Originally Posted by marcostt
I see... the problem is I can not see a pattern of behavior... sometimes I can work and save for a long time, and another I just can save once, and in the second try to save crashes. Maybe it crash more often when I move images and try to save after that, but noy always.
This is why Valloric is asking you to spend the time to experiment and find a way to always cause the crash. Because otherwise he just has a report that it sometimes crashes on save, and if it never does it on his system, the bug will take a long time to be fixed.
If you do spend the time and come up with a reliable way to cause the crash, it can probably be fixed very quickly.
It is possible that it's a random crash - dependent on some bit of memory have a particular value that isn't set by the code. In that case, the best you can do is find a sequence of events that will (say) cause the crash to happen one time in three. Even that would be a lot better as a bug report.
It's a sad fact, but reporting a bug well is often as much work as fixing the bug.