Skip to main content
Topic: [IRRESPONSIBLE QUOTING] Re: Dither vs Sampling Rate (Read 322 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

[IRRESPONSIBLE QUOTING] Re: Dither vs Sampling Rate

If one dithers to 16 bit, then it would seem consistent to dither again after processing (done at a higher bit depth) and exporting to 16 bit again.

If for example a 16 bit wave is dithered, the randomization produced by the dithering is represented when it is upsampled to a longer word length. If it is then downsampled to any word length 16 bits or more, the randomization of the quantization error is preserved.

For my examples, I generated a 1001 Hz tone at 24/44.

I downsampled it to 16/44 with dither.  The quantization error was randomized per spectrum analysis.

(1) If I upsample the 16/44 wave back to 24/44 and then downsample to 16/44 without dither then the 16 bit quantization error remains randomized by the original 16 bit dithering per spectrum analysis.

(2) If I downsample from 16/44 to 16/32 without dither then the 16 bit quantization error remains randomized by the original dithering per spectrum analysis.



 
SimplePortal 1.0.0 RC1 © 2008-2019