>Thank you very much Ed!
>I consider your reply a real class about the topic!
>Do you know any material which delves more deeply into the inner-working of VFP?

The OnLine Help file contains a tremendous amount of information, particularly on the subject of what VFP uses to resolve a query (there are discussions of the mechanisms for Rushmore, the query processing scheme which can make use of index tags to determine how to satisfy a query); some of the information is common sense - if there's no alternative to reading the file record by record, there's no choice in the matter. It discusses the structure of a table - what file(s) are involved, the information contained in them and the internal representations of data, especially Chapters 15 and 28 of the Programmer's Guide, the Appendix of the programmer's Guide, and the API Library Construction section of the Reference - Chapter 15, Optimizing Applications, discusses Rushmore and table access optimization from the VFP programmer perspective, Chapter 28 and the API Library Construction sections are targeted at C programmers; the Appendix discusses exactly what the various file structures for tables, memos, indexes and the like are.

FoxTalk magazine has discussed some of the issues of performance and implementation in its articles. From the standpoint of technical content, FoxTalk has the most detailed coverage of the environment.

A number of books from Hentzenwerke, particularly the "Hacker's Guide to Visual FoxPro 6.0" provide a lot of information.

Beating your head against the wall frequently helps, too.

I've come into VFP development from an unusual perspective - I was a systems programmer wrking in low-level operating system implementation long before I started working with database systems. Understanding what goes on below VFP in the OS helps a great deal - understanding what Windows is doing in terms of memory and disk services helps understand what makes VFP's data engine work.