Pawn Float bug ?

NexoR · 2020-04-19, 01:33 PM

Hello, I am working on height calculation. But some places seem incomplete and inaccurate. Can you help with this issue?

My code:

Code:
for (new i = 1; i <= 35; i++)

{

new Float: X = 1.50 + float(i) * 0.01;

printf("%.2f", X);

}

Output:

Code:
50

51

52

53

54

55

-> There is a deficiency here.

57

58

59

60

61

62

-> There is an error here.

62

63

64

65

66

67

-> There is a deficiency here.

69

70

71

72

73

74

75

-> There is an error here.

75

76

77

78

79

80

81

82

-> There is a deficiency here.

84

85

**Markski** · (This post was last modified: 2020-04-19, 05:53 PM by Markski.)

This is a floating point accuracy issue. It can be better seen when looking at the numbers on a larger scale.

A solution to this problem could be to just define i as a float from the let's go, and give it a value slightly higher than 1.0 to avoid floating point inaccuracy setting it below 1.0

Code:
for (new Float:i = 1.0001; i <= 35.0; i++) {

   new Float: X = 1.50 + i * 0.01;

   printf("%.2f", X);

}

This code prints as follows: