Hello, long time no blog!
Today, I want to talk in brief about my final project for my masters which is designed to be a piece of research in the general area of audio production. This is an informal post, meant to be gloss over certain points in an effort to keep things short and to the point in an effort to make the topic as accessible as possible. Once the project is finished, I will be able to make new posts about each part of the project in more detail! If you have any questions, just get in touch through here. The final dissertation can be read here.
Recently, I recorded a choir for surround sound playback. What this means is that when listening back to the recording in a room properly equipped, you will feel inside the room where the music was being performed. The sound will envelop you from all around you, simulating what it would have been like to be in the concert hall which is a very cool experience. In surround sound you have five speakers. Three at the front for the left, centre and right with two are the rear. Check out the below image from Sound On Sound Magazine about the placement of these speakers.
http://media.soundonsound.com/sos/jan01/images/surround2.gif
Recording Arrays
Many ways of recording surround sound have been developed over the years. The most common are what I call traditional arrays which is a catch all term for multi-microphone recording arrays. If you take note of the image above, there would be one microphone for each speaker. In general, the left, centre and right point at the respective areas of the stage and the rears point into the rear corners of the space. What the microphones “listen” to then gets recorded and played through their respective speakers outlined in the picture.
The collection of these microphones is called an array and then can usually subdivided into the front array and rear array as there can be a larger spacing between the front and rear. If you are at a concert and stand at the very front, you will get a great clean sound. If you stand at the back you get a more reverberant sound so an aim of these types of arrays is to capture both as best possible for use in the production. A pleasing recording for the listener can be achieved by placing the front array fairly close to the performers to get a clear sound and then by placing the rear array into something known as the reverberant field.
Here is a photo of a traditional array called INA5 from www.sanken-mic.com. You can see the distances and angles involved, especially between the front and rear.
The Soundfield
For the recording engineer, traditional arrays have around five microphones which means there are a lot of cables which are sometimes quite long, a lot of stands which are usually heavy and/or wobbly, a lot of measurements and angles which can be cumbersome to get correct and then possible headaches to worry about when setting everything up, for example, someone walking into or moving the stands.
There is a relatively new microphone called The Soundfield microphone. Roughly speaking, this is 4 microphones in one. This can be placed in a recording environment just like the front section of a traditional array can. The Soundfield microphones pickup can be seen as mainly based on figure of 8 microphone pattern, to keep things simple. A microphones capsule is in the shape of a big coin and listens in certain ways around it. Some microphones listen to just what is in front and on either side of it. Others listen all around it. The figure of 8 listens to the front and back while ignoring the sides. Imagine two tennis balls placed on either side of those big chocolate coins, this gives an impression of what directions the microphone is picking up from. Here is a photo of a microphone capsule from recordinghacks.com
Here is a diagram of a figure of 8 microphone from the Recording Review Forum, think of those tennis balls.
Basically, these types of microphones are listening in a shape on front and behind the microphone while ignoring what is going on at the sides (includes top and bottom). For the sake of example, imagine there are two of you and if you say something and your duplicate says the exact same thing in the exact same way. If the two of you are placed either side of the microphone and you both say something, anyone that is listening to what the microphone is listening to will hear nothing. That is because one side the microphone listens in a positive way and the other negative, which doesn’t mean one side is happy and the other angry, what it means is that everything can be boiled down to numbers.
What this means means is that what one of you saying could gets turned into the number 1 and the other gets turned into a -1. When the microphone combines these things you get 0, or nothing being heard. With that admittedly odd example out of the way, the Soundfield works in a similar enough way. Depending on what way you add and subtract the signal that the microphone creates, you can hear what is happening at any direction. Think of it as a 360 degree security camera. If you are watching something from the left and then move to the right, you use a control to point the camera in that direction. The Soundfield is similar, but for sound and instead of the camera moving around, the mathematics are being changed to adjust the direction of what the microphone is listening to.
Why mention all this?
Well, it is easier to setup that a traditional array. One stand, one main cable and less headaches. More importantly, after the recording is done and you are mixing the recording you can change where the microphone is pointing as you wish. This is because you are combining what was recorded from the microphone, not having to adjust the direction its pointing in on the day of the recording itself as you would with a traditional array. On top of that you can derive the five different directions at the same time which are the signals you need for the 5 speakers in a surround sound setup. With modern technology and the ease of having a powerful computer for audio production, this can be automated which means no more messing with angles and protractors at 6 feet in the air! With a traditional array things need to be set up exactly and if you make a mistake or something happens to the array which you didn’t know about, you can not fix the problems in the mix!
The Soundfield is also fairly expensive compared to the five standard microphones you need for a traditional array. Expensive not only in terms of money but also in versatility as if you are an engineer who does recording work with bands and close micing, the five standard mics can be much more useful to you than the single Soundfield mic.
Here is a photo of one of my recordings with the traditional array of five microhones in red and the single Soundfield in blue. (very well drawn eh?)
What my project is about.
The Soundfield may be much easier to set up and it may be versatile after the recording in terms of changing its settings to fix issue but does it sound as good as the traditional setup? Remember when I said that the rear microphones of a traditional array are placed further back into the reverberant field? You can’t do that with the Soundfield. The Soundfield is one microphone in the sense it is a single enclosed unit. If you move it further back to get more reverb, you make the front more reverberant too and lose the clarity. Compromise! Additionally, the two concepts simply sound different to each other, not necessarily worse than the other but finding out what a sample of listeners think could help in making the decision between what to use or what to buy.
What I want to do is find out what is “better” when asking a sample of listeners. I intend to record a choir and set of classical musicians with a Soundfield microphone and a traditional array simultaneously. Then, I want to play sections of the songs in a subjective listening test where expert and not so expert listeners can sit in a surround sound listening room and decide what is their favourite, without knowing which is which. The result of this, paired comparison test, will hopefully highlight which recording method is the most preferred. That said, they could be equally preferred, showing parity, and that would not be a bad result. All that means is that the engineer can be faced with a choice of what type of sound they want rather than facing one which could have an impact on listener enjoyment.
Thanks
Thanks for reading, this was intended to be an informal and accessible look into the background of the project. More detailed and precise information can be found in the final dissertation here. If you are new to the concepts I outlined here and are interested, do let me know and I can guide you to more formal information. =)