Technical Requirements

Before I start to record anything or even consider the process I thought it would be a good idea to look into the technical settings and procedures used when recording foley for the industry. All sound captured nowadays on modern field recorders is captured digitally in binary code and that comes hand in hand with varying quality. 

Digital capturing of the sound is basically a capsule taking a snapshot of the raw analogue signal and converting it into a digital signal. When doing this process you can control the amount of detail captured in each snapshot which will get you closer to the original raw sound but requires more storage space.

The goal is to get enough detail in the sound without recording obscene amounts of data and filling up all your cards. Not to mention the larger more detailed sound files hitting the CPU harder when it comes to editing and mixing the sound. 

To research this I found this article http://www.resoundsound.com/sample-rate-bit-depth/ very helpful. He mentions the different sample rates and bit depths and where and when they should be used. 

He and many others have referenced the standard working format for sound to be 24bit/48KHz as a standard to work from. Often the sound will be down sampled somewhat for release to allow more space on the DVD for other things. So that was my target and I'm fairly sure that is the quality that the sound recorder on set used to record the audio for the dialogue.

I have decided though to go higher than that when recording the foley as it will allow me more depth and control.

The sample rate and bit depth basically control the amount of detail in the captured audio and work similarly to the pixel density of a video/screen. The sample rate controls the amount of times a second the audio is being recorded and monitored. The bit depth controls how much information is being captured in each of those events.

So a 16bit 48KHz sound files is capturing audio at a rate 48KHz with a depth of 16bits. Audio captured at 24bit 96KHz is capturing audio more times per second and each time it captures audio that sample has more definition and fidelity.

It is like capturing a video in 720p at 24fps. It will be capturing 24 frames per second at 1280x720 resolution. If you captured the same thing in 1080p at 240fps, you would be getting 10 times as many frames per second and each frame would also contain more detail.

Staying with the video metaphor. If you were working on a project shot in 720p and you brought in a 1080p file, it could still be used and you would have more control and room for play with the file because it was shot at a higher quality in the first place.

For this reason I am choosing to record all of the foley assets in 24bit/96KHz as it is a very high quality sound file being captured that will give me a lot of room to play and more room to shape the sounds to fit into the current audio landscape.