From: John Stone (johns_at_ks.uiuc.edu)
Date: Fri Jun 05 2009 - 11:04:08 CDT

Hi,
  From what you're saying, it sounds like you get correct behavior
with the "nv" driver, but not with the NVIDIA provided display driver,
is that correct? If the "nv" driver works (but would obviously be
very slow and low on features), then that does seem to indicate that
this is an OpenGL driver issue. Can you try using a newer or older
version of the NVIDIA driver available from the nvidia web site
and see if the problem goes away:
  http://www.nvidia.com/

Cheers,
  John Stone
  vmd_at_ks.uiuc.edu

On Fri, Jun 05, 2009 at 05:56:04PM +0200, lucie delemotte wrote:
> Hi,
>
> >i would thus suspect that there is some problem or incompatibility
> >with your OpenGL library relative to your Xserver. since you have
> >a very new distribution, and it seems to contain somewhat experimental
> >software (you mentioned Tk8.6, which is not even officially released),
> >i would suggest you try a few more steps to narrow down the possible
> >causes. one option to try would be to disable some advanced OpenGL
> >features via using: env VMDSIMPLEGRAPHICS=1 vmd -e state.vmd
> >and see if that makes a difference.
>
> It makes no difference. Regrding the tk/tcl version, this is the only
> version available on this distribution (tcl-8.6-0.b1.1mdv2009.1.i586.rpm and
> tk-8.6-0.b1.2mdv2009.1.i586.rpm)
>
> >the second, more drastic option would be to uninstall the nvidia
> >driver and try the (limited functionality) "nv" driver that ships
> >with the distribution instead.
> >i would also make sure that you have all patches and updates installed.
> >new distributions, particularly those that are willing to ship beta
> >software, tend to have a lot of updates in the first few months.
>
> The OS is newly installed and all updates are done but it does not change
> anything. Changing the graphical to "nv" in the xorg file permits the state
> file to load but as it does not activate GL functions (or other stuff like
> that), using vmd in this configuration is painful. I also tried to use the
> version given by the mandriva distribution and gives the same behavior as
> with the drivers taken from the nivdia website.
>
> >Do you have Compiz or other
> >desktop special effects features enabled on this machine?
>
> Compiz is deactivated for the moment but if activated, it changes nothing
> regarding vmd.
> Besides, compiz works well when activated, and glxgears also gives good fps.
>
> >If all else fails,
> >trying alternative non-vendor drivers may be worthwhile to narrow down the
> >source of trouble.
>
> Can you be more specific on this information ?
>
> Thanks again,
> Lucie
>
> On Wed, Jun 3, 2009 at 9:05 PM, John Stone <johns_at_ks.uiuc.edu> wrote:
>
> >
> > Hi,
> > Yeah, I think Axel's suggestions here are probably on the mark.
> > I've never encountered the problem you're describing with the
> > NVIDIA's drivers however, so I'm somewhat surprised that you're
> > having this problem on your particular combination of hardware
> > and software. If changing the drivers affects the behavior, then
> > we may need to file a bug to NVIDIA. Do you have Compiz or other
> > desktop special effects features enabled on this machine? That's
> > another thing that may be worth checking on in addition to what
> > Axel already suggested. I would try and solve the problem while
> > retaining the use of the NVIDIA-provided OpenGL drivers if possible,
> > either by using an older driver version or by enabling/disabling various
> > options, since they will give the best performance. If all else fails,
> > trying alternative non-vendor drivers may be worthwhile to narrow down the
> > source of trouble.
> >
> > Cheers,
> > John Stone
> > vmd_at_ks.uiuc.edu
> >
> > On Wed, Jun 03, 2009 at 09:16:06AM -0400, Axel Kohlmeyer wrote:
> > > On Wed, 2009-06-03 at 09:36 +0200, lucie delemotte wrote:
> > > > Hi,
> > >
> > > lucie,
> > >
> > > i tried your input as well on a variety of linux machines
> > > (32-bit, 64-bit, "old" linux, "new" linux, nvidia, non-nvidia)
> > > and it always worked.
