Sound Card Performance Technical Benchmarks

TechTalk

The Distortion and Noise Model

Distortion and Noise are models of the common kinds of errors that are normally added to signals as they pass through sound cards. As shown in the drawing, below; distortion and noise are added to any signal that is processed by any sound card.

Error is a deviation from accuracy. The "Accuracy Model" of High Fidelity Sound Reproduction suggests that perceived sound quality increases as over-all system error decreases. Once error from a given source decreases to the point where it is far less than overall system error, or has decreased below what can be reliably perceived, further improvement has few, if any audible benefits.

Distortion and Noise are categories of errors that occur when a signal is processed by circuitry in a sound card.. Distortion is the name of the category of errors that is closely related to the input signal. Noise is the name of the category of errors that are generally unrelated to the input signal.

Distortion and noise are measured using FFT analyzers that separate and measure errors in the output signal. Noise is most commonly heard as a hissing or humming sound that was not part of the original program material, Distortion is most commonly heard as changes in the program materials tonal quality.

 

Examples of errors that are closely related to the input signal are errors in frequency response, signal timing, Harmonic Distortion (THD) , Intermodulation Distortion (IM), AM distortion, FM distortion, random noise, interfering signals, and modulation noise, as shown in the graphic, below.

 

Errors that change the relative size or relative timing of the various tones that make up the input signal are called linear distortion. Frequency response and phase response errors are examples of linear distortion. Linear distortion changes the tonal balance of program material.

Errors that add tones to the output signal that not present in the input signal but are related to tones in the input signal are called nonlinear distortion. THD, IM, AM, and FM distortion are examples of nonlinear distortion. Nonlinear distortion makes program material sound muddled, gritty, wavery, or adds artificial sweetness or hollowness.

Samples of pure sound and distorted sound, simulated by MIDI, can be heard by clicking here.

Noise also adds tones to the output signal that are not present in the input signal, but these tones are unrelated to the tones that are present in the input signal.

Noise can be predictable or unpredictable. Unpredictable noise is also called Random Noise and is usually due to thermal noise in components; or is in some sense a noise that is predictable, but has such a complex nature that we can't analyze it and determine how to predict it, yet. Hum and undesired pickup of signals from other equipment or near by broadcast equipment are examples of predictable noise.

Modulation noise is a general increase in noise that occurs when a signal is applied. Since this noise is composed of many frequencies, it is classed as a form of nonlinear distortion and noise combined.

Equipment with lower noise and distortion is generally considered to have better quality and be more desirable for the purposes of audio reproduction.

However, electronic musical instruments use intentionally created distortion to produce desirable musical sounds. The use of intentionally created distortion to create a desirable musical sound is called Euphonic Distortion. Euphonic Distortion is not consistent with the goals of High Fidelity Audio reproduction of music, despite its benefits for musicians who are making music.

If existing forms of noise and distortion are predictable and well understood, they can be compensated for by circuits that add noise and distortion that cancels, and thus reduces, existing noise and distortion.

PC AV Tech

Send questions or comments on the PC AV Tech  web pages to Arny Krueger

Website Art Direction by David Carlstrom Tests and Report Design by Arny Krueger

All material on this page is Arny Krueger's personal opinion, provided for entertainment purposes only. See web site disclaimer for details (c) 1998, 1999, 2000, 2001, 2002, Arnold B. Krueger. All rights reserved.

Please let The Webmaster know of any problems you encounter on this website. This Page First Posted4/2/1998 (abk) This Page Last Revised 05/13/2002 (abk)