>>
>I wouldn't jump to that conclusion just yet. The VFP code posted by Kevin can be optimized and it's only up to Kevin to do it. Until then the test results are irrelevant. Nonetheless, the data test queries from .NET are impressive.. after all they're using VFP technology :-)
>>
>
>But that is not the point. What you have here is for the first time, Fox code that operates on Fox Data - having to be optimized to reach levels of performance reached by .NET...

I am not sure I follow what you're trying to say. Most probably you have not read the Fox code that Kevin posted and the comments to that. Actually, in spite of Kevin's efforts which I appreciate, it looks more like the test was worthless based on the given premises. It basically compares the SQL processing performance of the VFP OLE DB with the performance of VFP, which unsuprisingly resulted in almost similar performance.
FWIW, I only noticed that the VFP data engine performs consistently both in the OLE DB and runtime. By the time I write this, I guess you must have read the rest of the thread and reconsidered your statement.