How to improve double precision?
显示 更早的评论
I have made below code
a=11;
b=256;
c=a/b
and the output said c=0, which was not I had expected
So I have used cast operator
a=11;
b=256;
c=double(a/b)
and the result returns the same value; 0
I have searched Matlab document and found that the default precision of double data type is 32
so It should return 0.04296875 instead of 0. But still get 0 for this kind of codes
How to fix it properly?
采纳的回答
Walter Roberson
2020-7-23
1 个投票
Either your a or your b is type uint8. When you work with int8, int16, int32, uint8, uint16, or uint32, then the operands are converted to double(), and the operation is done, and then the result is converted back to the most restricted type used.
Operations for int64 and uint64 are handled differently in some, because 64 bit integers cannot be converted to double without losing precision.
1 个评论
Could be any of the int classes, of course...what the
whos a b c
would have told us is which one(s) OP had...what OP didn't show was how a, b were created to begin with to become int class
更多回答（0 个）
类别
在 帮助中心 和 File Exchange 中查找有关 Data Type Conversion 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
