VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Wed Dec 16 2009 - 23:55:15 CST

Rich,
  From the startup messages you've provided below, it looks to
me like this is being caused by another bug in the GLX implementation
on the MacOS X11 server side. In particular, I see that the X11
server is claiming that it has stereo, and so VMD went ahead and
asked for a stereoscopic buffer. I seriously doubt however that
the Mac in question actually has a stereo-capable GPU, so something
is going amiss there. This would explain the flashing behavior,
as more likely than not there's some bogus activity going on with
buffer swaps if stereo is enabled but not actually supported by
the hardware. Try having your colleague set the environment
variable VMDDISABLESTEREO to 1 prior to running VMD, and see
if that cures the flashing when you use the real OpenGL:
/bin/bash:
  VMDDISABLESTEREO=1
  export VMDDISABLESTEREO

/bin/csh
  setenv VMDDISABLESTEREO 1
  
Cheers,
  John Stone
  vmd_at_ks.uiuc.edu

On Tue, Dec 15, 2009 at 02:18:49PM -0800, Rich Cook wrote:
> Sure here you go. To be honest, I suspect X11's involvement due to
> some of the description from the end user, which I have not
> duplicated. He claims that on his machine, the flickering occurs
> where the VMD window used to be as well as where it is, which
> definitely points at X11 involvement. However, since only VMD
> displays this, I thought I'd try you first.
>
> There is this of note as well:
> http://xquartz.macosforge.org/trac/ticket/252
>
> VMDDISABLEATILINUXWORKAROUND does not do anything to help.
>
> rcook_at_yana2 (~): vmd
> Info) VMD for LINUXAMD64, version 1.8.7 (August 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, 8 CPUs detected.
> Info) Free system memory: 10931MB (72%)
> Info) No CUDA accelerator devices available.
> Info) ATI Linux driver detected, limiting features to avoid driver bugs.
> Info) Set the environment variable VMDDISABLEATILINUXWORKAROUND
> Info) to enable full functionality on a known-safe driver version.
> Info) OpenGL renderer: ATI Radeon 9600 OpenGL Engine
> Info) Features: STENCIL STEREO MDE MTX NPOT PP PS
> Info) GLSL rendering mode is NOT available.
> Info) Textures: 2-D (2048x2048), Multitexture (8)
> vmd > Info) VMD for LINUXAMD64, version 1.8.7 (August 1, 2009)
> Info) Exiting normally.
>
>
> On Dec 15, 2009, at 1:20 PM, John Stone wrote:
>
> >
> >Rich,
> > Can you send me the VMD startup messages that get printed on the
> >text console when you're remote displaying on one of the machines
> >that flickers? Whether you render with Mesa or some other OpenGL
> >shouldn't have much effect since the rendering should be done on
> >the local Mac display and not on the remote machine, and in particular
> >since the glXSwapBuffers() call will be handled entirely on the Mac
> >side and not by the remote machine. So the results you got with your
> >Mesa test don't surprise me at present. The bugs in the ATI drivers
> >have historically affected only the Linux platform, and not MacOS X
> >or Windows, so I don't expect this has much to do with the graphics
> >chipset if you only see this behavior when using a remote X11
> >connection. If you saw the same flickering behavior on a locally
> >displayed VMD session, then I would be more apt to question the
> >ATI driver on the Mac. Send me those VMD startup messages on
> >the afflicted machines, and I'll ponder what might be causing this
> >other than some sort of bug on the MacOS X X11/GLX server side.
> >
> >Cheers,
> > John Stone
> > vmd_at_ks.uiuc.edu
> >
> >On Tue, Dec 15, 2009 at 01:03:07PM -0800, Rich Cook wrote:
> >>I'm wondering if folks here can help me troubleshoot this issue.
> >>
> >>We have VMD installed on our large computer clusters and users like
> >>to
> >>visualize their data and display back to their desktops. This is
> >>more
> >>convenient than moving the data.
> >>
> >>A few of my users are having flickering issues with VMD on their Mac
> >>laptops. The symptoms: VMD runs normally, but the screen flickers
> >>between a garbage image, mostly white, and the correct display. This
> >>makes VMD unusable for them in the way they'd like to be using it.
> >>
> >>They are running Mac OS X 10.5.8, using Apple's X11 (X.org) version
> >>2.1.6 or the newer XQuartz 2.4.0 from
> >>http://*xquartz.macosforge.org/trac/wiki and using ATI graphics cards
> >>(models 2600 and 4780). I have read "out there" that there are
> >>ATI issues
> >>with VMD. True?
> >>
> >>This problem exists on three different Mac laptops we've tested here.
> >>I don't have any problems displaying to my mac in my office, using
> >>10.5.8 and XQuarts 2.4.0 on an intel mac with nVidia GeForce 7300
> >>card. They do not have any problem when running VMD on their desktop,
> >>only when displaying back. I suspected GLX at first, it looked like
> >>something was up with glSwapBuffers. To test this, I tried setting
> >>LD_LIBRARY_PATH on the remote machine so that VMD loads OS Mesa and
> >>thus does software rendering, but this surprisingly does not
> >>eliminate
> >>the flicker.
> >>
> >>How to proceed to get more information? I can run in debug mode, but
> >>don't know what to look for. I tried changing graphics settings in
> >>the menu, but none made any difference.
> >>
> >>thanks!
> >>/* A function that takes a single integer argument and returns
> >> a pointer to a function that takes two integer arguments and
> >> returns a floating-point number. */
> >>float (*func2(int a))(int, int);
> >>Rich Cook
> >>rcook_at_llnl.gov
> >>
> >>
> >>
> >
> >--
> >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
> >
>
>
> /* A function that takes a single integer argument and returns
> a pointer to a function that takes two integer arguments and
> returns a floating-point number. */
> float (*func2(int a))(int, int);
> Rich Cook
> rcook_at_llnl.gov
>
>
> 

-- 
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