[gmx-users] NMA problem (again and again!)

[fabrizio.mancinelli at unina2.it]

Hello all,
I've got 2 more question about normal modes analysis.

1) at the end of the minimization step, I got a value of about e-04 for Fmax. 
But, when the mdrun(nm) step starts, it says that Fmax is about e+02. What 
could have been happened?

2) By the way, the hessian was built as well. During diagonalization step, the 
program halted with the following message:

-------------------------------------------------------
Program g_nmeig_d, VERSION 3.3
Source code file: eigensolver.c, line: 223

Fatal error:
Unspecified error from Arnoldi diagonalization:3
-------------------------------------------------------

The protein was a oligomer of 438 aa in vacuum. The electrostatic treatment 
was a shifted potential as suggested by David in an earlier post.
Thanks, 
Fabrizio