The Importance
of Full Spectrum Correction in the Time Domain

With current technology it is possible to design speaker systems to perform almost flawlessly under anechoic conditions. However, when you place a perfect speaker in an ordinary living room, the sound changes very dramatically. This is primarily due to standing waves, cancellations, and reflections. All of the room/speaker interface problems are time related. The direct (first) arrival of sound from the speaker is largely undisturbed by the room, but after the first arrival the sound propagates in the room in a very complex pattern. The greatest amount of energy is present at the lowest frequencies where most speakers radiate sound in a circular pattern - hence only a tiny fraction of the low frequency energy produced will arrive directly at the listening position. The room will usually change the frequency response from near perfect to something that is almost completely random - particularly at the lower frequencies. Deviation of more than +/- 10 dB is the norm rather that the exception. This response error is usually 100 - 1000 times greater than the combination of all other frequency response errors in the reproduction chain.

As if this is not bad enough, time behavior is even more compromised. Every room resonance occurs as a build-up of energy over time. In general, lower frequency resonances take longer to build up and longer to decay, generating a delay that can average close to one second compared to the higher frequencies. This explains why big speaker systems capable of low frequency reproduction often have a reputation for being slow. In reality, they are just more likely to excite lower frequency room resonances resulting in greater average delay of the resonances. The total energy of the resonances in a normal room can be significantly higher than the sound produced directly by the speakers. And in most cases, the resonances are much more powerful than the direct sound if you measure them just after a transient. The effects of the room problems are aggravated by the spectral complexity of musical instruments and voices: Almost every musical instrument is very broadband, meaning that even when a single note is played, the instrument is producing a wide range of upper and lower harmonics. An example: The cymbal - most would think of it as producing high frequencies only, but this is far from true. In fact, the frequency response extends from below 16 Hz and all the way up to ultrasonic frequencies. The low frequency output comes from the fundamental resonance of the suspension. This low frequency information is usually inaudible, because it is masked by the room problems which introduce not only frequency response errors but also a delay to the lower frequencies due to the resonances of the room. With the RCS, the information is accurately reproduced making the sound much more real and believable. (Unfortunately, many studios remove the low frequencies when recording cymbals because they are ignorant of their importance for the fidelity).

The TacT room correction system goes straight to the root of the problems. During the calibration process a number of impulses are sent to the speakers. The calibrated microphone picks up the impulses at the listening position. Based on the deformation of the pulse both fthe requency domain and the time domain response can be accurately determined. The actual frequency response is compared to the desired frequency response defined by the user, and the system calculates an extremely complex inverse filter for each speaker which will give the desired frequency response with the best possible time behavior. The filter is itself an impulse (in the time domain) which corrects both the time and frequency domain behavior of the speaker - room combination. As the correction takes place in the time domain, the level is the same before and after the impulse with no sacrifice of dynamic range. All processing is done with 48 bit word-length so that no noise or distortion is generated by the system. The system measures and calibrates the left and right main speakers and one or two subwoofers, so any difference between the left and right channels is also compensated.