>>>oops...one omission....This behaviour is observed on Pentiums.
>>
>>
>>Sorry, what behaviour is observed in Pentium?
>>
>>I think maybe I should elaborate on my problems at my user's machine.
>>
>>1. sum result wrong in report.
>>
>>The list of data in the details band is already in 2 dec points and in the summary band, the sum is not correct. I have taken a look at the cursor before it gets printed, the data (with only 2 dec points) is exactly the same as those appear in the report. To me, it is like FPW26 goes back to kindergarden school and still makes mistakes in simple additions.
>>
>>2. negative 0.00 appear
>>
>>Well, FPW26's SUM() thought there is such a number called -0.00 and it is not equal to 0.00.
>
>If SET DECIMALS TO 2 does not correct the problem then change your sum expressions to SUM((INT(value*100)/100)) {you could also use 1000 or 10000 if you are concerned with fractional numbers}
>
>For some reason (that I haven't figured out yet), the Pentium chip performs arithmetic with more precision than FPW can use, but it can still 'see' or something like that. I don't understand WHY it happens, I just know that it does. Manually stripping decimals works but using FPWs built in functions does not.
>
>HTH

My problems turn out to be a hardware / network problem. The program is running fine on another model of Pentium II system.

Thanks for all the help.