>
>>1) Performance. I *know* I'll come up with an application significantly faster with VFP against DBF's versus against SQL Server. Anyone with real world experience with this? His solution -- Hardware and denormalization.
>
>I did time trials on VFP, MS SQL Server, Oracle, and a AS400 with
>SQL queries (at the project I was on wealso barely ever had to
>update the tables). The dbf table was 1.8 gig. The queries were
>simple queries that were reflective of the ones we currently were
>using. VFP in terms of speed won hands down. There wasn't really
>any competition. (We chose anyways MS SQL Server because of
>the 2 gig limit and because it was from MS.)

I have a fair amount of real world experience with this. I've found that on a LAN VFP usually wins. Optimizing SQL Server is a lot more complicated than optimizing VFP. Lots of people move to SQL Server and find it very slow only to realize that they just didn't have it optimized well. Even optimized it's almost always slower on LAN. On a WAN SQL Server is much, much faster. There are a million variables here. If you have a kick-ass server and slow workstations SQL Server might be faster. If you have lots of power on the desktop and a fast network VFP will almost certainly win. The key is to balance performance requirements against other issues like security, replication, WAN performance, etc. You have to decide what's best for a particular application.

>>4) Network traffic -- Yes, there's less network traffic with SQL Server.
>
>How many users? How many hubs and routers? I've seen network
>bottlenecks that were quickly cured by additional routers and hubs.
>

Again, WAN's are a big issue here.