Hi all,
I’ve been looking into the sample rate conversion routines in SDL’s audio
subsystem. I am wondering, do the automatically generated up/downsampling
routines do any kind of interpolation, prevent aliasing etc? Naturally speed
is paramount since these routines can get invoked a lot, but what kind of
quality level do the conversions aim for?
Kind regards,
Philip Bennefall
I’ve been looking into the sample rate conversion routines in SDL’s
audio subsystem. I am wondering, do the automatically generated
up/downsampling routines do any kind of interpolation, prevent aliasing
etc? Naturally speed is paramount since these routines can get invoked a
lot, but what kind of quality level do the conversions aim for?
They do some extremely primitive interpolation (average of the current
sample with the previous one).
I imagine there’s a lot of room for improvement here, without
sacrificing a lot of performance.
–ryan.