Hi Walter,

>>Also, it is not how many fields in the current table that is the sole factor. It is the number_of_fields x number_of_variables x number_of_loops that must be looked at.
>
>What I was trying to say is that the numbers of fields is the sole factor for each variable reference evaluated.

OK.


>>Like I said in the conclusion of my post: "if you have loop code where the number of fields times the number of records being processed times the number of variables in the loop can be a significant number then mdot is essential for speed if nothing else."
>
>Sure, However, I like to point out (for the lurkers) that often there is something that is way more time consuming than the ommision of mdot. If you are doing for example database access, or (complex) graphical output, you may find that the time consumed by those statements make the omission of mdot.

Absolutely. That is why it's important to really understand where the bottlenecks are. Fortunately in this case I had code in all the relevant places that I could switch on/off to record the time for the different sections of the processing.

Later