Just a side note here. Several of the VFP 6 commercial applications I support experienced significant problems when they were ported from VFP 5. After extensive work with the Microsoft VFP development team on the 'confirmed' problem the developers of one of the VFP applications ended up 'programming around' the bug - i.e., Microsoft couldn't fix it. The closest thing to an explanation about the nature of the bug I have been able to get from either party (Microsoft or the Application Developers) is that a great deal of 'low level timing' code was changed in VFP 6 resulting in GPFs on many workstations (the exact configuration needed to reproduce the error is still a mystery). This explanation is woefully inadequate and leaves me and my colleagues with one of those empty feelings in the pit of your stomach. Microsoft still has no fix after 6 hard months of investigation - one of the suggestions was to run the application with the debugger turned on I assume so that the internal processing is slowed down to the point of not hitting whatever this 'timing' bug is - we laughed at that one.

My point is that even with Microsoft's undivided attention on any given problem you won't necessarily get a fix and be forced to work it out yourself by changing the way you code. I say, "If it ain't broke don't fix it." Too many improvements being put into the products in too short a time frame can be devastating to the developer base.

>I would like to thank George, Erik and Ed, for the reassurance that VFP (in general) is OK with Win2K, and everyone else that has contributed to this thread. I am still stuck with NT4 until the end of the year with my present client, but have a personal copy of Win2K Pro at home (but not the time to do
>anything with it).
>
>In particular I thought George's reply about the timing of Win2K's release and VFP 6 was particularly cogent. There can be little doubt that you (Colin) are experiencing the exception to the rule. That said, I would still like the opportunity to learn from your problem - there are still unanswered questions:
>
>Does your application:
>- Use frequent FLUSHing?
>- Open and close tables frequently?
>Digressing, if tables are already open, and a form's DS does the equivalent of a USE AGAIN - will this be faster than if the tables were not already open?
>- What kind of buffering is your Form DS using?
>- Indeed, are you using the Form's DS, or relying on the global VFP DS?
>- How many tables does your application typically work with?
>- What is the nature of the application - very large and sophisticated, or would you describe it as run of the mill? (Difficult to answer, I guess, because of undefined metrics).
>
>Regards,
>Houston