[Bug 755301] audioconvert: Integer->Float conversion creates values slightly smaller than -1.0

[GStreamer (GNOME Bugzilla)]

https://bugzilla.gnome.org/show_bug.cgi?id=755301

--- Comment #15 from sebastien.fenet at yahoo.fr ---
Hi Sebastian,

Could you enlighten me about the workflow in Gstreamer?

My initial usecase was to convert a 16 bits Integer raw audio to Float.

>From what I understood in your comments (correct me if I'm wrong), Gstreamer
converts, at some point, the 16 bits Integer audio to 32 bits integer.

And then, it goes from this 32 bits integer to a 32 bits float by dividing by
the factor you said before. 

Is that correct?

If it's indeed how it works, the problem is, in my opinion, this intermediary
conversion to 32 bits integer. I actually don't see its practical interest and
it will, for sure, pose a problem a resolution when you try to go from integer
to float with the same number of bits...

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.