> > >
> > > [...]
> > >
> > > > After that, I get no vmd> prompt. The openGL window opens but doesn't
> > > > show anything and no main window appears. I then have to kill the
> > > > process to quit.
> > >
> > > i would thus suspect that there is some problem or incompatibility
> > > with your OpenGL library relative to your Xserver. since you have
> > > a very new distribution, and it seems to contain somewhat experimental
> > > software (you mentioned Tk8.6, which is not even officially released),
> > > i would suggest you try a few more steps to narrow down the possible
> > > causes. one option to try would be to disable some advanced OpenGL
> > > features via using: env VMDSIMPLEGRAPHICS=1 vmd -e state.vmd
> > > and see if that makes a difference.
> > >
> > > the second, more drastic option would be to uninstall the nvidia
> > > driver and try the (limited functionality) "nv" driver that ships
> > > with the distribution instead.
> > > i would also make sure that you have all patches and updates installed.
> > > new distributions, particularly those that are willing to ship beta
> > > software, tend to have a lot of updates in the first few months.
> > >
> > > since it was working with -debug, it could be some sort of race
> > > condition somewhere (debugging slows down execution).
> > >
> > > > I also forgot to mention that every script I have works well in the
> > > > -dispdev text mode.
> > >
> > > which just confirms that this is an X-related problem.
> > >
> > > cheers,
> > > axel.
> > > > Cheers again,
> > > > Lucie
> > > >
> > > > On Tue, Jun 2, 2009 at 9:01 PM, John Stone <johns_at_ks.uiuc.edu> wrote:
> > > >
> > > > Lucie,
> > > > After fixing an invalid filename for the .coor file (was
> > > > looking
> > > > in a directory
> > > > "'/media/disk/LUCIE/ionchannels/kv1/kv1.5/kv1.5.26/"
> > > > which obviously doesn't exist on my machine), the state file
> > > > you provided loads fine for me with both VMD 1.8.7 beta 5 and
> > > > VMD 1.8.6.
> > > > If you run VMD 1.8.7 beta 5 and load the saved state via the
> > > > command
> > > > line -e flag, how far does it get? Can you mail me the
> > > > output from
> > > > that test? (no -debug)
> > > >
> > > > Cheers,
> > > > John Stone
> > > > vmd_at_ks.uiuc.edu
> > > >
> > > >
> > > >
> > > > On Tue, Jun 02, 2009 at 06:42:38PM +0200, lucie delemotte
> > > > wrote:
> > > > > Hi,
> > > > > I uploaded the files to biocore (called state.tar.gz).
> > > > > Thanks again.
> > > > >
> > > > > Lucie
> > > > >
> > > > > On Tue, Jun 2, 2009 at 6:15 PM, John Stone
> > > > <johns_at_ks.uiuc.edu> wrote:
> > > > >
> > > > > > Hi,
> > > > > > Can you tar up the saved state and related files, gzip
> > > > it, and
> > > > > > either post it on biocore or email it to me (if it is
> > > > small enough)?
> > > > > > I'd like to try loading it for myself so I can see what's
> > > > going on there.
> > > > > >
> > > > > > Cheers,
> > > > > > John Stone
> > > > > > vmd_at_ks.uiuc.edu
> > > > > >
> > > > > > On Tue, Jun 02, 2009 at 05:48:19PM +0200, lucie delemotte
> > > > wrote:
> > > > > > > Thanks for the quick answer!
> > > > > > > Yes, vmd works fine when not loading a script (all
> > > > windows open and I can
> > > > > > > load things from the graphical menu).
> > > > > > > Secondly, the script is supposed to load a pdb file plus
> > > > a namdbin file.
> > > > > > > (With the other computer or in the debug mode, I get the
> > > > same text as
> > > > > > above
> > > > > > > plus :
> > > > > > >
> > > > > > > Info) Using plugin pdb for coordinates from file
> > > > > > > kv1.5.26.0.pdb
> > > > > > > Info) Finished with coordinate file
> > > > > > > kv1.5.26.0.pdb.
