VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Wed Jul 15 2009 - 13:48:13 CDT

Hi,
  It turns out that the root of the problem here is that the MacOS X 10.5.7
X11 server is incorrectly reporting that it supports stereo, when it actually
does not. I just implemented a workaround for this bug in the MacOS X
X11 server that you and several others have encountered. Beginning with
VMD 1.8.7, there will be a new "VMDDISABLESTEREO" environment variable
you can use to disable stereoscopic OpenGL rendering regardless what the
remote X11 server claims to support. This allows you to retain GLSL and
the other advanced features of VMD but prevent a buggy X11 server (e.g.
Apple's) from telling VMD it can do stereo when in fact it cannot.

Until Apple fixes the bug in their X11 server, you can just set
the VMDDISABLESTEREO environment variable to 1, and things should
work as they should. This will be in the publically accessible CVS tree
late tonight, for those that compile from source.

Cheers,
  John Stone
  vmd_at_ks.uiuc.edu

On Fri, Jun 12, 2009 at 06:33:47PM +0200, Benjamin Bouvier wrote:
> >So, my opinion is that it looks to me like the new version of
> >MacOS X has a buggy X11 server (or video driver), and that we will
> >need to file a bug with Apple on this.
>
> Thanks John. I suspected as much -- lots of complaints have popped up on
> the web about things that this new X11 server has broken.
> Hopefully an update from http://xquartz.macosforge.org/trac/wiki/Releases
> will eventually fix the problem. Until then, I would advise people to
> stick to 10.5.6, as there is no going back :-(
> Cheers, B.
>
>
> >
> >Cheers,
> > John Stone
> > vmd_at_ks.uiuc.edu
> >
> >On Fri, Jun 12, 2009 at 05:45:37PM +0200, Benjamin Bouvier wrote:
> >>John,
> >>
> >>Here is the log. It appears to be more related to the Linux machine
> >>actually running VMD than to the Macbook doing the displaying, though...
> >>
> >>Info) VMD for LINUXAMD64, 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, 4 CPUs detected.
> >>Info) Free system memory: 7302MB (94%)
> >>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 X1600 OpenGL Engine
> >>Info) Features: STENCIL STEREO MSAA(6) MDE MTX NPOT PP PS
> >>Info) GLSL rendering mode is NOT available.
> >>Info) Textures: 2-D (4096x4096), Multitexture (8)
> >>vmd >
> >>
> >>On Fri, 12 Jun 2009 17:36:49 +0200, John Stone <johns_at_ks.uiuc.edu>
> >>wrote:
> >>
> >>>
> >>>Benjamin,
> >>> If you run one of the recent VMD 1.8.7 beta versions, the console
> >>>output will occur in whatever terminal you started VMD from. So this
> >>>would be on your original ssh/xterm session, etc. Check there.
> >>>
> >>>Cheers,
> >>> John
> >>>
> >>>On Fri, Jun 12, 2009 at 05:28:17PM +0200, Benjamin Bouvier wrote:
> >>>>Hi John,
> >>>>
> >>>>When I tried to find the VMD console to check the startup log, I was
> >>>>unable to find it, though it is listed as opened by the X11 server. It
> >>>>seems that *somehow* the console opens inside the OpenGL window, hence
> >>>>the
> >>>>flashing -- what I had described as white frames in my previous posts
> >>>>seem
> >>>>in fact to contain garbled text (it varies upon dragging the window
> >>>>around), which is reminiscent of the console window...
> >>>>
> >>>>Is there a startup switch to log the console output to some file? If
> >>>>so, I
> >>>>could retrieve the startup message there. I looked quickly in the
> >>manual
> >>>>but could not find one...
> >>>>
> >>>>Cheers,
> >>>>Benjamin
> >>>>
> >>>>On Fri, 12 Jun 2009 17:02:33 +0200, John Stone <johns_at_ks.uiuc.edu>
> >>>>wrote:
> >>>>
> >>>>>
> >>>>>Benjamin,
> >>>>> Can you send me the VMD startup messages you get when you
> >>>>>display on the remote machine? The description you're giving
> >>>>>sounds like they've somehow created a new bug/problem with
> >>>>>OpenGL double buffering. I don't recall if glxgears is using
> >>>>>double buffering or not, so the lack of trouble there may
> >>>>>not be indicative of anything. If you send me the VMD startup
> >>>>>messages, it will tell me more about what features their X11
> >>>>>server implementation is advertising to the client Linux system,
> >>>>>and with this we may be able to narrow down the source of the
> >>problem.
> >>>>>
> >>>>>CHeers,
> >>>>> John Stone
> >>>>> vmd_at_ks.uiuc.edu
> >>>>>
> >>>>>On Fri, Jun 12, 2009 at 04:48:29PM +0200, Benjamin Bouvier wrote:
> >>>>>>Hi all,
> >>>>>>
> >>>>>>I usually run VMD on a remote Linux machine (cluster frontend) and
> >>use
> >>>>>>MacOs's X11 implementation to forward the display to my MacBook Pro.
> >>>>>>This
> >>>>>>used to work fine.
> >>>>>>
> >>>>>>No so since I upgraded MacOS from 10.5.6 to the latest 10.5.7. The
> >>>>>>contents of the VMD OpenGL window flash very quickly (one frame is
> >>>>>>white,
> >>>>>>the second one is the actual VMD rendering), making VMD unusable.
> >>This
> >>>>>>occurs with VMD 1.8.6 and 1.8.7, both 32 and 64 bits.
> >>>>>>
> >>>>>>I read that Apple had done modifications to X11 in 10.5.7. However,
> >>>>the
> >>>>>>problem seems limited to VMD : glxgears for instance is forwarded
> >>OK,
> >>>>as
> >>>>>>are all non-openGL windows (including the VMD Main window).
> >>>>>>
> >>>>>>Are there any other users experiencing this ? Is there a simple
> >>cure ?
> >>>>>>
> >>>>>>Thanks,
> >>>>>>--
> >>>>>>Benjamin Bouvier
> >>>>>
> >>>
> >>
> >>
> >> 

-- 
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