While digital implementations of guitar distortion provide the flexibility to enable compact and feature-packed multi-effects in a single box, they often sound "cold" or "buzzy" to musicians. The goal of this research is to improve the technology of digital algorithms for guitar distortion so they sound as good or better than the analog pedals.

Digital algorithms for guitar distortion

Most algorithm designers work from the top-down, that is, they start with a black-box model of a distortion effect and use standard digital signal processing techniques to affect the sound. Commonly used tools are second-order filters (peaking filters, shelving filters), parametric EQ, and static nonlinearities (waveshaping).

Analog-based design

Analog circuits however, are typically low-order filters for cost reasons. In many circuits, there may be multiple stages of nonlinearity. A good understanding of how to design analog circuits for high-quality distortion will guide the design of digital algorithms. Analog circuits also provide a prototype by which to design digital algorithms, which should sound exactly the same as the analog circuits within the bounds of the approximations made. This site will provide analysis of the prototypical mechanisms for distortion used in guitar circuits.

Amp Modeling

Amp modeling is a hot topic these days. Usually this appears to be done in the same manner as distortion effect modeling, with a cabinet-miking model tacked on the end. The cabinet to microphone signal path is assumed to be a linear filter incorporating the frequency response of the cabinet, acoustic path, and microphone. The cabinet filter makes a significant difference in the character of the sound. Improving this segment of the algorithm could make great improvements in these distortion effects.

Amp Cabinet Acoustics and Miking Techniques

The design of guitar amplifier cabinets is very unusual in that they are often not designed with acoustics in mind. There is no tweeter and no crossover filter to limit the range of frequencies that drive the transducer. Therefore, the woofer works in a range that is not typical for most low-frequency transducers and itself acts like an acoustic instrument with a radiation pattern dictated by the vibrations of the cone. Many of the original vintage cabinets are open back, allowing for a complicated second path for the sound to radiate. The closed back cabinets do not seem to be designed for any particular frequency response. More recently there have been cabinets with ports that are probably designed for a particular frequency response.

The radiation pattern of the speaker cabinet is extremely complicated because of its unusual design. It is important to know the radiation pattern as a function of frequency so recording engineers may better know how to position the microphone to achieve a particular sound. Amplifiers are also commonly miked in the near-field of the transducer which has a complicated pressure field. As of now, amplifier miking is pure art, and requires extensive experience and experimentation. The goal of this research is to gather data on this practice and derive a theory that can guide the placement of microphones for recording electric guitar to streamline recording sessions.