Hi everyone!... I was wondering whether there is any point to converting your audio samples to 24 bit. I read somewhere that it improves overall quality even though the audio was originally at 16 bit and I was wondering if anyone can shed some light on that. My thinking is that once a sample has been recorded the quality cannot be improved and that only the format may be changed...and I fail to see where or how sound clarity can be bettered simply by enhancing the format of a good or mediocre sample. You are not introducing any new data into the sample that should alter the sound so.......help?
Would there be a difference if you changed all your samples to 24 bit as opposed to waiting till an entire project is done to convert...I know you should work in a higher bit environment from the get go and finally MOVE DOWN to 16 bit, but in case you can't do the 24 bits from the start...? Are there any merits here
Samples and Sample Conversion
-
- KVRAF
- 3345 posts since 8 Nov, 2003 from Amsterdam
No. It doesnt improve quality... It only adds a couple of zeroes, and effectively makes your sample directory 50% larger.
But: Once you start using them in a project, and process them (different pitch or even just the volume/gain/balance), then it's smart to store the resulting files in 24 bit.
Quoting a famous writer: If you put additional zeroes to 1.51 (1.5100), the quality does not improve. But once reduce the volume by 50%, it will be 0.755, and you need the extra digital position. (I hope I didn't confuse the English and Dutch decimal ',' and '.' here, which are reversed, because then the example won't work
)
But: Once you start using them in a project, and process them (different pitch or even just the volume/gain/balance), then it's smart to store the resulting files in 24 bit.
Quoting a famous writer: If you put additional zeroes to 1.51 (1.5100), the quality does not improve. But once reduce the volume by 50%, it will be 0.755, and you need the extra digital position. (I hope I didn't confuse the English and Dutch decimal ',' and '.' here, which are reversed, because then the example won't work
)-
- KVRian
- 769 posts since 2 Apr, 2005
But in a VST sampler, internal processing is 32 bits anyway. I presume the conversion is done at the point where the sample is loaded into memory, so I see no value in changing the storage format.