I think I know why:
When Phex gets into the queue of a source, it assignes a segment of the file to that source.
When it reaches the end of the file, chances are high that there will be some sources where Phex is in queue.
It will then attempt to get the already assigned segments only from these sources, and this means it will wait to get to the front of the queue, which can take some time.
We already have a feature requests to change this behaviour:
ge.net/tracker/index.php?func=detail&aid=1794260&group_id=27021&a tid=388895
Maybe you can write your experiences there (politely). |