Hi Paul,

SNIP
>
>
>What MS might have done in VFP8 is to make the FLUSH work not just in VFP's cache but also to flush the workstation's operating system caches. That would conceivably make a FLUSH take longer.

Hmmm, if that's what they did, then it will deserve prominent mention in the VFP8 "What's New - Behaviour Changes" section, for sure!

I've never been particularly bothered myself that the command does not actually cause data to be written right away. I've found it acceptable that it SEEMS to mean 'schedule the data to be written by the OS' and interpreted its purpose as a way to "override" SET REFRESH settings. I'm generally happy that the OS will take care of it asap, all things considered.

I hope that someone with the beta also tries this and reports the result, and if that esult is similar, then I hope someone from MS will help to clarify.

I'll tell you this... even if the OS cache, VFP cache and VFP buffers were all being cleared after each FLUSH, the time still seemed toooooooo long. Clearing memory happens real fast, and these are ALL INSERT INTOs, so prior cached data is irrelevant all around.

thanks for the info.