LabWindows/CVI Idea Exchange

cancel
Showing results for 
Search instead for 
Did you mean: 
tstanley

Implement the 'long double' type with 80 bits of precision.

Status: New

Currently the 'long double' type has 64 bits of precision, which is the same as the 'double' type.  It would be handy on occasion to be able to use the full 80 bits of precision like is available with the Intel compiler.

2 Comments
Wolfgang
Trusted Enthusiast

In addition to that, functions generating long double results are needed, e.g. expl, sqrtl...

Wolfgang
Trusted Enthusiast

Following the comments of Bob here we should ask for a long double with 128 bits of precision...

 

Btw, I don't care so much about its implementation in hardware - if I need extended precision I am willing to wait longer for the result...Smiley Wink