Okay, time to solve this for once and for all ...
But things are going on, of which I currently don't know what it is. So please help me a bit;
First of all people earlier (months ago) reported about what you just told, which is not consistent with "in 0.9j it was ok".
So : it would be helpful if you could determine the last version which was okay
for youOn the other hand, you could skip this if you could email me the smallest track you can find, which should be sufficient to copy the behaviour (and please note I refuse to install/activate WMP on any machine I have, so now I'm not "able" to create those files
).
Do not cut the source, because
a. the program concerned will reassign the "header data" and it won't be a representative of the problem
b. when you'd really cut it (with an editor) the header data won't be consistent with the audio data.
Sidenote : Engine#1 and #2 as well as other software are able to deal with, say, inconsistencies because errors in the header data can be tweaked out during playback. In Engine#3 this, so far, was not done.
Thanks,
Peter