One-Bit Delay

Most digital delays store audio as 16–24 bit samples at a fixed sample rate, delivering clean, highly accurate repeats with minimal coloration. The One-Bit Delay is different because, like classic delay chips from the 90's, the One-Bit Delay operates using an internal ADC to create a 1-bit stream saved into a small (44k) internal memory running at a variable sample rate. This has interesting effects: the tonal character shifts with delay as because it changes the sampling rate and also there is a "warmer" sound due to the presence of modulation and demodulation filters.

Noise is not a flaw, but inherent to the one-bit quantization process of classic delay chips and the One-Bit Delay plugin. With only a single bit per sample instead of the 16–24 bits used in conventional digital systems, quantization error is present. One-Bit Delay uses Delta-Sigma Modulation to manage this noise - shaping most of the noise energy above 20 kHz - although some remains audible, typically around −68 dB at 16× oversampling. As with the classic chips, the noise floor decreases with higher sampling rate (typical chips used 2-22 MHz). The One-Bit Delay plugin includes selectable oversampling from 1× to 32×, effectively increasing the sampling rate to achieve the same effect (less noise, if you want).

Yes! The realism of the One-Bit Delay comes from modeling the signal path of the PT2399 delay chip, rather than simulating its byproducts. Even down to the 44,000-bit ring memory, two-pole demodulation filter, and voltage-controlled oscillator behavior derived from the original datasheet equations. Because the structure itself is reproduced, the characteristic noise profile, harmonic distortion, bandwidth reduction at longer delays, and clock-dependent coloration arise naturally from the mathematics of the ADC-DAC instead of being artificially layered on top.