Over the last couple of years, I've begun purchasing music in "high resolution" format.
In the past, I've used 16-bit 44.1 KHz lossless audio files. They could be easily manipulated with DSP effects.
However, some of the audio files I use now are 24-bit 48 KHz or higher. This gets more complex because it requires additional processing steps.
GTA V only accepts AAC-LC .m4a audio files, and they must be 16-bit 44.1 KHz CBR. To make matters worse, the audio engine mixes custom music quietly. This is bad, because it has to compete with everything else happening in the in-game environment, especially vehicle engine noise. Tracks with high dynamic range would be drowned out by ambient noise, so dynamic range compression is necessary (yes this is sinful, but it's justified by the use case).
To make custom audio tracks work decently with GTA V, here are the DSP effects I'm considering using (in no particular order):
What is the best order for this DSP processing? I don't want clipped/distorted samples, but I need it to be "loud", or as close to dBFS without clipping or brick-walling the audio too much. Yes, audio quality isn't the most important thing here, but I want to have the best quality possible, as the game does not support lossless audio.
In the past, I've used 16-bit 44.1 KHz lossless audio files. They could be easily manipulated with DSP effects.
However, some of the audio files I use now are 24-bit 48 KHz or higher. This gets more complex because it requires additional processing steps.
GTA V only accepts AAC-LC .m4a audio files, and they must be 16-bit 44.1 KHz CBR. To make matters worse, the audio engine mixes custom music quietly. This is bad, because it has to compete with everything else happening in the in-game environment, especially vehicle engine noise. Tracks with high dynamic range would be drowned out by ambient noise, so dynamic range compression is necessary (yes this is sinful, but it's justified by the use case).
To make custom audio tracks work decently with GTA V, here are the DSP effects I'm considering using (in no particular order):
- Resample to 16 bit, 44.1 KHz
- Normalization to -0.01 dBFS
- Apply dynamic range compression/limiter (depending on dynamic range)
- Convert to AAC-LC 256k CBR .m4a file (iTunes QAAC encoder)
- Add the required metadata tags manually
What is the best order for this DSP processing? I don't want clipped/distorted samples, but I need it to be "loud", or as close to dBFS without clipping or brick-walling the audio too much. Yes, audio quality isn't the most important thing here, but I want to have the best quality possible, as the game does not support lossless audio.
Comment