[pgw3]

Having decided to learn myself Cocoa I wrote a simple calculator in Java and everything worked fine except for whenever I tried to use decimal numbers. The result is almost right but not exactly, e.g. 1.1 + 1 would become 2.099999... Oh, well I thought, must be something wrong with my Java-knowledge. So I tried to write a simple program in Objective-C that just adds two numbers and to my surprise the same thing happens, i.e. 1 +1 becomes 2 but 1.1 + 1 becomes 2.099999904632568. I include my source code below and I hope someone can shed some light on it as I really haven't got a clue what could be wrong...

---test.h

#import <Cocoa/Cocoa.h>

@interface test : NSObject
{
IBOutlet id number1;
IBOutlet id number2;
IBOutlet id sum;
}
- (IBAction)calculate id)sender;
@end

---test.m

#import "test.h"

@implementation test

- (IBAction)calculate id)sender
{
float value1, value2, calculation;
value1 = [number1 floatValue];
value2 = [number2 floatValue];
calculation = value1 + value2;
[sum setFloatValue:calculation];
}

[parallax]

Try using 'double' instead of 'float'

[pgw3]

Originally posted by parallax:
Try using 'double' instead of 'float'

Thanks for the suggestion, it worked. But why is it like that? Can't one trust float and is there ever a reason to use float instead of double? I thought that one could use float if one only needs a few decimals.

[davechen]

That's just the nature of representing a decimal fraction with binary
numbers. Using double just pushes the error out to fewer decimal places.
It's like trying to represent 1/3 in decimal. At some point you have to
cut it off.

The standard thing to do is decide how much precision you want and then
round off your numbers.