Viewing 15 posts - 4,096 through 4,110 (of 6,036 total)
Or you don't insist on presision on those numbers anymore?
And now you do agree with me that 0.66666666 is valid perpresentations for all numbers between 0.66666666(0) and 0.66666666(9), including 2/3...
June 11, 2007 at 4:42 am
Are you sure .66666666000000000000 = 2/3?
I checked:
@d * @b-2 = 1.99999998000000000
It's not = 2.
So, can you please provide script where BR is rounding 2/3, but not any numbrer around?
June 11, 2007 at 4:24 am
That's funny:
Bankers Rounding cannot round value 2/3.
Can anybody provide a code disproving this statement?
![]()
June 10, 2007 at 11:51 pm
No, that's set of data YOUR test provided.
My test did not convert DECIMAL(8,3) values to money, so it would not fool anyone.
In my test it was:
Level of precision:
4 digits ...
June 10, 2007 at 10:28 pm
> On the 51st pass in that loop, the untruncated, unrounded number was indeed 0.0050.
So, what did you mean when you wrote this:
"Why in the world would you consider a...
June 10, 2007 at 8:16 pm
> Your test was a series of numbers between 0.0000 and 1.9999, inclusive, with no 2.0 to be found.
2.0 indicates start of the next interval, from 2.0 to 4.0.
2.0...
June 10, 2007 at 7:53 pm
You ran imprecise numbers.
Your result is imprecise as well because of that.
There was no 0.0050, it was 3 digits precision measurement of 0.00591232349283423847234... : 0.005.
It's a matter of fact:
Average of...
June 10, 2007 at 6:44 pm
David,
What's an AVERAGE for all numbers evenly distributed between 0 and 2?
Do the MATH, not ARITH.
The average is 1. Or 0.(9) if you wish.
Difference SUM(Values)/COUNT(Values) and "1" shows level of...
June 10, 2007 at 2:12 pm
Here is another script.
Hope it satisfies all your criteria:
DECLARE @A DECIMAL(8, 4), @b-2 DECIMAL(8, 3), @Step DECIMAL(8, 4)
SET NOCOUNT ON
SELECT @A = 0.0000, @Step = .0001
IF Object_Id ('tempdb..#BRounding') IS...
June 10, 2007 at 6:40 am
David, are you drunk?
BR returns the same result when precision allows to see the next number.
So, BR returns different results depending on precision of source value is measured.
Don't know which...
June 10, 2007 at 5:35 am
As I see you have nothing to say anymore for defence of BR.
June 9, 2007 at 10:06 pm
"If on the cell with an elephant you see the label "Hyppo" - don't believe your eyes".
The name of this author will not say you anything.
June 9, 2007 at 5:29 pm
Viewing 15 posts - 4,096 through 4,110 (of 6,036 total)