Hi!

I'm pretty new to using NR and TNT.

Questions:
(1) I want to be able to use TNT, yet also use the functions in NR. Should I use TNT declarations, then convert to NR when I need to use the NR functions? I tried passing TNT vectors and matrices directly to NR functions (and vice versa) and it chokes.

(2) Also, what is an easy way to convert regular arrays (1 and 2D) into TNT Vectors and Arrays, and also TNT Vectors and Arrays (assume type double) to Vec_DP and Mat_DP classes.

(3) How do I resize a Vec_DP or Mat_DP object? How do I dynamically allocate Vec_DP or Mat_DP object? Is there a way to call a destructor explicitly (so that I can resize an object)?

sorry about so man questions. I am starting a project and I would like to use this library -- so it's sink or swim time!

Best,

AO

In my opinion, there's no general way sticking with one vector or matrix implementation by far.

STL should publish a matrix class, and then everybody stick with it.

The other problem is that fortran ppl uses different indexing for matrix. And lots of good c++ library translate them. Therefore it's just nightmare for c++ ppl.

IMHO, NR are recieps that you want to understand and then copy to your own code instead of just using them directly. (This only applys to C++ version ) Hope i don't get punched by saying this.

I know i didn't answer your question.

Fang