The first parameter is the rounding precision (1 rounds the nearest integer, 0.1 to the nearest 0.1, etc). The offset is an amount added to the rounded value after rounding (0 by default). I must admit that I have never used the offset in any QV model that I have developed.
Hope that helps
Ok, I tryed this:
Round(-7.4, 1, 0.3)
so let me see if I got this stright; the idea of the offset is to modify in a way the steps at which the rounding is calculated....
so, for the former example, I would get -7 if no offset was added,,,,
but since 1-0.3 = 0.7, this would be the new step used to calculate the rounding.... then, -7.4 is closer to -7.7 than it is to -7....
Is my reasoning right here? thank you in advance!
My understanding was that the offset is applied after the rounding. But after studying your example in your original post, this is what I think is happening.
First the offset is being applied to the round precision to get the rounding points
Around -7.4, these would be -6.7, -7.7 (the steps remain 1, but offset by 0.3)
Then the value (-7.4) is being rounded to the nearest value (7.7)
Also look at 4.6
The rounding points are 3.3, 4.3, 5.3 ... (step 1, offset by 0.3)
The nearest rounding point is 4.3
Does that make it clearer?