser’s Scatter control adds micro-delay density to individual early reflections, breaking up the evenly-spaced, slightly mechanical feel a fixed reflection pattern can have on its own.

The LR diffuser stretches the tail’s attack envelope instead of adding density after the fact, moving the reverb’s onset anywhere between a door slamming open and a curtain slowly drawing back. Each diffuser runs its own envelope independent of the other, so an early field can stay tight and quick while the tail still blooms in gradually.

Both diffusers layer on top of the 14 diffusion models rather than replacing them, so stacking heavy Scatter with a slow LR bloom on an already dense diffusion model can smear a transient-heavy source like a snare hit past the point of clarity. Mix engineers chasing a specific onset character — snappy versus gradual — get direct control here that a single diffusion-model dropdown doesn’t offer elsewhere.

Bass Lingers Longer Than Treble

The MAT filter applies frequency-dependent absorption to each reflection cluster in the Early engine, modeling the way carpet soaks up high frequencies while concrete stays bright — the same reflection can read warm or hard depending on where the split points sit. The DCY filter works on the Late engine instead, controlling how reverb time itself varies across the spectrum so bass can ring longer than treble, the way it does in a room with soft ceiling treatment.

A bidirectional routing switch links the two filters directly: independent, MAT feeding DCY for physically plausible spaces, or DCY feeding MAT for the inverse, more synthetic result. Each band can pan independently across L/R or M/S, so a filtered low band and a filtered high band can sit at different points in the stereo field instead of moving together.

REVERSIDE’s per-engine 3-band EQ sits separate from MAT/DCY, with bell, notch, shelf, cut, and band-pass types and slope choices from 12 to 48 dB per octave — steep enough to carve a reverb tail out from under a lead vocal without touching the vocal’s own EQ. Running MAT, DCY, and both 3-band EQs on one instance is more filtering than a single reverb typically exposes, and a quick send-effect role on a rough mix doesn’t usually call for four independent filter stages at once.

No Single Flat Control Page

Easy and Main serve as two hub pages, but MAT/DCY routing, per-channel decay shaping, and diffuser envelopes each live on a dedicated sub-page reachable only by leaving both hubs. The Easy page surfaces reverb time, size, pre-delay, and mix as a small set of top-level controls, enough to get a usable sound from a preset without opening any editing page at all.

Rewritable parameter defaults mean a producer’s own starting point replaces the factory default once set, so a new instance opens close to a personal baseline instead of a generic room. The Main page adds three live mini displays for reflections, position, and reverb tail, so shape, placement, and decay stay visible at once instead of requiring a page switch to check each one.

A/B/C/D memory slots hold four full parameter states for direct comparison, and parameter locks keep specific controls fixed while browsing presets — useful for holding a dialed-in room size steady while auditioning different tail characters. Factory content runs to hundreds of presets across rooms, halls, and plates, with contributions from RumbaCodex and Nirto of NSONICS alongside DEVPROAUDIO’s own bank.

A mix moving fast under deadline spends more time in menus than a plugin built around one flat parameter list, since reaching the deeper routing and shaping controls still costs a page switch every time. Sound designers building a signature space preset-by-preset get the most from the four-way recall and locks; someone applying one reverb across a whole session in a few clicks will lean on presets more than the deeper editing tools.

FAQs