AMBER Archive (2003)

Subject: Re: Re: can you give me more advice about: vlimit exceeded when sander calculate free energies using TI

From: sxzheng (sxzheng_at_mail.shcnc.ac.cn)
Date: Sun Jun 29 2003 - 20:05:55 CDT

Dear David A. Case£¬
I have tried the Sander7_TI with ntr=1 (500kcal/mol ) in nonperiod condition, it can run normally, but I found two queston:
1. ntr=1 will decrease the computational velocity
2. the Etot will be positive for the restrained energy will be added. Thus the results seem not reasonable.
This questions have not been found in period condition simmulations.

So can the routines in set.f are making modifications for the regular Hamiltonian, and also for the perturbed Hamiltonian.

The same phenomena has also displayed in Gibbs7, that ibelly=1 and nonperiod conditions with watercap will not go normally.

Thank you
sxzheng

ÔÚ 03-6-16 8:12:00 ÄúдµÀ£º
>On Mon, Jun 16, 2003, sxzheng wrote:
>
>> I have test the system: first, just take the ligand in waterbox it can run
>> normally even with TI method second, do the complex's dynamics in capwater
>> condition with ibelly=0, it can run normally also with TI method third, do
>> the complex's dynamics in capwater condition with ibelly=1, it can run
>> normally with no Ti method, but vlimit exceeded with TI method from the
>> first step in clambda=0.0.
>
>It looks like the routines in set.f are making modifications for the regular
>Hamiltonian, but not for the perturbed Hamiltonian, and this is leading to
>memory corruption.
>
>The simple advice is: don't use ibelly in this way; (use ntr=1 instead). You
>could try to debug it by not calling the "set" routines and seeing if that
>helps.
>
>I am still leaning toward removing the ibelly option altogether in future
>releases, at least until it can be implemented without so many side effects.
>
>...thanks for your report....dac
>
>--
>
>==================================================================
>David A. Case | e-mail: case_at_scripps.edu
>Dept. of Molecular Biology, TPC15 | fax: +1-858-784-8896
>The Scripps Research Institute | phone: +1-858-784-9768
>10550 N. Torrey Pines Rd. | home page:
>La Jolla CA 92037 USA | http://www.scripps.edu/case
>==================================================================

                    ÖÂ
Àñ£¡

            sxzheng
            sxzheng_at_mail.shcnc.ac.cn