> > > > > > > 0
> > > > > > >
> > > > > > > Info) Using plugin namdbin for coordinates from file ...
> > > > > > > ..........
> > > > > > > ........
> > > > > > > ........
> > > > > > > ........
> > > > > > > [New Thread 0xa8914b90 (LWP
> > > > > > > 25252)]
> > > > > > > [Thread 0xa8914b90 (LWP 25252)
> > > > > > > exited]
> > > > > > > [New Thread 0xa8113b90 (LWP
> > > > > > > 25253)]
> > > > > > > [Thread 0xa8113b90 (LWP 25253)
> > > > > > > exited]
> > > > > > > [New Thread 0xa8113b90 (LWP
> > > > > > > 25254)]
> > > > > > > [Thread 0xa8113b90 (LWP 25254)
> > > > > > > exited]
> > > > > > > [New Thread 0xa8914b90 (LWP
> > > > > > > 25255)]
> > > > > > > [Thread 0xa8914b90 (LWP 25255)
> > > > > > > exited]
> > > > > > > [New Thread 0xa8914b90 (LWP
> > > > > > > 25256)]
> > > > > > > [New Thread 0xa8113b90 (LWP
> > > > > > > 25257)]
> > > > > > > [Thread 0xa8914b90 (LWP 25256)
> > > > > > > exited]
> > > > > > > [Thread 0xa8113b90 (LWP 25257)
> > > > > > > exited]
> > > > > > > {{1 0 0 0.654825} {0 1 0 -0.0264271} {0 0 1 -9.97098} {0
> > > > 0 0 1}}
> > > > > > {{-0.728706
> > > > > > > -0.681299 -0.0693803 0} {-0.060971 -0.0363649 0.997481
> > > > 0} {-0.682102
> > > > > > > 0.731099 -0.01504 0} {0 0 0 1}} {{0.0266042 0 0 0} {0
> > > > 0.0266042 0 0} {0 0
> > > > > > > 0.0266042 0} {0 0 0 1}} {{1 0 0 0} {0 1 0 0} {0 0 1 0}
> > > > {0 0 0
> > > > > > > 1}}
> > > > > > > 0
> > > > > > >
> > > > > > > 0
> > > > > > >
> > > > > > > [New Thread 0xa8113b90 (LWP
> > > > > > > 25258)]
> > > > > > > [New Thread 0xa8914b90 (LWP
> > > > > > > 25259)]
> > > > > > > [Thread 0xa8914b90 (LWP 25259)
> > > > > > > exited]
> > > > > > > [Thread 0xa8113b90 (LWP 25258)
> > > > > > > exited]
> > > > > > >
> > > > > > > vmd >
> > > > > > >
> > > > > > >
> > > > > > > and the graphical window works fine.
> > > > > > >
> > > > > > > Lucie
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > On Tue, Jun 2, 2009 at 5:39 PM, John Stone
> > > > <johns_at_ks.uiuc.edu> wrote:
> > > > > > >
> > > > > > > >
> > > > > > > > Hi,
> > > > > > > > Does VMD run correctly for you by itself? The main
> > > > menu should
> > > > > > > > appear if it was installed correctly and you don't
> > > > have an empty .vmdrc
> > > > > > > > in your home directory. Once you've verified that VMD
> > > > itself is
> > > > > > working
> > > > > > > > fine, the next question I'd ask is whether you're
> > > > loading any
> > > > > > coordinates
> > > > > > > > with that PSF file or just the PSF file. Unless you
> > > > load coordinates
> > > > > > as
> > > > > > > > well, nothing will be displayed in the VMD graphics
> > > > window. VMD needs
> > > > > > > > the atomic coordinates to be able to draw anything.
> > > > The PSF file only
> > > > > > > > contains the structure data (e.g.
