electric machines of android dreams studio

emad studio, inc.

We are building the next generation of oversampled data converters for high-resolution audio. Our Delta-Sigma-C architecture replaces passive noise shaping with active control, delivering provable stability and performance that exceeds the state of the art.

// technology

All modern oversampled data converters — in devices from smartphones to professional recording studios — are based on delta-sigma modulation, a technique developed in the 1960s. While effective, delta-sigma modulators suffer from a fundamental trade-off: as designers push for higher accuracy, the system becomes increasingly prone to instability.

The Delta-Sigma-C (DS-C) architecture takes a fundamentally different approach. Rather than relying on the indirect mechanism of noise shaping, DS-C formulates data conversion as a tracking-control problem and employs a nonlinear controller to explicitly minimize the conversion error. Stability is guaranteed via Lyapunov theory — not left to empirical tuning.

r y H loop filter e_o state feedback e C controller y_c Q quantizer y

DS-C signal flow: inputs r and y → loop filter H → controller C → quantizer Q → output y, with feedback

provable stability

Mathematical guarantees via Lyapunov theory. No empirical margins, no instability surprises.

flat noise floor

No spectral notches or spurious tones. Clean, transparent sound with no audible artifacts.

active error control

A nonlinear controller explicitly drives quantization error to zero, rather than relying on passive noise shaping.

spectral linearity

No measurable intermodulation distortion above the noise floor. Linear SQNR vs. input amplitude.

// performance

In simulation, the DS-C architecture achieves signal-to-noise-and-distortion ratios (SNDR) that substantially exceed what is possible with conventional delta-sigma designs of comparable complexity. For reference, CD-quality audio requires ~98 dB, and the best commercial converters achieve ~120–130 dB.

Configuration Oversampling Quantizer SNDR
5th-order filter 512× 1-bit >160 dB
9th-order filter 512× 1-bit ~184 dB
15th-order filter 256× 1-bit ~210 dB

All configurations use 1-bit quantization, making the output directly compatible with the Direct Stream Digital (DSD) format used in Super Audio CD and high-resolution audio streaming.

// applications

professional audio

High-resolution ADCs for recording studios, mastering facilities, and live sound capture. Transparency and dynamic range beyond current products.

consumer hi-fi

DACs and ADCs for high-end audio equipment, portable players, and streaming hardware. DSD, PCM, and emerging high-resolution formats.

measurement

Precision data converters for test equipment, scientific instruments, and industrial measurement. Exceptional linearity and dynamic range.

// research

The DS-C architecture was conceived through original research in nonlinear control theory and has been published in peer-reviewed venues.

T. Zourntos, "Oversampled Encoding without Delta-Sigma Modulation: A Novel Alternative based on Nonlinear Control," Proceedings of the American Control Conference (ACC), Portland, OR, 2005.
T. Zourntos and D. A. Johns, "Variable-structure compensation of delta-sigma modulators: Stability and performance," IEEE Transactions on Circuits and Systems I, vol. 49, pp. 41–53, 2002.

A comprehensive article for the Journal of the Audio Engineering Society (JAES) is currently in preparation, presenting the full DS-C theory and simulation results for audio applications.

// contact

For partnership inquiries, technical discussion, or general information:

takis.zourntos@emads.org

Takis Zourntos — Founder
emad studio, inc. · Canada