[vtk-developers] 2D API: Bug in older OpenGL point sprites?

[Marcus D. Hanwell]

Hi Eric,

On Fri, Jan 7, 2011 at 9:23 AM, Eric Monson, Ph.D. <eric.monson at duke.edu> wrote:
> Hey Marcus,
>
> I was trying to get my application running under 32-bit Fedora 13 in a virtual machine (VMWare Fusion) on my Mac, and had trouble with the color-by-array rendering on the ChartXY but not on the Parallel Coordinates. Looking through the dashboards I was able to find one machine that exhibits exactly the same behavior (hythloth.kitware):
>
> http://www.cdash.org/CDash/testDetails.php?test=78582182&build=818285

I fixed that bug yesterday - just pull.
>
> When I run similar tests on other machines it always says "Using OpenGL 2 for the render" in the terminal and looks fine. On this Fedora VM it doesn't say that, so I'm assuming it's using the older OpenGL implementation of the point sprite renderer, and maybe this is where the problem is. I never tested that backup (non OpenGL 2) color-by-array routine, so maybe there's a bug in it someplace...?

I removed that debug output in the same set of commits.
>
> Let me know if you have any quick ideas. I don't have time to track it down right now, but I'll try to take a look in the near future. I can also try to get C++ tests written up for the color-by-array functionality since it looks like not all machines build VTK with Python support and thus never run the current tests.
>
Just pull the latest sources, and let me know if this is still an
issue. It is an OpenGL 1 card/mesa and so the rendering may not be as
fast.

Marcus