>>>>> Alex M writes:
Alex> That is actually documented I believe. It's not limited to
Alex> jdk1.2. I don't know why it can't be just 0.3000000000000
Alex> but it is. This is why the language spec says not to
Alex> compare floats and doubles without casting, you will always
Alex> get negative results even though you think you should have a
Alex> positive one.
Alex> So to compare 0.3d with 0.3f you would do:
Alex> double d = 0.3;
Alex> float f = 0.3f;
Alex> if ((float)d == f) { do something };
Don't use '==' to compare floating point values, instead compare
their difference against a small value:
double epsilon = 0.0001;
double a, b;
...
if (a - b < epsilon) ...
Alex> You will lose precicion on your double variable though, so
Alex> make sure you don't need it before you do it.
Alex> Back to why you can't store 0.3 as a double. I'm guess this
Alex> has more to do with the limitations of binary representation
Alex> of floating point numbers. It's been a long time since my
Alex> intro to CS courses when I learned all this, anybody have
Alex> anymore info on this?
See http://www.validgh.com/goldberg/paper.ps (What Every Computer
Scientist Should Know About Floating-Point Arithmetic')
Juergen
--
Juergen Kreileder, Blackdown Java-Linux Porting Team
http://www.blackdown.org/java-linux.html
----------------------------------------------------------------------
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
