32 bit calculation

[bralanth]

In a project [sort of calculator], I need 9 digits and floating point calculation; I know that ccs c supports 32 bit floating points (IEEE) which provides 24 bits for mantissa. that makes a 8-digits number. but I need at least 30-bit mantissa for 9 digits.

Is there a way for me to calculate both 32 bit floating point and store 9 digit numbers as a whole? can anybody help?

[Ttelmah]

Not without writing your own maths library.
However remember that you could (for instance), use a 32bit integer, and another int8, as a scale factor. Depending what arithmetic you want, this could be relatively easy to code.

Best Wishes