Concerning your request for a talk, I must respectfully decline - lets stick to written communication here on phabricator.

Missunderstood your comment then. Please think about radiologist's versus neurologist's view: rendered slices are twosided and twodimensional, so inverting an axis in a slice means changing the point of view, not changing handedness. Yes, thats an excellent source of confusion.

thank you very much for your input.
as you have experienced, documentation can be incomplete or unclear.

cannot reproduce.
awaiting example data file and exact specification of affected version.

https://phabricator.mitk.org/file/data/kdjzwxzsn56duqb64xg3/PHID-FILE-zept6drjkpwlbu3id653/LPS_12x12x12_100010001.nrrd

Yes, misunderstood.
MITK uses LPS in world coordinates, therefore (0,0,0) is located in Right-Front-Bottom corner ( assuming 1. octant, that means all 3 positive half-axes ), of course the patient's right is on the left on the screen, as usual in medical imaging.

According to the documentation, the world geometry origin should be in the bottom-left-back corner.

same as T20070, fixed in master.

Still can reproduce this error-on-closing MitkDiffusion, and it persists over multiple sessions until I restart $> MitkDiffusion --BlueBerry.clean
with --BlueBerry.clean.
To me this issue smells like an off-by-one in the destructor callchain between BlueBerry and Qt. Since this issue seems to only create a coredump _on_closing_ and does not affect the session/usage/functionality of the MitkDiffusion-Workbench, as far as I can see, really low priority... but a vexing problem none the less.

Pushed new branch T20070-fix-cpu-volume-rendering-crash.

It works:

diff --git a/Modules/MapperExt/src/mitkGPUVolumeMapper3D.cpp b/Modules/MapperExt/src/mitkGPUVolumeMapper3D.cpp
index 117e91f..bc7b96b 100644
--- a/Modules/MapperExt/src/mitkGPUVolumeMapper3D.cpp
+++ b/Modules/MapperExt/src/mitkGPUVolumeMapper3D.cpp
@@ -18,6 +18,9 @@ See LICENSE.txt or http://www.mitk.org for details.
 #define GPU_WARN MITK_WARN("mapper.vr")
 #define GPU_ERROR MITK_ERROR("mapper.vr")

According to @kolbch, mitk's volume rendering code will be removed and the rendering backend replaced, soon-ish.
VTK now can do all that stuff, and more.

this is where the fun begins. where should this missing define be located?

Seems to be a VTK problem.

No, FA-scaling is okay. Maybe I messed up using vtkGlyph3D?

Doublicate of T19632, working on it, seems to be a nasty problem.

My branch "bug-19632-FindInvertedAxis" is useless, because I had the stupid idea that this problem would be easily solved by correct math.

similar or same as T19513

Same appearance, but still functional in Mac OS 10 elCapitan, unlike ubuntu 14 where one cannot click anything anymore as long as rotation symbol is on the mouse cursor.

The question is HOW:
*change rendering somehow,
*change index2world-matrix,
*change window titles to XYZ instead of AxialCoronalSagittal,
*do a global reinit instead,
*just inform the user, or
*forgettaboutit?

backtrace

Possibly dublicate of T19165, futher investigation needed.

backtrace

CLI output

[fe3b7b]: Merge branch 'bug-19607-ControlVisualizationPropertiesView-m_MultiWidg

Merged commits:

2016-03-14 14:51:34 Martin Hettich [682c55]
increase robustness and sanity.

The main issue was

screenshot

screenshot

screenshot

screenshot

screenshot

screenshot

[ed8990]: Merge branch 'bug-19484-integra-FiberFoxFixArtefactsFlag'

Merged commits:

2016-06-08 21:05:27 Martin Hettich [7bec47]
COMP: fix merge conflict frak-up.

[580ab8]: Merge branch 'bug-19484-FiberFox-FixArtefactsFlag'

Merged commits:

2016-02-20 19:53:48 Martin Hettich [6fa004]
add checks of AddNoise boolean flag and do some cleanup of indentation and add missing braces.

[369b8c]: COMP: Merge branch 'bug-19484-FiberFox-FixArtefactsFlag', resolve over

Merged commits:

2016-02-20 19:53:48 Martin Hettich [119e6b]
add checks of AddNoise boolean flag and do some cleanup of indentation and add missing braces.

[b94ec8]: Merge branch 'bug-19484-integra-FiberFoxFixArtefactsFlag'

Merged commits:

2016-06-09 00:24:20 Martin Hettich [ec9b74]
COMP: fix missmatched braces after merge-conflicts.

[224b92]: Merge branch 'bug-19481-EmptyMotionVolumesTag'

Merged commits:

2016-03-16 21:09:40 Martin Hettich [882096]
fix T19841: crash on empty tag.

[41ab7f]: Merge branch 'bug-19474-ParseGradientsBeforeAddingMotion_FiberfoxParam

Merged commits:

2016-02-29 16:36:35 Martin Hettich [14c879]
parse the gradients before adding the motion, so that the image information is complete at that moment.

