The Ultimate Recording Method
By Robert Dennis from April, 1994 Engineer's Quarterly Article
What is the ultimate way of recording and what is the ultimate recording system? In this age of change in recording methodology, you'll get a lot of opinions. The first big debate is whether to use analog or digital recording. The second biggest debate is whether to use analog or digital control of the signal, especially in the area of signal processing such as equalization and compression. It is easy to find one professional who will vote for analog and another for digital. Is one set of professionals right and the other wrong?
Let's look at some of the pros and cons of each of these recording elements.
Analog Recording:
What is it?
The continuous changes of energy in a sound wave are converted to continuous changes in electrical energy giving an audio signal that can run through wires and consoles. The tape recorder converts the continuous changes in electrical energy to continuous changes in magnetic energy. Those changes are recorded onto a moving strip of material that can be magnetized (recording tape).
Disadvantages:
Analog recording is inherently noisy. Analog recording inherently changes the waveform shape, causing distortion. The random magnetization of magnetic materials and their inability to exactly magnetize the way that the incoming magnetic energy fluctuates causes the inevitable noise and distortion.
The noise made by analog recording is somewhat masked by the recorder boosting the high-frequency energy of the signal before recording and then lowing the high frequency energy by the same amount on playback. Even with this built-in scheme to reduce recorded noise, analog recording is noisy, producing at least 256 times as much noise as the console and the other electronics that the signal must flow through.
The circuits that boost the high frequencies in the analog recorder during recording cause the majority of the objectionable qualities of analog recording. The objectionable qualities I am referring to are overload and phase distortion.
Overload:
Boosting the high-frequency energy before recording means that overload and distortion will happen to the high-frequency component in the audio first. With multitrack recording we are recording things like cymbals that have severe amounts of high-frequency energy and they distort fast. On the vocal track, we will notice the "S" sounds (high frequency in nature) distort first. As a matter of fact, even at low levels the high-frequencies have many times the rate of distortion of low frequencies as an inherent part of analog tape recording.
Phase Distortion:
The circuits that boost the high-frequency energy in analog recording cause a phase-distortion. Phase distortion causes a slight difference in time between the high-frequency energy and the low-frequency energy. This results in the transients (sudden bursts of energy as a waveform begins) coming through at a different time than the body of the sound. It also results in the harmonics (the frequencies that give distinctive quality to instruments) coming through at different times than the fundamental frequencies (those that give the pitch recognition and body). The end result is that the recorded sound is less life-like and has a "flat," "recorded" sound.
Advantages:
Every musician and most recording engineers will tell you that some kinds of distortion tend to be an enhancement and other types of distortions are a distraction. Almost all distorting devices add harmonics that are added are considered "pleasing," the presence of distortion may be considered to enhance the sound. Analog tape recording distortion is described by many to be pleasing, if there isn't too much of it.
How quickly something goes into distortion is another factor. If a device or system goes into severe distortion immediately when the signal is a bit too high, the device has a harsh overload characteristic. If the device, like analog tape, goes gradually into distortion, the effect can be pleasing. Many more things than you expect operate with some degree of overload. Analog recording is a prime example of this and this quality makes the analog recording more pleasing to the ear for many professionals.
Digital Recording:
What it is?
In digital recording the incoming waveform is "sampled" every 44,100th of a second and measured as to its volume level at that instant. The sample's level for that instant is converted into a number that describes the level. The number is stored as a series of pulses; the spacing of the pulses determines the number being stored. On playback these pulses are fed back into a converter that makes "spikes" of audio waves at the correct level. Finally the recorder smoothes these spikes back into the continuous waveform.
Disadvantages:
In digital recording, any signal too high in level for the system causes the system to run out of numbers to describe the level. ANY amount of overload results in severe and immediate distortion of the worst kind. You just don't overload digital recorders by ANY amount if you want a useable recording.
While it's true that digital recording has very low distortion, the kind of distortion it does have is very unmusical. The result is that small amounts of distortion affect the listener much more than analog distortion. Many people feel that digital recordings have a more irritating "edge" than analog recordings. The most recent recorders are less objectionable.
Advantages:
One of the biggest advantages of digital recording is lack of noise. The amount of noise made by a digital recorder is comparable to the noise of the console and other electronic items that the audio must run through.
Another big advantage is the lack of phase-distortion and the low level of distortion on the high frequencies. Because of these factors, the sound seems very lifelike and clear. These qualities are very noticeable especially at high frequencies.
