Strange fprintf output

[danep2]

Hello all

This is a really strange problem. I have code that performs a few
calculations based on input from a joystick, and writes these values
to a file using basically the following code:

fprintf(fpResul ts, "i: %4.2f, j: %4.2f", i, j );

This works correctly (literally) 99.9999% of the time. However, once
every million or so writes I get the following output:

i: -1.#J, j: 1.#R

Does anybody know what this means?

[James Kanze]

On Apr 20, 10:11 pm, danep2 <danepow...@gma il.comwrote:

This is a really strange problem. I have code that performs a few
calculations based on input from a joystick, and writes these values
to a file using basically the following code:

fprintf(fpResul ts, "i: %4.2f, j: %4.2f", i, j );

This works correctly (literally) 99.9999% of the time. However, once
every million or so writes I get the following output:

i: -1.#J, j: 1.#R

Does anybody know what this means?

Perhaps your compiler vendor? This can't happen in a conforming
implementation unless your code has undefined behavior. Perhaps
the vendor could indicate what sort of undefined behavior might
cause it. (It will probably depend on how floating point values
are represented.)

--
James Kanze (GABI Software) email:ja******* **@gmail.com
Conseils en informatique orientée objet/
Beratung in objektorientier ter Datenverarbeitu ng
9 place Sémard, 78210 St.-Cyr-l'École, France, [phone]

[danep2]

On Apr 21, 4:18*am, James Kanze <james.ka...@gm ail.comwrote:

On Apr 20, 10:11 pm, danep2 <danepow...@gma il.comwrote:

This is a really strange problem. *I have code that performs a few
calculations based on input from a joystick, and writes these values
to a file using basically the following code:
fprintf(fpResul ts, "i: %4.2f, j: %4.2f", i, j );
This works correctly (literally) 99.9999% of the time. However, once
every million or so writes I get the following output:
i: -1.#J, j: 1.#R
Does anybody know what this means?

Perhaps your compiler vendor? *This can't happen in a conforming
implementation unless your code has undefined behavior. *Perhaps
the vendor could indicate what sort of undefined behavior might
cause it. *(It will probably depend on how floating point values
are represented.)

--
James Kanze (GABI Software) * * * * * * email:james.ka. [e-mail]
Conseils en informatique orientée objet/
* * * * * * * * * *Beratung in objektorientier ter Datenverarbeitu ng
9 place Sémard, 78210 St.-Cyr-l'École, France, [phone]

I am developing using Microsoft Visual Studio 2005. Prior to
migrating to VS 2005, we also experienced this problem with VC++ 6.0.
For what it's worth, this *usually* happen when the values of i or j
are very close to zero. (I don't know for a fact what they are, I am
simply inferring from the data immediately surrounding this output.)
Also, usually both variables are affected - i.e., if the output for i
is corrupt, the output for j will also be corrupt. However, other
floats being written in the same statement are never corrupted.