>>Naomi,
>>
>>I messed with the original message but it is interesting-
>>
>>

>>
>>? 24.93*.33
>>
>>? 24.93 *33/100 &&same result, right?
>>
>>

>>
>>Try this:
>>
>>

>>CREATE CURSOR cc (p1 n(11,2),p2 Y)
>>
>>lntotal=0
>>lyTotal=$0
>>
>>FOR lncounter = 1000 TO 20000 STEP 99
>>	lnNewval=lncounter/3
>>	insert into cc values (lnNewval,lnNewval)
>>	lntotal=lntotal+lnNewval
>>	lytotal=lyTotal+lnNewval
>>ENDFOR
>>
>>SUM p1 TO lnP1
>>SUM p2 TO lnP2
>>
>>?lntotal,lnp1
>>?lyTotal,lnp2
>>

>
>Interesting. So, does it mean I should use 11,2 (0r 9,2) in order to match exactly with my apparently not precise original calculation?

No, the sum has the same result.
The example is not correct because compare an (implicit) rounded sum with an not rounded sum

CLEAR
CREATE CURSOR cc (p1 n(11,4),p2 Y)

lntotal=0.0000
lyTotal=$0

FOR lncounter = 1000 TO 20000 STEP 99
	lnNewval=lncounter/3
	insert into cc values (lnNewval,lnNewval)	&& implicit round(,4) round(,4)
	lntotal=lntotal+round(lnNewval,4)
	lytotal=lyTotal+lnNewval	&& implicit round(,4)
ENDFOR

SUM p1 TO lnP1
SUM p2 TO lnP2

?lntotal,lnp1
?lyTotal,lnp2

I'll try this and see, if it makes a difference.