[24edfb]: Merge branch 'bug-19475-integra-FiberFoxMiniAppOutputName'

Merged commits:

2016-02-13 17:53:56 Martin Hettich [de3aed]
fix output filepath and filename handling in MitkFiberfox MiniApp and itkTractsToDWIImageFilter Module.

[c37e05]: Merge branch 'bug-19381-add-comments-to-mitkPlaneGeometry-initializeSt

Merged commits:

2015-10-26 15:57:12 Martin Hettich [152846]
correct indentation and spelling.

May be related to bugs #11477,#18662, trouble with reinit vs. global reinit etc..

Missing documentation and comments in mitkplanegeometry.cpp and unclear code.

New bug for this particular function:
http://bugs.mitk.org/show_bug.cgi?id=19582

Invalid since 18167 was reverted.

actual master (1aac94dc16abbc1ff5b54a4cafd61afd788bbfa5)
works visibly, but gives many gl-Errors, and sometimes a 'soft' crash
(see screenshot)

In Developer Mode -> DataManager -> stdmulti.widget[13] -> details => negative Normal vector ( who, how, what, why, wtf does this ? ), while stdmulti.widget2 => positive Normal vector. This matches this bug.

segfaults at AddContour, both branches and my own stashed branch - which sucks, because it used to work fine on those branches/commits last week...

[74572d]: Merge branch 'bug-19237-implementExclusionListFiberFoxMotionGeneration

Merged commits:

2016-06-20 13:09:43 Martin Hettich [6bce31]
minor syntax error corrections.

[398dd6]: Merge branch 'bug-19237-implementExclusionListFiberFoxMotionGeneration

Merged commits:

2016-07-01 21:08:56 Martin Hettich [38f485]
also implement #19237: list of negative numbers meaning exclusion in
QmitkFiberfoxView plugin and amend mouse-over help text.

[c34ce0]: Merge branch 'bug-19228-implement-ellipsoid-glyphs-for-diffusion-tenso

Merged commits:

2015-08-28 14:22:58 Martin Hettich [d1602a]
COMP: typo corrected, s/abs()/fabs()/2:mitkOdfVtkMapper2D.txx.

[41358c]: COMP: Merge branch 'bug-19228-implement-ellipsoid-glyphs-for-diffusion

Merged commits:

2015-08-28 15:05:12 Martin Hettich [ff2ea6]
COMP: <cmath> vs. MSVC, and Minor Bug: GUI buttons reenabled.

[298e52]: COMP: Merge branch 'bug-19228-implement-ellipsoid-glyphs-for-diffusion

Merged commits:

2015-08-28 18:52:29 Martin Hettich [6e80b9]
Merge remote-tracking branch 'origin/bug-19228-implement-ellipsoid-glyphs-for-diffusion-tensors' into bug-19228-implement-ellipsoid-glyphs-for-diffusion-tensors

[9c086c]: Merge branch 'bug-19228-implement-ellipsoid-glyphs-for-diffusion-tenso

Merged commits:

2015-08-11 17:49:00 Martin Hettich [6ff148]
Ellipsoid Glyphs for dti implemented, minor bugfix, comment cleanup.

[e9eaa2]: Merge branch 'bug-19219-typoInDocu'

Merged commits:

2016-03-16 14:56:18 Martin Hettich [4145fc]
correct minor typo in docu.

[840bec]: Merge branch 'bug-19165-DiffusionWorkbenchCrashes'

Merged commits:

2016-04-27 20:56:05 Martin Hettich [0e4bde]
add checks for nullptr after each dynamic_cast.

bug-19133-old-docu-links-fixed

git diff of my bugfix

Modules/QtWidgetsExt/src/QmitkAboutDialogGUI.ui
Modules/DiffusionImaging/MiniApps/Registration.cpp
Modules/Core/resource/Interactions/Legacy/DisplayConfigMITKTools.xml
Examples/BlueBerryExampleLauncher/Configurations/CustomViewer.cmake
Examples/BlueBerryExampleLauncher/Configurations/ExtensionPointDefinition.cmake
Examples/BlueBerryExampleLauncher/Configurations/MinimalApplication.cmake
Examples/BlueBerryExampleLauncher/Configurations/MultiplePerspectives.cmake
Examples/BlueBerryExampleLauncher/Configurations/SelectionServiceMITK.cmake
Examples/BlueBerryExampleLauncher/Configurations/SelectionServiceQT.cmake
Plugins/org.mitk.gui.qt.extapplication/documentation/UserManual/QmitkMITKWorkbenchUserManual.dox
Documentation/Doxygen/3-DeveloperManual/Concepts/Persistence.dox

git diff of my bugfix

(In reply to mdhettich from comment #2)

Created attachment 2092 [details]
git diff of my bugfix

bug-19133-old-docu-links-fixed

crazy, it works when there is a positive geometry node present in DataManager

http://www.itk.org/Doxygen/html/namespaceitk_1_1SpatialOrientation.html

Seems sometimes those glyph are visible after all, but not placed correctly.