What Is a Graphic Equalizer?

The graphic equalizer, English name is Graphics EQ, also referred to as "EQ", but the real "EQ" (equalizer) stands for broader meaning, that is, EQ includes "graphical equalizer". The graphic equalizer uses a sliding controller (slider) as a multi-band variable equalizer for parameter adjustment. The logo under the sliding controller corresponds to its frequency response. Each music player has an EQ, called "Adjusting Sound Effects." For example, there is an EQ in Qianqian Jingting, Windows Media Player and Real Player. The center frequency and bandwidth of each band in EQ are fixed.

Graphic equalizer

The following uses Cool Edit Pro as an example to introduce graphic equalizers (such as illustrations).

do

Analog EQ, digital EQ aspect ratio: The most primitive EQ is the use of the capacitor's so-called "capacitive reactance" to adjust the tone of the sound. The so-called "capacitive reactance" means that the capacitor has such a physical phenomenon. For capacitors with different specifications, they may weaken or boost AC signals at different frequencies. After the sound is converted from the MIC (microphone), it will become an alternating current signal, and the current I will be proportional to the sound amplitude (in fact, it can only be approximately proportional). I enters the EQ through the wire. Let's use a theoretical circuit of a 3-segment EQ as an example:

Three capacitors with different specifications are responsible for adjusting high frequency, intermediate frequency and low frequency. Because the three capacitors have different sensitivity levels for high, medium, and low frequencies, one can generate the EQ effect by adjusting the current transmission efficiency of each capacitor. This method of using physical phenomena is wise and labor-saving, and quite accurate! However, with the development of digital recording technology, sound engineers began to like to add EQ at the later stage, and traditional EQ cannot meet the needs. As a result, more and more digital EQs have appeared in people's eyes. To adjust the EQ in a digital signal whose sound signal has been quantified, a mathematical algorithm must be used to solve it. Everyone must have heard of the concept of "sampling rate". In digital audio signals, the change of the waveform cannot be continuous, but stringed up one by one.

This design creates a problem-when analyzing the frequency of the sampling point, it is difficult to find another sampling point that exactly coincides with the amplitude state of this point:

Therefore, the digital EQ must connect the sampling points like threading, so as to approximate the two points with the same state. It's easier said than done. Computers are not human brains. They can only be "threaded" mathematically. The oldest method, which I call "straight path", is to connect the sampling points with a straight line. This method is very simple, but everyone knows that there cannot be a straight line connection between the sampling points and the sampling points, which will cause great errors! Later, people connected the sampling points with the curve closest to the original waveform according to a formula in the high number (forgot the name), which I called "analog path". As shown in the figure:

The error of this method still exists, after all, it is calculated by theory and not the real waveform. However, there is little difference from the original waveform. Most popular digital EQs today use this design.

The principle of digital EQ:

Although there are many types of digital EQ, the principle is the same, that is, the input signal "x" is established to correspond to the output signal "Y", and Y = f (X), where the function (f () includes one and " x "corresponds to a function of frequency" k ". Expand the function expression corresponding to "X", that is, Y = g (k) * X. Where g () changes with the adjustment of the EQ parameter.

Example: The principle of ancient digital EQ.

This is an ancient 3-segment EQ, using a "straight path". We increased the intermediate frequency by 2 times and the high frequency by 3 times. At this point, the function's function becomes:

Y = 1 * X (k belongs to 0Hz to 400Hz)

Y = 2 * X (k belongs to 400Hz to 2500Hz)

Y = 3 * X (k is from 2500Hz to infinity)

It can be seen that this kind of EQ adjustment is has a corner and an angle, the amplitude of 399.9Hz is still a little unchanged, and it suddenly increases by 2 times to 401Hz. The addition of this EQ has produced the voice of the devil ......... EQ now has not only a "simulation path", but also a function function with a gradual change. For the same 3-band EQ, the intermediate frequency is increased to 2 times, and the high frequency is increased 3 times. The function image will become very smooth.

This "staircase" is very smooth. Although the intermediate frequency starts from 400Hz, it has started to increase the amplitude to produce a gradual effect from around 350Hz. Everyone can try, even if we reduce the high frequency of EQ to 0, we can still hear a little high frequency. And because the "analog path" is used, the frequency analysis is more accurate! Easier to adjust. But these two optimization algorithms cost more system resources than the ancient EQ.

The reason we want to talk about the ancient EQ that is no longer useful is because it makes it easier for people to understand EQ. Some friends always ask: Since the EQ effect can change the frequency of the sound, will the E-tune of C-tunes become B-reduced after being tuned? ? Lower the bass (bass) low frequency. Does the bass sound like an 8 degree rise? Do you remember the concepts of "music frequency" and "sound frequency" mentioned earlier? We use this concept to explain these two issues, starting with the ancient EQ.

Let's look at the formula of the ancient EQ: Y = r * X (k belongs to a Hz to b Hz). As mentioned earlier, the pitch of a sound is only related to the "music frequency". That is, to prove that the EQ effect can change the frequency of the sound without changing the pitch, just prove that the EQ effect can change the frequency of the sound without changing the tone frequency.

According to the definition of musical tone frequency, it must be the reciprocal of the length of time between the zero points of the same state (the first zero, the third zero). We set the time at 1 o'clock as t1 and the time at 3 o'clock as t2. Music frequency f = 1 / (t2-t1). Let's prove that there is no change at time t1 or time t2: for any input signal "x" there is an output signal Y = r * X (k belongs to a Hz to b Hz). At any time t, the signal processed by EQ can be changed to any value. However, because the X value of points 1 and 3 is 0, no matter how we adjust the EQ parameter, Y = r * 0 = 0, so at point 1, 3, the X value is always equal to the Y value of 0. That is to say, all the time points with an amplitude of 0 are processed by EQ, and the amplitude is still 0, so the time interval between the first zero point and the third zero point does not change with the parameter change.

This is why EQ effects can change the sound frequency without changing the pitch, so everyone (especially beginners) can use EQ with confidence. In fact, with the advancement of technology, digital EQ algorithms have begun to become more diverse. Just before the completion of this manuscript, I heard that there are novel and clever tricks to determine the frequency by calculating the slope (that is, the speed at that point) before and after two points at any frequency point, but the purpose of EQ is unchanged- only change The same tone. Sound frequency and music frequency such as 440Hz in music are not a concept. It is impossible to lower the high frequency music without the high voice, and the bass will not disappear because of lowering the low frequency.