ff123 wrote:
So far no one else has reported any problems with 0.12.5b -- which is not to say there couldn't be any, but there could also be other causes.
Umm, there's that 25 to 30 msec truncating problem at the end of files. Since Magnus's problem is also at the end of files, the two problems could be related.
The truncation problem derives from the fact that the last audio frame is (currently) not decoded unless there is something following it, such as an ID3 tag. The technical reason for this is *precisely* to avoid the possibility of a memory error leading to a crash, because it's possible for the process of decoding to read a few bytes past the end of the current frame under certain (perverse) conditions.
Magnus's problem could be related, yes, but it is an orthogonal issue.