The VFP Coverage Profiler may help you in identifing the problem. Check also Coverage Profiler 1.17.58 File #17507

>
>I have a large prg (app. 7000 lines) with mathematical and array operations. The prg is also working with local dbfs,
>but the focus of this on mathematical calculation, not working with data.
>
>Has anyone an idea, why a trial (with the same basic data) takes in FPW2.6 approx. 430 seconds
>and in VFP8 and VFP9 approx. 930 seconds, more than twice the time.