Digital vs. Analog Audio Signal Handling:
Analog signal handling is done while keeping the signal in its waveform state of continuous energy changes. Mixing and level control are accomplished by resistors which allow a certain portion of the energy through and different energies to combine in certain proportions. The resistors resist energy by the same proportion regardless of its frequency or initial level. Equalizers (Tone Controls) have elements that resist energy at some frequencies more than they do others. Compressors are "automatic level controls" which change how much they resist energy depending on the level of the signal coming in. Your traditional recording/mixing console, equalizers and compressors are all analog.
Digital signal processing is accomplished by changing the incoming audio signal to numbers which represent the level of the audio waveform every 44,100th of a second. In this aspect, the digital and recorder are alike. All controls of the signal for things like level, equalization and compression are done by number-shifting. It works approximately like this: to increase level by 1 dB (approximately the smallest change in level that can be heard) the system has to multiply each of the numbers by 1.12. To achieve equalization, it only multiplies numbers that define a range of frequencies. To achieve compression, it divides numbers higher than a certain amount. In short, mathematical functions are applied in some pre-determined manner to the numbers that define the waveform. This is simple for a digital system to do.
The biggest difference in quality comes in the difference between the analog and digital equalization.
Analog Equalization:
What it is:
Any equalization is an increase or decrease of level for signals of some band of frequencies. After the equalization is done, this band of frequencies will by higher (or lower) in level than signals of other frequencies.
Analog equalization achieves this by putting elements in the circuit that impede (resist) signals of one kind of frequencies more than other kinds of frequencies. This extra impedance (resistance) lowers the level of those frequencies.
Advantages:
There are no real advantages to analog equalization except that people are used to it. Digital equalization sounds different and, as "the new kid on the block," is rejected by some. Anytime that a system changes the sound in a certain manner, there will be people that like that method better. There is actually nothing wrong with this approach. When you have a device that gives you the sound you're looking for, you simply use that device regardless of what is "supposed to be" better.
Disadvantages:
The same elements that resist one set of frequencies more than another cause a time delay for those frequencies. The result can be a "less-real" or even "harsh" sound. The same disadvantages of analog recording's phase-distortion apply to analog EQ and even more so.
Digital Equalization:
Digital equalization has been pretty much explained. It has no phase distortion and has many of the advantages listed for digital recording. If you want something to sound brighter or fuller, digital equalization will do this without it sounding "false." The only big disadvantage of digital equalization is its price (which is coming down) and its availability (which still needs to be worked on).
To Be Digital or To Bo Analog:
Recently, I decided to do all of my multitrack productions digitally. I felt that the new breed of digital recorders (read Tascam DA-88) on an overall basis has a far better sound. With these new recorders, digital recording was much less expensive and I jumped with both feet into being a producer that only dealt with digital recording.
However, I was in the middle of some projects that started out analog. I added some digital voices to these projects and mixed them. I was astonished by how good they sounded. It made me realize that there were elements of analog recording that I liked. Without changing my mind about doing new productions digitally, I began to think. What is the ultimate format of recording? What if cost was not a factor? What if I could do anything I wanted without restrictions? What would be the ultimate recording method?
Other engineers who do what they want have tended to combine digital and analog. One prominent engineer insists on cutting the drums and bass analog 16-track and doing the rest of the recording on 32-track digital. Many engineers would record on older analog ("classic") console and mix on the latest and greatest consoles.
I mad up my mind. THE ULTIMATE METHOD WOULD BE TO RECORD DIGITAL AND ANALOG.
The Ultimate Recording System:
The ultimate recording system would consist of the following:
Advantages to the "Ultimate" Method:
If we pull out all the stops we have to realize that some things sound better analog and some things sound better digital. If foot drums and toms sounded better digital than the "heavies" of the industry wouldn't cut them analog. Vocals are the center of attention in 99% of the recorded music and there is 256 times the noise and "edgy S-sounds" because analog can't take the energy. This makes the noiseless, clear recording of digital very desirable.
The idea of blending all instruments with a combination of analog and digital sound is, to my knowledge, not being done today. This technique is an outgrowth of blending different amounts of different types of reverb during mixing. When one instrument has one type of reverb that is not present on other instruments, it tends to sound like it was recorded in a different room. Having a little bit of that type of reverb on the other instruments tends to make the mix sound like one performance. Having some instruments recorded analog and some digital tends to make the different instruments sound like they were recorded differently and at different times. Blending both techniques together helps the illusion of one technique, one recording and one performance - the Ultimate image.
Analog is dead (or dying), long live analog and digital combined!
Copyright 1994, 1999, Recording Institute Of Detroit - All Rights Reserved