Precision of zgetrf

[FrankM]

Hi,

I use the zgetrf-routine for calculating the determinant of a complex matrix. Unfortunately, I think the round-off errors of the routine are in the order of 1e-15 and I need to calculate the determinant with a higher precision. So my question is, whether there is a way to increase the accuracy of the routine? In fact, I use them in C and I want to use "long double" instead of "double" precision.

Thanks in advance,
Frank