>>G'day.
>>Got VFP6. Thinking about using it to deliver the (nearly-finished) app in runtime, rather than VFP 5.0a. Especially with my wierd "invalid seek offset" problem. Question: Is it stable enough? Would any of YOU ship a system based on VFP6? It SEEMS to run just fine.
>>
>>2nd question. Performance. Subjectively, it seems slower working with the screen. And I have a quantitative measurement as well. I have a VERY complex form (a class scheduling calendar) with a LOT of textboxes on it, and a fair bit of refresh code in each. On my Pentium 120, using VFP 5.0a, it takes 1 second to refresh the form (I have timing code in it). Under VFP 6, the same form takes 1.8 seconds. I haven't started doing anything yet to see why the big difference (I haven't changed a byte of code), but if this is the way VFP 6 writes to the screen, I probably won't be able to use it for this app - the 1 second refresh is just acceptable - nearly twice that isn't. Anyone else noticed any performance issues with VFP 6?
>>
>>Larry
>What is this needing to refresh text boxes? can you explain a bit more?


Lets be honest here. The issue isn't refreshing the text boxes - sure Larry might be able to speed it up a little by tweaking the code here and there, but 6.0 refreshes slower than 5.0 and 5.0 slower than 3.0. I've got numerous screens and noticed this as soon as I got ahold of 6.0. Put this issue to test on a page frame and see what happens. I set a timer on my screen and add objects to the other pages after the screen displays initially. Visually - it appears to be as fast as I once had it in 3.0 but even after it displays I don't allow you to do anything till all the objects are added to the pages and the data is bound to the form. Visually 6.0 is not a winner in my book, but I can make it do anything - so I've opted to stick with it.

Robert