A New Paradigm – Linear Transfer
Before delving into the nuts and bolts it is necessary to examine the desired goal of measurement. After all, how can one get somewhere without knowing the destination?
After much exposure to Smaart in the “early days” of 1997 it became quite clear to the author a personal rethinking of the equalization process was in order. What is the goal? Of course the easy answer is “make it sound good” but after watching Sam Berkow and Jim Brawley equalize several systems a methodology revealed itself. The goal: get out exactly what you put it.
This seems somewhat intuitive but upon examination a clear distinction can be made between the usual equalization process practiced by those not using test equipment and a scientific method supported by measurement and reasoning. The perfect integration of art and science can yield stunning results, yet it is crucial to understand when to apply each.
Robert Scovill is a master integrator of art and science. He describes the system equalization goal as “linear transfer”, and his goal is to create a system response free of resonance with no frequencies “hyped”. Ultimately, he expects the system to accurately track EQ changes made on the console channel strip. When the system is properly equalized this goal can be realized, and he is free to creatively mix on the console rather than fight a system that impedes the creative process with inaccurate response. With proper equalization his mix decisions are transferred in a linear fashion to the ears of the listeners, no more and no less. Linear transfer equals freedom to mix creatively.
Three Dimensional Sound
For many aspiring audio mixers and technicians, one of the most difficult challenges is to be able to identify frequencies. Since the most common tool at our disposal is a graphic equalizer, it is only natural that we begin our audio odyssey by considering sound in terms of frequency and amplitude. Go to the graphic equalizer, select the closest slider to the frequency of interest, and adjust the amplitude. Many of us will not advance past this point. However, this is akin to seeing in only black and white. There is a missing component in the big picture and until it reveals itself we are certain to fail in our attempt to marry art and science. This missing component is time, and the implications of not considering the relationship between multiple sound sources and the times at which they arrive at our eardrums are huge.
When we begin to consider what happens when a given sound source arrives at different times a new set of problems reveals itself. Adjusting that slider on the graphic EQ won’t solve it – if there is a problem in the time domain it must be addressed in the time domain. The listener may perceive the problem to be frequency/amplitude related, and in fact it is but the root cause of the problem is non-synchronized arrival of the same sound. When the “same sound” arrives at two different times, cancellation occurs and the frequencies at which this occurs are determined by the difference in the arrival times. To the listener, however, the problem simply lies in the frequency/amplitude domain although it cannot be fixed there.
Recalibrate Your Thinking
Next show, look all around you for sound sources. Backline, monitors, left and right PA, possibly a center cluster, subwoofers on the ground with a flown PA. Each of these delivers sound to your position at different times, unless you happen to be standing in the magical position in which each source is equidistant. Where is that spot? Does it exist at all? Where would you like for it to be?
When we begin to think of sound in terms of arrival time a whole new set of possibilities reveals itself. Is the problem truly only a frequency response problem, or is it time related? Is it a null caused by cancellation, or merely non-linear frequency response? With a tool like SIA Smaart Live the answers are there if you learn to interpret the information presented.
Comparing Inputs and Outputs
Smaart Live compares your audio source, presumably a console output, to the sound arriving at the measurement microphone. The audio from both sources is converted to digital format and compared to each other. The resulting display, or transfer function, is the difference between what is coming from the console and what is coming from the loudspeaker system. When you learn to interpret this information it is possible to equalize a system so that linear transfer can take place.
Since the signal present at the console output is traveling at the speed of light and the signal present at the measurement microphone is traveling at the speed of sound there is an offset involved. Smaart calculates the difference and gives the operator the numbers needed to time align the two signals. The signal present at the console output is the reference signal and the signal present at the measurement microphone is the measurement signal.
Think about it: how desirable and useful is the ability to determine the difference between an electrical audio source and that same source after it has been sent through a loudspeaker system? If you have the tools you can adjust the arrival time of various sources for synchronous arrival, as well as equalize out anomalies introduced by nonlinearities in the loudspeaker system.