[gmx-users] rvdw vs cut-off / different types of non bonded interactions

[David van der Spoel]

On Wed, 12 Dec 2007, Claus Valka wrote:

> Dear Sir,
>
> I would try to be more precise.
>
> example1 mdp file:
> rlist = (greater or equal to rvdw)
> vdw-type = Switch
> rvdw-switch=5.56
> rvdw=8.9
>
are you aware that gromacs usues nm for distance?

you can find the functional form of the shift/switch function in the 
manual.

> Between the value 5.56 and 8.9 gromacs uses cubic
> spline?
>
> example2 mdp file:
>
> rlist=5.56
> vdw-type = Cut-off
> rvdw-switch=0 (this value here should be irrelevant)
> rvdw=8.9
>
> Gromacs uses cubic spline after the value 8.9 till LJ
> to go to zero?
>
> And which is the difference of rlist if someone uses
> vdw-type = Cut-off instead of vdw-type = Switch? It
> would have been nice to have images of these options
> so as to be better understood.
> Because it is stated in the manual:
> rlist: (1) [nm]
> cut-off distance for the short-range neighbor list
> and also:
> Switch
> ...The neighbor search cut-off rlist should be 0.1 to
> 0.3 nm LARGER than rvdw...
> Isn't this a little bit contradictive?
>
> As far as twin cutoffs are concerned if we want to
> find out the total energy of the non bonded
> interactions we have to add LJ-14 + LJ(SR) + LJ(LR)?
>
> With tabulated potentials can I use different cutoffs
> for different types of atoms? Judging from your answer
> about system wide cutoffs I understand that even
> though I might use tabulated potentials, I won't be
> able to use different cutoffs for the different types
> of atoms I have. Am I correct?
>
> Yours Sincerely,
> Nikos
>
>
> --- Mark Abraham <Mark.Abraham at anu.edu.au> schrieb:
>
>> Claus Valka wrote:
>>> Hello,
>>>
>>
>>> 3rd question: Which is the difference between the
>>> S(hort)R(ange) LJ and LJ 1-4? For me the short
>> range
>>> interactions are the 1-4LJ, yet it seems that this
>> is
>>> not the case.
>>
>> GROMACS uses twin-range cutoffs (see manual), so
>> LJ-SR is inside the
>> shorter cutoff, and LJ-LR is outside.
>>
>>> 4th question: I want after 1,45*&#963; till
>>> 2,33*&#963; to use the cubic spline in such a way
>> that
>>> the LJ potential will go smoothly to zero at
>>> 2,33&#963;. If I use the biggest &#963; I have,
>> then
>>> in the mdp file rvdw-switch=5.56 and rvdw=8.9.
>> Between
>>> these two values does the cubic spline work as a
>>> switch function using vdw-type = Switch ?
>>
>> Please find a way to describe your problem with
>> plain ASCII text.
>>
>>> 5th question: Given the above if I use vdw-type
>>> =Cut-off and rvdw=8.9 then gromacs will use cubic
>>> spline till the LJ will reach zero for a value
>> greater
>>> than 8.9, yet to me uknown? If this is not the
>> case
>>> how LJ goes to zero with this option?
>>
>> I don't understand what you're asking.
>>
>>> 6th question: If I want to use a quintic spline
>>> between 1,45*&#963; till 2,33*&#963;(cut-off
>> value)
>>> for the different combination of atoms(they have
>>> different &#949; and &#963;), then should I use a
>>> tabulated potential? And if I do that which will
>> be
>>> the meaning of cutoffs in the mdp file as I would
>> like
>>> them to be different for each of the different
>> types
>>> of atoms I have? Can I use different cut-offs for
>> the
>>> different types of atoms I have?
>>
>> The cut-offs are a system-wide property.
>>
>> Mark
>> _______________________________________________
>> gmx-users mailing list    gmx-users at gromacs.org
>> http://www.gromacs.org/mailman/listinfo/gmx-users
>> Please search the archive at
>> http://www.gromacs.org/search before posting!
>> Please don't post (un)subscribe requests to the
>> list. Use the
>> www interface or send it to
>> gmx-users-request at gromacs.org.
>> Can't post? Read
>> http://www.gromacs.org/mailing_lists/users.php
>>
>
>
>
>      Jetzt Mails schnell in einem Vorschaufenster überfliegen. Dies und viel mehr bietet das neue Yahoo! Mail - www.yahoo.de/mail
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>