[Insight-developers] Random number reseeding

[Brad King]

Lorensen, William E (Research) wrote:
> On cygwin, why not seed the first time with a specific seed? I think the
> other compilers are OK. The regression tests do a reseed with a specific
> value to guarantee reproducable results.

This problem was already fixed in the VXL upstream source:

http://cvs.sourceforge.net/viewcvs.py/vxl/vxl/core/vnl/algo/tests/test_cholesky.cxx?rev=1.1&view=auto

They use this code for the random initialization:

   vnl_random rng(1000);
   vnl_matrix<double> A(3,3);
   test_util_fill_random(A.begin(), A.end(), rng);

Clearly this is a fixed seed.

-Brad