mfkne has released PULSEMELT, a free VST3 space and character engine for macOS and Windows.
It is not one box doing one thing: a tempo-synced dual-tap delay, a reverb blended in behind it, a continuous morph from clean through tape saturation into bitcrush and asynchronous pitch drift all share a single signal path, and the whole space gets out of its own way. The tail ducks under the signal that feeds the plugin, so transients stay in front while the modulated texture fills the rhythmic gaps behind them. There is no sidechain wiring, no bus routing and no companion plugin: put it on the track as an insert and turn MELT up.
The detector runs 3 ms ahead of the audio path and reports that lookahead to the host for delay compensation, so the tail is already out of the way before the transient lands. The dry path gets no filter, no saturation and no compression, only the compensated 3 ms and a level factor, so the attack shape of every hit passes through exactly as it came in. All the character is added behind it.
Four controls:
- SPACE sets room size and delay feedback. The delay locks to the host tempo, 1/8 note on the left and dotted 1/8 on the right with a little cross-feed between the taps, and higher settings move the balance from separate echoes towards one continuous tail.
- GRIT is one continuous morph on the effect signal only: clean at zero, then tape up to 55 % (asymmetric soft saturation, head and gap loss down to 3.2 kHz, gentle squash), then bit and sample rate reduction up to the top (16-bit down to 6 bit, sample and hold up to 24 times). Between 45 and 55 % both stages run together, so tape and bitcrush blend into each other instead of switching.
- DRIFT adds asynchronous pitch modulation from three LFOs per channel at unrelated rates, detuned between left and right, so the instability stays organic and never settles into an audible pattern.
- MELT is the macro. It scales the wet/dry ratio and the depth of the internal ducking at the same time, reaching 85 % wet and 28 dB of gain reduction at the top, while the dry signal holds its full level up to 60 % and eases down by only 3 dB at full travel.
Fixed and deliberately not adjustable: a 150 Hz high-pass in the wet path right before the output mix, so pitch modulation and saturation cannot pile up low end where the bass and kick live. Peak-normalised drive and a makeup taper keep GRIT from moving the level, and all four parameters ramp over 50 ms, so automation stays quiet.
Price: Free
Read More