> > > > bonds/angles/dihedrals/etc). When
> > > > > > you
> > > > > > > > loaded it in VMD 1.8.6, did the saved state also load
> > > > a PDB file or
> > > > > > some
> > > > > > > > other coordinate file, or not?
> > > > > > > >
> > > > > > > > Cheers,
> > > > > > > > John Stone
> > > > > > > > vmd_at_ks.uiuc.edu
> > > > > > > >
> > > > > > > > On Tue, Jun 02, 2009 at 05:31:26PM +0200, lucie
> > > > delemotte wrote:
> > > > > > > > > Hi all,
> > > > > > > > >
> > > > > > > > > I have a vmd state script (state.vmd) that was
> > > > working properly with
> > > > > > vmd
> > > > > > > > > 1.8.6.
> > > > > > > > > After upgrading from mandriva 2008.0 to mandriva
> > > > 2009.1 and changing
> > > > > > to
> > > > > > > > vmd
> > > > > > > > > 1.8.7, the script stopped loading (either in command
> > > > line with vmd -e
> > > > > > > > > state.vmd or in the Tkconsole with the play
> > > > command). The openGL
> > > > > > window
> > > > > > > > pops
> > > > > > > > > up but nothing appears in it while the vmd main
> > > > window doesn't open.
> > > > > > In
> > > > > > > > the
> > > > > > > > > console, the regular vmd stuff appears :
> > > > > > > > >
> > > > > > > > > Info) VMD for LINUX, version 1.8.7beta5 (June 1,
> > > > 2009)
> > > > > > > > >
> > > > > > > > > Info) http://www.ks.uiuc.edu/Research/vmd/
> > > > > > > > >
> > > > > > > > > Info) Email questions and bug reports to
> > > > vmd_at_ks.uiuc.edu
> > > > > > > > >
> > > > > > > > > Info) Please include this reference in published
> > > > work using VMD:
> > > > > > > > >
> > > > > > > > > Info) Humphrey, W., Dalke, A. and Schulten, K.,
> > > > `VMD - Visual
> > > > > > > > >
> > > > > > > > > Info) Molecular Dynamics', J. Molec. Graphics
> > > > 1996, 14.1, 33-38.
> > > > > > > > >
> > > > > > > > > Info)
> > > > -------------------------------------------------------------
> > > > > > > > >
> > > > > > > > > Info) Multithreading available, 2 CPUs detected.
> > > > > > > > >
> > > > > > > > > Info) Free system memory: 326MB (16%)
> > > > > > > > >
> > > > > > > > > Info) No CUDA accelerator devices available.
> > > > > > > > >
> > > > > > > > > Info) OpenGL renderer: Quadro FX 3450/4000
> > > > SDI/PCI/SSE2
> > > > > > > > >
> > > > > > > > > Info) Features: STENCIL MSAA(16) MDE CVA MTX NPOT
> > > > PP PS GLSL(OVF)
> > > > > > > > >
> > > > > > > > > Info) Full GLSL rendering mode is available.
> > > > > > > > >
> > > > > > > > > Info) Textures: 2-D (4096x4096), 3-D
> > > > (512x512x512), Multitexture
> > > > > > (4)
> > > > > > > > >
> > > > > > > > > psfplugin) Detected a Charmm31 PSF file
> > > > > > > > >
> > > > > > > > > Info) Using plugin psf for structure file
> > > > kv1.5.26.0.psf
> > > > > > > > >
> > > > > > > > > Info) Analyzing structure ...
