>We have a few people here who have tried VFP applications on Win2k compared to previously running on NT. They say that Win2k is noticably slower (very noticably) but this conflicts with what I've heard that Win2k should be noticably faster than NT.
>
>Who's right here? We're close to pulling each others hair out here.

My experience is that on a machine with adequate memory and equivalent file systems Win2K wins, hands down. OTOH, my idea of an adequate Win2K box starts at 64MB and goes up rapidly from there as apps compete for RAM, and if you don't take steps to curb VFP's appetite with SYS(3050) it will force swap file activity in almost all cases where you deal with any significant amount of data. NT will run on a smaller machine. My net at home runs NT Server adequatey for light data loading with an IE and VFP development with 128MB- adequately meaning that it switches apps smartly and responds faster than I type most of the time (IE is sluggish at times) and is noticably slower running Win2K Server. Change from a 128MB system to the identical hardware and 256+MB, and no question, Win2K is faster. Aside from my laptop and my test platform box, intentionally left as a 32MB Pentium 200 for testing (with Win9x), everything has at least 256MB in it. RAM is lots cheaper than dinking with hardware tweaking - an added 64MB costs less the time to tweak NT properly at this point in time in places where it would count, and Win2K is easier to install and manage.

Depends on your perspective of what you think should be used to run things, clearly - in theory, you can run NT in 12MB and Win95 in 4MB. Win2K's stated base is 64MB and a P166 as I rememeber...