Game Development Reference
In-Depth Information
memory, if we use MP3, that memory will contain
lower quality
audio data that if we
use WAV format. Given that we are going to get at least an 8X data footprint reduction
during the sample optimization work process that we are about to learn in the next sec-
tion of the chapter, and that all our digital audio assets are going to be optimized to
between 4KB and 14K of data footprint, MP3 compression is not going to give us any
real JAR file data footprint reduction relative to the reduction in sample quality that it
will “cost” us. Game audio is short-burst sound effects and musical loops, so we can
use WAV file format and still get a good result, and not have to use
any compression
.
The other advantage is the data size of the WAV file that you see in our file manage-
ment software is also the amount of memory the sample will use.
AudioSampleResolutionandFrequency:Optimizingfor
Your Memory Footprint
The first step in the memory data footprint reduction process will be to take the raw
32-bit data sampling rate and reduce that by 100%, from 32-bit floating point data to
16-bit PCM data, as is shown in
Figure 15-5
on the bottom left of the screenshot. Find
the drop-down arrow in the gray information panel on the left side of the Audacity
sample editing area, which I have circled in red in this screenshot as it is kind of diffi-
cult to find if you are not used to using it. This will give you a menu that allows you to
set the data display (Waveform or Spectrogram) and set your Sample Data Format,
which is the submenu that we want to use to select the
16-bit PCM
option rather than
the 32-bit float option. Do not set your sample rate using this menu, as it will slow
down your voice (you can try this if you want to use this as a special effect later on).
Next, we will be looking at the correct work process for setting the sample rate.
