Most loudspeakers are measured on axis at a distance of one meter. This is often the only measurement that you will see, if you even see that. While an on-axis measurement can show one aspect of a loudspeaker's performance, this measurement tells only a small part of the story - a very small part.

Ai measures the response of a loudspeaker from -90° to 90° off-axes in two planes, vertical and horizontal. A listener's perception of the sound quality of a loudspeaker system in real rooms is more dependent on the power response than the response along any single angle. The axial response is a very unique direction where path lengths from the source all tend to be equidistant and it only represents a small portion of the total sound field. For these reasons the axial response is an almost meaningless indicator of the actual perceived response of the loudspeaker system in real spaces. The polar response is of paramount importance.

At Ai our loudspeakers are designed for a uniform response at all off-axis directions up to about 30°. This polar angle is highly correlated with the total power response since most of the sound lies within this cone. We don't consider the axial response to be highly relevant. No one axis is considered to be significant. Looking at an Ai frequency response map you can see how uniform the responses of our loudspeakers are, as-well-as exactly where there are compromises. A single off axis curve or even several curves are simply not enough. The polar response is of such paramount importance that highly detailed information is required to determine a good speaker from a bad one. We do polar responses at 7.5° increments for a very high resolution of the total polar response.

Having a uniform polar response means that not only does a wide range of positions in front of the loudspeaker have an accurate frequency response, but the direct sound field and the reverberant field will both have the same response yielding a much better perception of the sound field. The narrow directivity of our loudspeakers means that interaction with room boundaries and poor room acoustics is minimized since the response in the critical midrange ultimately drops off to nothing as you move further off axis. Room placement and acoustics becomes less critical than a poorly designed loudspeaker.

Our systems are true Constant Directivity designs not just "CD" claims without any supporting data. Many loudspeakers claim to be CD but a careful analysis of them shows that they are not at all constant in polar output across the important midrange bandwidth. Our systems yield a uniform polar pattern with a near seamless transition between high and low frequency transducers ( a small response hole exists in the vertical plane as the result of the source displacement in the vertical direction).

A constant narrow directivity at low frequencies is virtually impossible to obtain, but fortunately at frequencies below about 500 Hz our perception system is insensitive to this physical limitation. We are not very good at localizing or discriminating sounds below about 500 Hz. We can readily detect the pitch of these sounds quite accurately, but we are not very sensitive to spatial and temporal aberrations at these low frequencies. Above this frequency however our hearing system is highly tuned to its perception task in localizing and discriminating sounds. A problem in the frequency range of 1 kHz through 6 kHz will be disastrous as evolution has created in the human species an incredible refinement of our hearing acuity in this range.

The power response is the area weighted average of the sound emitted in all directions. The axial response of a loudspeaker represents a very small area and hence its weighting in the power response is also very small. In essence the axial response is a negligible factor in the perception of a loudspeaker in a real room.

Constant Directivity or CD means that the frequency response of the system is the same no matter what direction one looks at this response. True CD is virtually impossible at very low frequencies and one is usually most interested in frequencies above about 500 Hz. CD has become so misused in the field as to be misleading. It is often used to define horns which are not in fact constant in their directivity. True CD is rarely found in practice and virtually never does a manufacturer produce sufficient data to determine if it is CD or not.