Jump to content

The big red rectangle


leszekp

Recommended Posts

I sent a few PTM files to a friend, and he reported that when he opened the files in the main RTIViewer window was filled with solid red - no view of the PTM image. The image in the lower right hand corner shows up correctly, and if you save a snapshot, the view is also correct. He also sees the same problem with the HP PTMViewer program. The original problem was on a Windows 7 system, but we've duplicated it on two Windows XP systems of his as well. What's more, PTM files that he was able to view successfully in the past now also have the same problem. His graphics drivers and Java version are fully up to date. I've tested these files on three of my Windows systems, and they work perfectly on all of them. I'm stumped - any ideas?

Link to comment
Share on other sites

I thought this had been reported on the forum before, but I can't find it.

 

This problem occurs rarely on windows machines.  It seems to be machine specific and does not appear to be a problem with the files, which can be opened on other machines.  We have never been able to reliably reproduce it for the developers to figure it out.  It does not seem to be specific to a particular OS, or any other configuration data we have.  It has only happened a handful of times (that we know of)

 

We are working on a new version of RTIViewer, and I asked someone who reported this bug recently to tryout a test version, and unfortunately, it was not resolved.  The weird thing is that our code is identical for Windows and Mac, we just compile for the different platforms using a set of open source libraries called QT.  I'll note that the RTIViewer is written in C++, RTIBuilder is written in Java. So, my guess is that the problem is actually in the libraries, since we don't do any low level rendering to the screen.  The fact that the data is there, and that you can get a snapshot, means it has something to do with the rendering to the screen.  It has not been reported very often. but when people do tell us about it (a few cases came up before we had the forums) the behavior is consistent with what you report.

 

I don't have a workaround for it.  At this point, we don't think we have a way to fix it (especially since the developers can't see it on any machines they have access to).  I wish I had a better answer.

 

Carla

Link to comment
Share on other sites

Thanks for the reply. It's just weird that the program worked on his main computer before, but no longer doesn't. We'll take a look at his install history, to see if we can spot any potential problem programs that were installed after RTIViewer.

Link to comment
Share on other sites

The developer thinks it may be an issue with the OpenGL libraries on the computer.

 

I have a way to test this, but I'll need to send you a piece of software.  Can either of you who reported this email me outside the forum, and we can set up this test?  please send a note to info@c-h-i.org

 

Carla

Link to comment
Share on other sites

I have that problem in OSX Lion, but sometimes the image appears by "moving" the lightball a little bit, or at worst by closing and relaunching the viewer. It happends me around 50% of times I open an file. Nothing to worry a lot in my case, like the lack "continuity" when changing light angle (must actually click in other part of the ball in order to get the right light angle, by dragging the cursor is not enough).

 

So if sometime you want me to test a new version, count me on.

Link to comment
Share on other sites

Just to follow up on the situation.  Byl and Leszekp contacted me and ran the needed script.  That told us exactly what openGL libraries were on the computers exhibiting the problem.

 

The issue on Windows is that the RTIViewer, both the existing and the new version, require OpenGL libraries of 2.1 or higher.  These specs have been out since 2003.  So, if you see this problem it could be due to an old computer, and you can update the drivers for your graphics card OR it could be that your computer has an integrated Intel GPU.  Intel has chosen not to stay up to date with the OpenGL version for these integrated GPU systems, so computers as late as 2010 with the Intel integrated GPU might exhibit this problem. 

 

These are the wiki pages where you can check the GPUs that have the OpenGL 2.1 (or later) support:
http://en.wikipedia.org/wiki/Comparison_of_Nvidia_graphics_processing_units
http://en.wikipedia.org/wiki/Comparison_of_Intel_graphics_processing_units
http://en.wikipedia.org/wiki/Comparison_of_AMD_graphics_processing_units

 

So the mystery is solved, and for some people updating drivers will solve this.  I'm not sure what the solution is for folks with the integrated Intel GPUs (and as my computer graphics pals point out, isn't actually a GPU).

 

As for mjm's report of seeing an intermittent issue on the Mac, with Lion, that appears to be a very different issue with different behavior.

 

The issue with the system not keeping up with moving your mouse is discussed in another post here: http://forums.culturalheritageimaging.org/index.php?/topic/237-rti-viewer-trackball/

It is actually a feature - read all about it by following the link.

 

I've never seen or heard of the red rectangle problem (where the entire rendering area consists of a large red filled in rectangle) occurring in a Mac environment.  The fact that it is intermittent indicates a different issue.

 

Carla

Link to comment
Share on other sites

For me, somebody with not really experience in coding, the red screen it seemed to be an issue related to the lack of memory, maybe similar in that way to the feature of the trackball and the continous changing...

 

Thanks a lot for your answers.

Link to comment
Share on other sites

We're going to try updating graphics drivers for these systems again; first attempts failed, as Windows said that they were already up to date. But if it's the Intel integrated graphics that's the problem, this may be a futile endeavour, as both systems use that. And they're both laptops, so they can't be upgraded.

Link to comment
Share on other sites

  • 1 month later...

Just a brief update. Intel states on the following page that drivers for their graphics chipsets obtained directly from Windows Update or the OEM may not support OpenGL:

 

 

https://downloadcenter.intel.com/Detail_Desc.aspx?DwnldID=20046

 

And suggests downloading drivers for these chipsets directly from Intel for full OpenGL support. They also suggest going into the control panel to check the settings:

 

http://www.intel.com/support/graphics/sb/CS-030506.htm

 

We tried that with the problematical laptop we've been strugglng with (which uses a 945 Intel graphics chipset), but it didn't solve the problem. However, it might work for other Intel chipsets, or for other computers with the 945 chipset. Worth a shot, in any case.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...