Cita Iniciado por ManuelBC Ver mensaje
Qué sentido tiene mover el micro si el oyente no se mueve?

Es decir, si uno vive solo y siempre se sienta en el centro mismo del tv, para que mover micro?

Yo la primera que hice la medición, usé las 4 plazas del sofá, moviendo el micro como se indica el asistente en 6 u 8 puntos (ya no recuerdo las veces).
La segunda y tercera vez, ya solo usé el asiento central, que es el mio

Muy pocas veces vemos cine los 4 juntos, a lo sumo 2 a la vez.
Aunque hay usuarios que dicen que calibrando en un sólo punto los resultados son buenos, en realidad va contra la lógica de Audyssey.
Para realizar la calibración correctamente necesita información de la sala, cuanta más mejor, distinta en cada posición del micro. Si le damos siempre la misma información (mismo punto) no va a ser capaz de realizar el trabajo como se espera, por eso es importante que las posiciones del micro sean distintas. Si sólo hay un único punto de escucha lo mejor es que usemos las 8 posiciones del micro (6 puede valer), pero en un área muy reducida, es decir, en un radio de unos 8 - 15 cm con respecto a la primera posición. Es importante que un par de esas posiciones sean a distinta altura (8 - 15 cm también).

Aquí, Chris Kyriakakis (co-fundador de Audyssey), explica cómo funciona:

The technology is a result of a multi-year university research effort to understand the key factors that influence sound reproduction. From this research came two key findings:

  1. Acoustical problems in the room are more accurately measured in the time domain. This type of analysis provides information about the direct sound and the effects of reflections from room surfaces. This was a departure from traditional EQ methods that only looked at data in the frequency domain.
  2. Measuring in a single location does not capture sufficient information and often results in equalization artifacts. Multiple measurements are required to capture the spatial distribution of acoustical problems, particularly in the low frequencies where the problems are more serious.

MultEQ captures multiple measurements in the time domain and then groups them in clusters based on similarities in the data. Using Fuzzy Logic mathematics, the clusters are allowed to overlap so that each measured response belongs to each cluster with a certain probability. In each cluster a representative response is then created that is weighted by the acoustical problems in that cluster, but also by those in the other clusters. MultEQ then re-combines these representative responses to create a final room representation and then inverts that to create the correction filter for each loudspeaker. The type of filter used by MultEQ simultaneously corrects the time and frequency domain problems to produce a smooth response.
MultEQ also measures the time it takes for the signal to arrive from each speaker to the first microphone position. Delays are then applied to the speakers that are closer to match the timing of the signals coming from the speakers that are farther away. Finally, the sound pressure level produced by each speaker is adjusted with the trim controls so that they match each other.