On 23 Jan 2005 at 20:22, [EMAIL PROTECTED] wrote: [quoting me:] > >On 22 Jan 2005 at 21:31, [EMAIL PROTECTED] wrote: > > > >> My theory is that at this point, there is a problem with the way > >> Finale has been programmed. . . . > > > >I think *that's* indisputable! > > > >> . . . A problem, which results in the Overwrite > >> Bug, that when there is more memory in a machine, doesn't occur. > >> But when there isn't enough memory and the system gets to the > >> point of thrashing, then the mistakes made in the way that Finale > >> was written result in this problem. > > > >I don't know that anyone has identified low memory as a requirement > >to produce the bug. > > That is true...no one has identified that for sure. But given the > way the machines act about the time the bug bites, and Christopher at > least has said that it has never happened to him on a newly booted > machine, to me, it seems possible that there may be a correlation > between memory and this bug...but I'm also just speculating :-)
I think you're mixing unrelated reports (see below). > >> Can you think of anything in the way a program is coded that would > >> fit into this scenario and possibly cause the overwrite bug? > >> Problems with the way the program uses threads? I/O issues? > >> Caching? etc....? > > > >I can't think of a reason why low memory could suddenly cause > >corruption of such a minor type. When a program can't handle low > >memory conditions, usually the results are fairly catastrophic. > > Well, for those that are losing the work, it can be catastrophic!! :-) Well, yes, from a human point of view that's catastrophic, but Finale cannot tell that there is corruption of data, because it's not *structural* corruption, but content corruption. That kind of corruption simply can't happen from something like virtual memory problems because it's not a random enough result. > And I think that Christopher mentioned that after the bug hit on his > machine, Finale crashed shortly after that. >From low memory or from some other cause? > I just looked back at a couple of posts one from Chuck and one from > Mark. Chuck mentioned that his system seemed to start to slow down > after a long Finale session and that is when the icon for the selected > tool in the tool palette turns into the Finale icon. (I've had this > happen to me too.) Mark had a similar thing happening with icons > appearing incorrectly....with the wrong graphic and then a garbled > graphic. His system was also in the state where it seemed that memory > was "tight" as he put it. Were those reports of the file overwrite bug, or independent reports of low memory/performance problems. Has Darcy reported low memory as a part of his problems? And if low memory is the *cause*, it, of necessity, must be present every time the bug hits. > >Of course, it depends on the type of low memory condition. > > Can you elaborate here? It depends on your definition of "low memory." Certain kinds of non- RAM memory structures can run low and that gets reported as low memory (the details of that are specific to the OS and applications, though). Many times the term "low memory" is used when what is meant is "low system resources." -- David W. Fenton http://www.bway.net/~dfenton David Fenton Associates http://www.bway.net/~dfassoc _______________________________________________ Finale mailing list [email protected] http://lists.shsu.edu/mailman/listinfo/finale
