[Bug 755301] Problem with audioconvert from INT16bits to FLOAT64bits

[GStreamer (GNOME Bugzilla)]

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

--- Comment #5 from Sebastian Dröge (slomo) <slomo at coaxion.net> ---
Thanks, the problem here is that we divide by 2147483647.0. Which is the
maximum signed integer value for 32 bits, but -2147483648 is also in range. So
-2147483648 would be converted to something slightly below -1.

Want to provide a patch for fixing these conversions? Alternatively one could
also clamp the results to the [-1,1] range instead of changing the factor we
use for the division.

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