>>>So there can be suboptimal vfp code, which shows *more* of the bad side effects on SSD and Raid 5.
>>>But that should not result in "serious issues" (trace logs, better definition ???)
>
>Actually one of the things I did was use LAN_SpeedTest.exe to prove that the READ was slower than the Write and that speed varied enormously over time (up to 50% variation).

unless a write-cache controller is involved, that does not sound right as well...

mystified

thomas