Re: [ESPResSo-users] Can't get virtual sites to work

[Tristan Bereau]

Hi Olaf,

Thanks for your message. No, I didn't unfortunately. I tried digging
into the virtual.c file, but weird things are going on. I seem to
follow the protocol written in the user manual, but the center of mass
routine doesn't call the right particles, and there seems to be memory
allocation problems. I'll try digging into it a bit more when I have
some more time (I can never dismiss a mistake on my side, but I can't
find what I'm doing wrong). Unfortunately, that doesn't seem to be a
feature that's heavily used by the community. The other
VIRTUAL_SITES_RELATIVE does have a testsuite, but that doesn't help me
too much.

Thanks again for getting back to me.

Best,
Tristan

On Fri, Mar 1, 2013 at 9:23 AM, Olaf Lenz <address@hidden> wrote:
> Hi Tristan!
>
> Did you get a reply to this mailing off-list?
>
> Olaf
>
> On 02/08/2013 10:20 AM, Tristan Bereau wrote:
>> Dear all,
>>
>> This must be a stupid mistake on my part, but I can't seem to spot it.
>> I was trying to setup an example script to work with virtual sites
>> (defined with respect to the center of mass of the molecule,
>> VIRTUAL_SITES_COM). I set up two molecules of two particles each and
>> define their virtual sites by the following script:
>>
>> part 0 pos  0.0  0.0 0.0 mol 1
>> part 1 pos  1.0  1.0 0.0 mol 1
>> part 2 pos  9.0  9.0 9.0 mol 1 virtual 1
>> part 3 pos  5.5  0.0 2.0 mol 2
>> part 4 pos  4.5  1.0 2.0 mol 2
>> part 5 pos  8.0  8.0 8.0 mol 2 virtual 1
>> analyze set 1 "0 1 2" 2 "3 4 5"
>> puts "call 'analyze set': [analyze set]"
>>
>> foreach i [part] { puts $i }
>> puts ""
>> setmd time_step 0.001; thermostat off; setmd skin 1.0
>>
>> puts "call 'integrate 1'"
>> integrate 1
>>
>> foreach i [part] { puts $i }
>>
>> puts "\ncode_info: [code_info]"
>>
>> *************
>>
>> The output is below. As you can see, the call to "integrate 1" does
>> not properly place virtual site 1 (it stays at its initial position)
>> and virtual site 2 becomes NaN. Could anyone tell me what I'm doing
>> wrong?
>>
>> Thanks,
>> Tristan
>>
>>
>>
>> call 'analyze set': { 1 } { 0 1 2 } { 2 } { 3 4 5 }
>> 0 pos 0.0 0.0 0.0 type 0 molecule 1 mass 1.0 v -0.0 -0.0 -0.0 f 0.0
>> 0.0 0.0 virtual 0
>> 1 pos 1.0 1.0 0.0 type 0 molecule 1 mass 1.0 v -0.0 -0.0 -0.0 f 0.0
>> 0.0 0.0 virtual 0
>> 2 pos 9.0 9.0 9.0 type 0 molecule 1 mass 1.0 v -0.0 -0.0 -0.0 f 0.0
>> 0.0 0.0 virtual 1
>> 3 pos 5.5 0.0 2.0 type 0 molecule 2 mass 1.0 v -0.0 -0.0 -0.0 f 0.0
>> 0.0 0.0 virtual 0
>> 4 pos 4.5 1.0 2.0 type 0 molecule 2 mass 1.0 v -0.0 -0.0 -0.0 f 0.0
>> 0.0 0.0 virtual 0
>> 5 pos 8.0 8.0 8.0 type 0 molecule 2 mass 1.0 v -0.0 -0.0 -0.0 f 0.0
>> 0.0 0.0 virtual 1
>>
>> call 'integrate 1'
>> 0 pos 0.0 0.0 0.0 type 0 molecule 1 mass 1.0 v 0.0 0.0 0.0 f 0.0 0.0
>> 0.0 virtual 0
>> 1 pos 1.0 1.0 0.0 type 0 molecule 1 mass 1.0 v 0.0 0.0 0.0 f 0.0 0.0
>> 0.0 virtual 0
>> 2 pos 9.0 9.0 9.0 type 0 molecule 1 mass 1.0 v 0.0 0.0 0.0 f 0.0 0.0
>> 0.0 virtual 1
>> 3 pos 5.5 0.0 2.0 type 0 molecule 2 mass 1.0 v 0.0 0.0 0.0 f 0.0 0.0
>> 0.0 virtual 0
>> 4 pos 4.5 1.0 2.0 type 0 molecule 2 mass 1.0 v 0.0 0.0 0.0 f 0.0 0.0
>> 0.0 virtual 0
>> 5 pos -nan -nan -nan type 0 molecule 2 mass 1.0 v -nan -nan -nan f 0.0
>> 0.0 0.0 virtual 1
>>
>> code_info: ESPResSo-3.1.1
>> { Compilation status { FFTW } { CONSTRAINTS } { BOND_ANGLE_HARMONIC }
>> { COMFIXED } { NPT } { COMFORCE } { PARTIAL_PERIODIC } { LENNARD_JONES
>> } { LJ_ANGLE } { OVERLAPPED } { MODES } { EXTERNAL_FORCES } {
>> TABULATED } { METADYNAMICS } { LJCOS } { LENNARD_JONES_GENERIC } {
>> MASS } { MOLFORCES } { LJCOS2 } { DPD } { VIRTUAL_SITES_COM } {
>> BOND_VIRTUAL } { LJ_WARN_WHEN_CLOSE } { EXCLUSIONS } }
>>
>
>
> --
> Dr. rer. nat. Olaf Lenz
> Institut für Computerphysik, Allmandring 3, D-70569 Stuttgart
> Phone: +49-711-685-63607