On Wed, Feb 20, 2008, Z. Nevin Gerek wrote:
> Thanks for your suggestions. I figured out the problem and wanted to
> share it.
> I tried with small protein and I found that you need to define nsnb
> in mm_options.
> When I wrote
> mm_options( "cut=999.0, ntpr=50, gb=1, nsnb=99999" );
> It worked without any problem.
This is quite odd, since nsnb defaults to a value of 10, which is not the best
value, but should not cause behavior like this:
from conjgrad:
> >>ff: 0 -8576.17 7735.94 -1750.28 -10638.24 64.21 -3987.79
from newton:
> >>ff: 1 25349557291.95 25349573605.27 -1751.25 -10638.73 64.21 -3987.55
It sounds like there may be a problem with the list-builder or something, but
I don't see any problems when I try including nsnb vs. leaving it out in my
tests.
Can you post the input pdb file and the nab program you used? It sounds like
there is (still) a bug lurking somewhere, but we need to be able to have it
fail here before we can find it.
Even if it is "operator error", we should figure this out so that the examples
and documentation can be improved.
...thanks...dac
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Sun Feb 24 2008 - 06:07:11 PST
