This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(35.2, 84%, 13%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EHP%3C/text%3E%3C/svg%3E)
So I'm using double datatype for most of my calculations and when I compare my results with the calculations from an online calculator sometimes there are differences in the decimal part of the number. Also even during calculations 3 * 3.5 = 9.600000000000001. Is there anyway I can fix this mistakes without using decimal as my datatype?
@Hrishikesh P, Welcome to Microsoft Q&A, could you provide the related code about your precision problems?
I used the following code to make a test and did not get your result:
double d1 = 3;
double d2 = 3.5;
var result = d1 * d2;
"Double" is an un-precise float-point data type and and calculation results can have more decimals as expected.
Also even during calculations 3 * 3.5 = 9.600000000000001.
3 * 3,5 = 10.5; if you get 9.6, then you are doing something completley wrong.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(211.20000000000002, 61%, 30%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EBS%3C/text%3E%3C/svg%3E)
the issue is that double is floating point number, but in base 2 instead of base 10. you probably know in base 10, 1/3 = .33333..... in base 2 there are other repeating numbers. 1/10 in base 2 is a repeating number. which make floating bad for money calculations.
the decimal datatype gets around this but storing a numeric value and an implied decimal point. The math operations are done as integer match, than rounded to the scaling factor
the usual fix if using floating point is to first multiple by the scaling factor (number of decimal places you want), do the calc and then divide by the scale factor.