From: John Stone (johns_at_ks.uiuc.edu)
Date: Thu Dec 15 2016 - 11:37:46 CST

Thanks, this is very useful information, particularly that you have
the same problem whether 32-bit or 64-bit. I assume from this that your
32-bit FLTK build was done using Cocoa rather than Carbon, right?

Based on what you described before in addition to what you say below,
there could be a problem with the sequence of OpenGL operations vs.
the active Window and currently active OpenGL context.

VMD uses FLTK for the MacOS X OpenGL display, but on Windows and Unix,
we use our own code. An issue with this is that if FLTK has done something
like making the VMD OpenGL Window's rendering context non-current, I could
envision some problems cropping up if we have an OpenGL call sequence that
occurs at any time when the context is non-current for the Window owned by
MacOS X.

Another possibility is that there is some new issue that is related
to MacOS X desktop compositing going on. If it is a compositing issue,
that also might explain why you get a good screen shot from the
VMD snapshot renderer, but it doesn't look right in the interactive display.

I'm probably going to have to dig into this because I'm fairly sure it's a
non-trivial issue, either with VMD itself, or its interactions with
FLTK, based on the results you've already collected.

Cheers,
  John Stone
  vmd_at_ks.uiuc.edu

On Thu, Dec 15, 2016 at 01:42:20PM +0100, FX wrote:
> Yet some more information: the bug is not 64-bit specific, as I can reproduce it with a 32-bit based on fltk-1.3.4-1
>
> FX

-- 
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/