Thread: spontaneous round-off type error in loop

Hello, looking for some help with some peculiar behavior. Check out the code and output below. BTW, I am on a Win2k box using MinGW to compile.

Code:
=======
.
.
.
float i=0.0,j=0.0;
.
.
.

for (j=0.0; j<=0.01;j=j+.001 )
{
// fprintf(FID, "%6.3f", j);
fprintf(stderr, "\n%.14f",j);
}
.
.
.

Output
=====

0.00000000000000
0.00100000004750
0.00200000009499
0.00300000002608
0.00400000018999
0.00500000035390
0.00600000051782
0.00700000068173
0.00800000037998
0.00900000054389


Where is the .0004750 error coming from on the 1st time through the loop? The problem is that this is kicking me out of my for loop early.

Thanks in advance,
Jesse

Hello, looking for some help with some peculiar behavior. Check out the code and output below. BTW, I am on a Win2k box using MinGW to compile.

Code:
=======

Code:

.
.
.
float i=0.0,j=0.0;
.
.
.

for (j=0.0; j<=0.01;j=j+.001 )
{
     // fprintf(FID, "%6.3f", j);
     fprintf(stderr, "\n%.14f",j);
}
.
.
.

Output
=====
[code]
0.00000000000000
0.00100000004750
0.00200000009499
0.00300000002608
0.00400000018999
0.00500000035390
0.00600000051782
0.00700000068173
0.00800000037998
0.00900000054389
[/code/]

Where is the .0004750 error coming from on the 1st time through the loop? The problem is that this is kicking me out of my for loop early.

Thanks in advance,
Jesse

What every computer scientist should know about floating-point arithmetic

howdy,

thanks for the quick replies.

I downloaded that pdf, and will take a look at it later.

I usually use doubles, and am using them now on this program per your advice.

thanks again!

later,
Jesse