[gmx-developers] Performance checklist

[Berk Hess]

On 10/12/2012 04:05 PM, Shirts, Michael (mrs5pt) wrote:
>> But in this context this is not really an issue, because the virtual
>> sites remove
>> the fastest modes and thus decreases the acceleration, making this error
>> about equally large at 4/5 fs as without vsites at 2 fs.
> This is a reasonable theory, that is likely right.  It should be tested
> though, rather than assumed.
>   
I have looked at potential energies and rdf some years ago, I don't recall
the exact numbers, but there were not severe deviations.
For water I checked it more recently and there deviation is very small 
at 4 fs
and "small" at 5 fs.

Cheers,

Berk
>> This would also matter with md-vv, as that has a worse Ekin estimator.
> Agreed.  Quantifying this amount would be useful.
>
> Best,
> ~~~~~~~~~~~~
> Michael Shirts
> Assistant Professor
> Department of Chemical Engineering
> University of Virginia
> michael.shirts at virginia.edu
> (434)-243-1821
>
>
>
>> In itself this is a good point.
>> But in this context this is not really an issue, because the virtual
>> sites remove
>> the fastest modes and thus decreases the acceleration, making this error
>> about equally large at 4/5 fs as without vsites at 2 fs.
>>
>> This would also matter with md-vv, as that has a worse Ekin estimator.
>>
>> Cheers,
>>
>> Berk
>>
>>> Best,
>>> ~~~~~~~~~~~~
>>> Michael Shirts
>>> Assistant Professor
>>> Department of Chemical Engineering
>>> University of Virginia
>>> michael.shirts at virginia.edu
>>> (434)-243-1821
>>>
>>>
>>>> From: Berk Hess <hess at kth.se>
>>>> Reply-To: Discussion list for GROMACS development
>>>> <gmx-developers at gromacs.org>
>>>> Date: Fri, 12 Oct 2012 15:21:54 +0200
>>>> To: Discussion list for GROMACS development <gmx-developers at gromacs.org>
>>>> Subject: [gmx-developers] Performance checklist
>>>>
>>>> Hi,
>>>>
>>>> I added a performance check-list on the Gromacs site:
>>>> http://www.gromacs.org/Documentation/Performance_checklist
>>>>
>>>> If something is unclear or you have something to add, pease tell me.
>>>> Note that about half of the points on the list only apply to version 4.6.
>>>>
>>>> Cheers,
>>>>
>>>> Berk
>>>>
>>>> -- 
>>>> gmx-developers mailing list
>>>> gmx-developers at gromacs.org
>>>> http://lists.gromacs.org/mailman/listinfo/gmx-developers
>>>> Please don't post (un)subscribe requests to the list. Use the
>>>> www interface or send it to gmx-developers-request at gromacs.org.