AMBER Archive (2008)

Subject: Re: AMBER: SHIFTS compile error

From: David A. Case (case_at_scripps.edu)
Date: Wed Jul 23 2008 - 12:22:32 CDT

On Tue, Jul 22, 2008, IN SUK JOUNG wrote:

> While I compiled SHIFTS-4.2, I got this error message:
>
> nab -o lsq lsq.nab
> lsq.c: In function `lsqfunc':
> lsq.c:63: error: too many arguments to function `rand2'
> cc failed!
> make: *** [lsq] Error 1
>

The way nab handles random numbers has changed, and shifts needs an update.

You can either skip the compiliation of lsq (I'm not sure that it is needed),
or apply the following patch to lsq.nab:

-----------------------------------------------------------------------------
--- lsq.nab 14 Sep 2001 17:10:35 -0000 1.8
+++ lsq.nab 23 Jul 2008 17:22:53 -0000
@@ -37,12 +37,12 @@
 // can be put here: just compute the function value in "f" and its
 // gradient (with respect to x) in "g".
 
- iseed = -(9975631 + 7511*ijack);
+ iseed = rseed();
     f = 0.; used = 0;
     for (ivar = 1; ivar <= n; ++ivar) g[ivar] = 0.;
 
     for (iobs = 1; iobs <= m; iobs++) {
- if( ijack == 0 || MAXJACK*rand2( iseed ) > 1.0 ) {
+ if( ijack == 0 || MAXJACK*rand2() > 1.0 ) {
                        calc = 0.;
                        for (ivar = 1; ivar <= n; ivar++) {
                        calc += a[iobs, ivar] * x[ivar];
-----------------------------------------------------------------------------

Thanks for the report...regards....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