476 Recording Studio Design protection in such small boxes gives rise to the need for steep protection filters, but the time response suffers accordingly. The time response of the Tannoy, at low frequencies, can be seen to go completely off the scale. 19.9 General discussion of results The three loudspeakers under discussion in this chapter were chosen because they are all good performers. They are all above average in the range of loud- speakers typically used in professional studios as close or mid-field refer- ences. Despite this, no aspects of their performances match. Neither their pressure amplitude responses, phase responses, time responses, harmonic distortion responses, nor their diffraction characteristics need an expert to sep- arate them. They are all obviously different. It is therefore not difficult to conclude that the loudspeakers all sound different; which they do. A parallel exists with microphones, which are more or less loudspeakers in reverse. Nobody would pay 3000 for an old Neuman microphone if a 200 Shure could be equalised to sound the same. The characteristic sounds of electro-mechanico-acoustic transducers are highly complex combinations of the convolution of their time and frequency domain responses, along with the spacial responses, which also affect their overall responses in non-anechoic conditions. One great problem for investigators looking into the audibility of various aspects of loudspeaker performance is that of separating combined effects. Where a diffraction problem causes spacial smearing by introducing secondary sources, then what is being heard? The time domain effects, the frequency domain effects, the spacial effects, or the combinations? Despite all the money involved in the music, hi-fi and recording worlds, an absolutely extraordinarily small amount is spent on subjective/objective correlation. This is a sad reflection on how the professional industries have been usurped by the business people. The current situation is that there is no loudspeaker that is optimal for all rooms. This fact can be clearly seen from the on and off-axis responses of the three loudspeakers studied in this chapter. No loudspeaker is optimal for all music. Of the three types of low frequency responses shown in Figure 19.5(a), one really would not want to play classical music through `b' or disco music through `a', because they would simply sound much more appropriate in the reverse order. Obviously, no loudspeaker can sound optimal in a bad room, though the Tannoy would probably sound better than the Westlake because of its very smooth total power response, as shown in Figure 19.3(c). All the designs are affected very differently by their circumstances. High definition monitor systems with low distortion, good transient accur- acy, flat and extended frequency responses, high SPL capabilities and decent sensitivity are expensive to make and require expensive components. They also tend to be large. Commercial realities therefore force compromises on loudspeaker designs, and the wide range of loudspeakers that are commercially available reflect the need for choice if best-fit compromises are to be eco- nomically matched to circumstances. Unfortunately, the people who choose the loudspeakers are rarely fully familiar with the reasons for the compromises. They hear a loudspeaker in one set of circumstances, presume that the sound