AMBER Archive (2008)

Subject: Re: AMBER: [Fwd: segmentation fault in nmode]

From: David A. Case (case_at_scripps.edu)
Date: Thu Jun 19 2008 - 19:22:34 CDT

On Tue, Jun 17, 2008, Guillaume Renvez wrote:

> I also send the input files I used, if someone else wants to try...
> thanks!

The code is clearly wrong, in its initialization of the scr2() array (in
forces.f).

Try the following patch: (just commenting out three lines:)

diff -c -r10.0 forces.f
*** forces.f 15 Apr 2008 23:23:18 -0000 10.0
--- forces.f 19 Jun 2008 23:07:31 -0000
***************
*** 133,141 ****
              iptatm,scr7,scr8)
  
        lim = 3*natom*(3*natom-1)/2
! do jn=1,lim
! scr2(jn) = 0.d0
! end do
        
        ! subroutine pol2der(natom,x,f,p,q,h,ipair,
        ! - jpair,xrc,xij,r2,fw,vt,
--- 133,141 ----
              iptatm,scr7,scr8)
  
        lim = 3*natom*(3*natom-1)/2
! !do jn=1,lim
! ! scr2(jn) = 0.d0
! !end do
        
        ! subroutine pol2der(natom,x,f,p,q,h,ipair,
        ! - jpair,xrc,xij,r2,fw,vt,

For me, this removes the seg. fault, but I don't know if the results are
otherwise correct.

Note that your structure is not at a minimum (as least as nmode sees it), so
you will need to make sure you get a minimum in nmode (with the same energies
as in sander), before trusting the modes.

...hope this helps...dac

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu