OOF: Finite Element Analysis of Microstructures

Known bugs in PPM2OOF

Please send bug reports to [email protected]. Let us know what version of the program you're using, and on what kind of computer you're running it. A log file is always helpful. You can save log files with the "log/save" menu item.

Bugs that were fixed as soon as they were discovered may not be listed here. This page lists bugs that have been outstanding for a while. If you think you have found a bug, and you're not using the latest version of PPM2OOF, check the Recent Changes file to see if the bug has been fixed in a newer version.

1.1.28
(and earlier?)
  • If some mesh triangles don't have an assigned material, the mesh can't be saved, but ppm2oof doesn't warn you about it.
  • If some mesh triangles don't have an assigned material, the program will crash when saving a configuration file. To avoid this, make sure that a material is explicitly assigned to each pixel.

1.1.8
and earlier
  • Loading a new image with the "ppmfile" command sometimes screws up the interpretation of mouse clicks in open graphics windows. The problem occurs only if the new image is a different size than the old image, but the last click on the old window and the first click on the new window both occured on displayed images of the same size (eg, both images had been scaled to fill the window). A workaround is to close the graphics window and open a new one with the "graphics/open" command. FIXED in 1.1.9.

1.1.7
and earlier
  • A triangle's material sometimes isn't updated when its nodes are moved, leading to wrong material assignments. FIXED in 1.1.8.

1.1.6
and earlier
  • When using the "empty" material type and adaptive meshes, elements that should be empty at the border of the empty region are sometimes given the wrong material type. FIXED in 1.1.7.

1.1.4, 1.1.5
  • ppm2oof fails to read configuration files, due to a silly bug introduced in 1.1.4. FIXED in 1.1.6.

1.1.3
and earlier
  • Deleting a pixel group can sometimes lead to a core dump or a corrupt .goof file. FIXED in 1.1.4.

1.1 beta 30
  • Writing a configuration file with "adaptive_mesh/stack/stacksize=1" leads to an infinite loop and a very large file. Change stacksize to 2 to avoid the problem. FIXED in 1.1 beta 31.

1.1 beta 23
and earlier
  • Burning a very large area in a very large image can cause a segmentation fault on some systems. FIXED in 1.1 beta 24.

1.1 beta 19
  • If a new image with the same size as the current image is loaded with the "ppmfile" command, the new image isn't displayed immediately. It can be displayed from the Image Gallery dashboard, though. FIXED in 1.1 beta 20.

1.1 beta 17
  • The "quit" command in the main menu will make PPM2OOF dump core when certain dashboards are open. The core dump is annoying but harmless. The "Quit" button at the top of the menu does not have this problem. FIXED in 1.1 beta 18.

1.1 beta 14
  • The program uses a lot of extra memory upon repeated application of "adaptive_mesh refine all". This problem only appears in the SGI and SUN versions (I think) and has been FIXED (I think) in version 1.1 beta 15 for the SGI only.
  • PPM2OOF will crash if you set "adaptive_mesh stack stacksize" before creating a mesh. FIXED in version 1.1 beta 15
  • Setting adaptive_mesh stack stacksize=1 doesn't prevent PPM2OOF from making a copy of the mesh each time you modify it, it just prevents you from reverting to the copy. This is a memory leak. FIXED in version 1.1 beta 15.

Send comments to the OOF Team.