[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 #8 from sebastien.fenet at yahoo.fr ---
(In reply to Sebastian Dröge (slomo) from comment #7)
> Let's divide by the bigger number then :)

I agree with this. I don't think it would be very nice to have an asymmetric
conversion, depending on the sign of the input sample (linear PCM assumes that
the space between each quantified value is equal).

If you are interested, there's a listing of various softwares' methods here:
http://blog.bjornroche.com/2009/12/int-float-int-its-jungle-out-there.html

You'll notice that it also raises the question of the conversion back... Should
we multiply by the lower number to ensure that we will not overflow INT16?
(which means we sacrifice bit-transparency...).

There's no perfect method anyway, since we got from one asymmetric range to a
symmetric one.

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