Thanks for the explanation!
On 09/12/2012 07:39 PM, Ray Luo, Ph.D. wrote:
> Jan-Philip,
>
> If you want absolute control of the grid dimension in the final
> fine-grid calculation, you have to set nfocus to 1, i.e. no focusing
> should be used. Since fillratio only applies to the dimension of the
> first grid, i.e. the coarse grid, when nfocus = 2.
>
> Ray
>
> On Wed, Sep 12, 2012 at 10:32 AM, Jan-Philip Gehrcke
> <jgehrcke.googlemail.com> wrote:
>> Ray,
>>
>> Actually I missed your answer below. Excuse me. I will look into how
>> nfocus changes the behavior I observed. Do you expect fillratio to have
>> no effect with nfocus=2?
>>
>> Thank you!
>>
>> Jan-Philip
>>
>>
>>
>> On 08/22/2012 04:25 PM, Ray Luo wrote:
>>> I think fill ratio is the only thing you need to worry about. Of course you don't want to turn on electrostatic focusing, ie nfocus should be set to 1.
>>>
>>> Ray Luo, Ph.D.
>>>
>>> On Aug 20, 2012, at 6:24 AM, Jan-Philip Gehrcke <jgehrcke.googlemail.com> wrote:
>>>
>>>> Let me try to put my question in simpler words:
>>>>
>>>> I use pbsa for calculating the electrostatic potential of proteins. How
>>>> can I enlarge the sampling volume (the absolute grid dimensions) for
>>>> this type of calculation? I tried `fillratio`, but failed.
>>>>
>>>> Any suggestions would be helpful.
>>>>
>>>> Thanks,
>>>>
>>>> Jan-Philip
>>>>
>>>>
>>>>
>>>> On 07/24/2012 07:13 PM, Jan-Philip Gehrcke wrote:
>>>>> Hello,
>>>>>
>>>>> I'd like to increase the grid dimensions in a PBSA run while keeping the
>>>>> spacing constant (just to see more of the distribution of the potential
>>>>> in space). I figured that this should be possible through the
>>>>> `fillratio` parameter. I did not find any other relevant parameter while
>>>>> browsing the documentation.
>>>>>
>>>>> Hence, I increased fillratio -- however, I did not observe a change in
>>>>> the grid dimensions in the DX output file. For my system, I get 55x69x63
>>>>> gridpoints in the DX file independent of the fillratio setting.
>>>>>
>>>>> So, my question is: how do I increase the grid dimensions in a way that
>>>>> a *larger* volume is sampled and that the data of the entire sampled
>>>>> volume is actually written to the volumetric data file?
>>>>>
>>>>> What follows now are some more details. I have played around a bit with
>>>>> the fillratio setting and used npbverb=1. What I've seen in the pbsa
>>>>> output file was the following:
>>>>>
>>>>> - "Grid dimension at level 2 55 69 63" did not change in
>>>>> dependence of fillratio. The data of this grid seems to end up in the DX
>>>>> file.
>>>>>
>>>>> - "Grid dimension at level 1 11 15 13" changes in dependence
>>>>> of fillratio (1.0 in this case).
>>>>>
>>>>> What are the two grid levels good for? Can you help me out here? The
>>>>> code in subroutine `pb_setgrd` is way too unstructured for me to follow
>>>>> its logic within a reasonable amount of time.
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Jan-Philip
>>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> AMBER mailing list
>>>> AMBER.ambermd.org
>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Sep 12 2012 - 11:00:06 PDT
