VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Tue May 20 2014 - 10:12:50 CDT

Hi,
  Use the CPU-only version when running the NAMD energy plugin for now.

Cheers,
  John Stone
  vmd_at_ks.uiuc.edu

On Tue, May 20, 2014 at 05:40:05AM +0800, sunyeping wrote:
> Dear professor Vermaas,
> Thank you for taking time to explain the usage. I have found the
> namdenergy.tcl script using env command and added +idlepoll after exce
> $namdcmd. But when I run the script the following error comes out:
> FATAL ERROR: Pair interaction calculation is not supported in CUDA version
>
> while executing
> "exec $namdcmd +idlepoll ${jobname}-temp.namd > ${jobname}-temp.log"
> So in what cuda version is the Pair interaction calculation supported?
> Then I switched the NAMD of CPU version. This time the script worked.
> Thank you very much.
> Yeping
> ------------------------------------------------------------------
> aa**a:>>P:a:-o-oi 1/4*Josh Vermaas <vermaas2_at_illinois.edu>
> aa**e**ae*P:e*'i 1/4*2014aa^1'5ae**19ae*YEN(ae**ae**a:,*) 05:11
> ae*P:a:>>P:a:-o-oi 1/4*aa*a:,*aa^1^3 <sunyeping_at_aliyun.com>; Axel
> Kohlmeyer <akohlmey_at_gmail.com>
> ae**a**e**i 1/4*vmd-l <vmd-l_at_ks.uiuc.edu>
> a:,>>a**e-c-*i 1/4*Re: vmd-l: REi 1/4*vmd-l: proper enviroment for the VMD
> namdenergy plugin
>
> Hi Yeping,
>
> What Axel is saying is that because you need to add +idlepoll to the
> command line in NAMD 2.9 when running a GPU-enabled build, the error you
> are getting is actually perfectly normal (although undesirable). What you
> need to do to fix this is change a line in the tcl script that runs the
> plugin. To do this, first figure out the directory where the NAMDEnergy
> plugin is stored. env is your friend here. From the tkconsole of VMD, the
> "env" command will give you a bunch of information, but you want what is
> at the end, the plugin directories. Find the NAMDENERGYDIR line, and open
> the namdenergy.tcl file in that directory. You'll find many lines like
> this:
>
> exec $namdcmd something.namd > something.log
>
> I think its just a simple matter of adding +idlepoll between the $namdcmd
> and the configuration file, and it should no longer give you that error.
>
> This was the technical solution. Most of the time non-technical solutions
> are better. Neither the CPU-only version of NAMD nor the current
> GPU-enabled CVS builds of NAMD need the +idlepoll argument, so you could
> just try using those. :)
> -Josh Vermaas
>
> On 5/18/14, 1:00 PM, sunyeping wrote:
>
> Yeping Sun
>
> Institute of Microbiology, Chinese Academy of Sciences
>
> ------------------------------------------------------------------
> aa**a:>>P:a:-o-oi 1/4*Axel Kohlmeyer <akohlmey_at_gmail.com>
> aa**e**ae*P:e*'i 1/4*2014aa^1'5ae**18ae*YEN(ae**ae**ae*YEN) 21:08
> ae*P:a:>>P:a:-o-oi 1/4*aa*a:,*aa^1^3 <sunyeping_at_aliyun.com>
> ae**a**e**i 1/4*vmd-l <vmd-l_at_ks.uiuc.edu>
> a:,>>a**e-c-*i 1/4*Re: vmd-l: proper enviroment for the VMD namdenergy
> plugin
>
> On Sun, May 18, 2014 at 1:30 AM, sunyeping <sunyeping_at_aliyun.com> wrote:
>
> Dear VMD users,
>
> a**[...]a**
>
> My cuda version is cuda_5.0.35_linux_64_rhel5.x-1.run, my VMD version is 1.9.1
>
> In order to run the namdenergy plugin properly, which version of cuda
> do I need? should I upgrade or downgrade it?
>
> a**neither. you obviously have to use a CPU version of NAMD.
> axela**
> Dear axel,
> Thank you for you reply! I have installed NAMD of CPU version:
> NAMD_CVS-2014-05-18_Linux-x86_64-multicore. But there is still one thing
> that I don't understand: how should I make the namdenergy.tcl get the
> path of namd2 in this CPU version of NAMD ( say, /opt/NAMD_CPU) rather
> than that of my previously installed NAMD of CUDA version (in
> /opt/NAMD_CUDA)? Should I fillful it by modifying the namdenergy.tcl
> script and how?
> Best regards,
>
> Yeping
>
> --
> Dr. Axel Kohlmeyer akohlmey_at_gmail.com http://goo.gl/1wk0
> College of Science & Technology, Temple University, Philadelphia PA, USA
> International Centre for Theoretical Physics, Trieste. Italy. 

-- 
NIH Center for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
http://www.ks.uiuc.edu/~johns/           Phone: 217-244-3349
http://www.ks.uiuc.edu/Research/vmd/