Resampling in Batchprocessor question

What is the difference behind the scenes between these 2 scenarios (screenshots below)? If I set the output format of the batchprocessor to 96/32F, do I need to instantiate a Resampler in the Custom Plug-in Chain? I know that I don’t (I do this regularly); the result will be the format specified in the Output File Format field.

But are the audio files handled in exactly the same way in both scenarios? Same resampling algorithm used? (if resampling set to “best” in the Global Preferences)

Is there any reason to use the plug-in if you’re only doing resampling? I suppose if you needed to dither to 24 or 16 you would, because then you’d also add a dither plugin.



thanks,
bob weston
chicago

They should be exactly the same because they both use the Best setting. But you could confirm that by running a File Compare from the Analysis tab on two test result files.

There are the same and should give the same result, but “behind the scenes”, as you ask, the resampler is not located at the same place (“in the Master Section” vs “after the Master Section”).
This being said, form #2 is preferred because you can add a peak limiter after the resampler. This is important to limit certain peaks that could be risen by the resampling process (else you could get small clipping for waveform near 0 dB).

If the output format is 32-bit floating point, would you need to worry about new peaks above 0 after resampling? 32-bit float can store samples above 0dBFS, correct?

-bob

Yes, but delivery media use fixed point, so the issue will arise at some point.

Paul

Yes, understood. Thanks, Paul.

-bob