> > > > > > > > >
> > > > > > > > > Info) Atoms: 305464
> > > > > > > > >
> > > > > > > > > Info) Bonds: 221738
> > > > > > > > >
> > > > > > > > > Info) Angles: 172116 Dihedrals: 117638
> > > > Impropers: 4748
> > > > > > Cross-terms:
> > > > > > > > > 1552
> > > > > > > > > Info) Bondtypes: 0 Angletypes: 0 Dihedraltypes:
> > > > 0
> > > > > > Impropertypes:
> > > > > > > > > 0
> > > > > > > > > Info) Residues:
> > > > > > > > > 85614
> > > > > > > > > Info) Waters:
> > > > > > > > > 83392
> > > > > > > > > Info) Segments:
> > > > > > > > > 21
> > > > > > > > > Info) Fragments: 84058 Protein: 4 Nucleic:
> > > > > > > > > 0
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > I tested versions beta.2, beta.3 and beta.5
> > > > unsuccessfully. I also
> > > > > > > > checked
> > > > > > > > > that the script is still working on another
> > > > computer. It also appears
> > > > > > > > that
> > > > > > > > > vmd 1.8.6 is not working at all (vmd main window not
> > > > appearing) on
> > > > > > the
> > > > > > > > new
> > > > > > > > > installation.
> > > > > > > > > The Nvidia drivers are well installed with the
> > > > latest version :
> > > > > > 180.51
> > > > > > > > > Tcl/Tk version 8.6
> > > > > > > > > gcc (GCC) 4.3.2
> > > > > > > > >
> > > > > > > > > I tried the debug mode to check what was wrong and
> > > > interstingly, my
> > > > > > > > > state.vmd file loads properly (gdb version :
> > > > 6.8-6mdv2009.1).
> > > > > > > > >
> > > > > > > > > Any help would be greatly appreciated.
> > > > > > > > > Lucie
> > > > > > > >
> > > > > > > > --
> > > > > > > > NIH Resource for Macromolecular Modeling and
> > > > Bioinformatics
> > > > > > > > Beckman Institute for Advanced Science and Technology
> > > > > > > > University of Illinois, 405 N. Mathews Ave, Urbana, IL
> > > > 61801
> > > > > > > > Email: johns_at_ks.uiuc.edu Phone:
> > > > 217-244-3349
> > > > > > > > WWW:
> > > > http://www.ks.uiuc.edu/~johns/>
> > <
http://www.ks.uiuc.edu/%
> > > > 7Ejohns/><
> > > > > > http://www.ks.uiuc.edu/%7Ejohns/>
> > > > > > > > Fax: 217-244-6078
> > > > > > > >
> > > > > >
> > > > > > --
> > > > > > NIH Resource for Macromolecular Modeling and
> > > > Bioinformatics
> > > > > > Beckman Institute for Advanced Science and Technology
> > > > > > University of Illinois, 405 N. Mathews Ave, Urbana, IL
> > > > 61801
> > > > > > Email: johns_at_ks.uiuc.edu Phone:
> > > > 217-244-3349
> > > > > > WWW: http://www.ks.uiuc.edu/~johns/>
> > > > <
http://www.ks.uiuc.edu/%7Ejohns/>
> > > > > > Fax: 217-244-6078
> > > > > >
> > > >
> > > > --
> > > > NIH Resource for Macromolecular Modeling and Bioinformatics
> > > > Beckman Institute for Advanced Science and Technology
> > > > University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> > > > Email: johns_at_ks.uiuc.edu Phone: 217-244-3349
> > > > WWW: http://www.ks.uiuc.edu/~johns/> Fax: 217-244-6078
> > > >
> > > >
> > >
> > > --
> > > =======================================================================
> > > Axel Kohlmeyer
akohlmey_at_cmm.chem.upenn.edu http://www.cmm.upenn.edu
> > > Center for Molecular Modeling -- University of Pennsylvania
> > > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
> > > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425
> > > =======================================================================
> > > If you make something idiot-proof, the universe creates a better idiot.
> >
> > --
> > NIH Resource for Macromolecular Modeling and Bioinformatics
> > Beckman Institute for Advanced Science and Technology
> > University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> > Email: johns_at_ks.uiuc.edu Phone: 217-244-3349
> > WWW: http://www.ks.uiuc.edu/~johns/ <http://www.ks.uiuc.edu/%7Ejohns/>
> > Fax: 217-244-6078
> >

-- 
NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
Email: johns_at_ks.uiuc.edu                 Phone: 217-244-3349
  WWW: http://www.ks.uiuc.edu/~johns/      Fax: 217-244-6078