[SVN revision 21518]
ENH (#3182): thick slices as described in feature request

fixed in commit #20711

memory requirements has been already addressed in the 3m3 volume visualization bug, and the gpu mapper requires no additional memory now

[SVN revision 20450]
ENH (#3068): more intelligent "auto-categorization"

[SVN revision 20454]
COMP (#3068): and again mbits/public

[SVN revision 20473]
ENH (#3068): more keywords

[SVN revision 20711]
FIX (#3068): fixed 64bit-warnings & signed/unsigned bug

[SVN revision 20475]
ENH (#3068): other keywords

[SVN revision 20451]
FIX (#3068): fix some blueberry logging

[SVN revision 20452]
FIX (#3068): fix some std::cout

[SVN revision 20456]
ENH (#3068): better autocategories

[SVN revision 20453]
COMP (#3068): trying rerun test

[SVN revision 20449]
ENH (#3068): windows console logger now "auto-categorizes" log messages

[SVN revision 20443]
FIX (#3068): deactivated debug messages / categorized tools

[SVN revision 20439]
FIX (#3068): showing now categories in console logging output / "openCherry->BlueBerry"

[SVN revision 20445]
ENH (#3068): added colored console log output for windows

[SVN revision 20448]
ENH (#3068): reverted std::cout recoloring as it remained even after quitting mitk

[SVN revision 20446]
ENH (#3068): normal std::cout messages are now shown in purple to distinct them from regular log messages

[SVN revision 20447]
ENH (#3068): the console window under win32 has now a title "MITK - Log"...

[SVN revision 20357]
FIX (#2910): new histograms for transferfunctions widget

implemented own version for histograms
(currently support only signed short)

volume rendering requires that much memory

for 3m3 there is no possibility to reduce further memory requirements.

[SVN revision 20119]
FIX (#2743): fixed global reinit to ignore widget-planes

seems, that old mitk behaviour is already enabled in current source tree and this "bug" already fixed

the original "global reinit" function should be split into two separate methods:

reopened, there was a small glitch at that existing global-reinit after load, it did not ignore the 2d-widget-planes and so never became smaller.
following commit fixes this

[SVN revision 20438]
FIX (#2733): removed debug messages

[SVN revision 20765]
ENH (#2733): reduced mapper memory requirements /

[SVN revision 20540]
ENH (#2733): enabled at least rendering of 3dt images/though no histogramm, GPUVolumeMapper3D has now better parameters and more properties

[SVN revision 20819]
ENH (#2733): initial support for RGBA volumes

[SVN revision 20407]
FIX (#2733): fixed memory leaks and reduced memory requirements when switching between gpu/cpu by de/initializing gpu/cpu mappers

[SVN revision 20412]
FIX (#2733): selection now updated immediately after creating view

[SVN revision 20358]
ENH (#2733): gpu volume mapper now adds his new properties @ SetDefaultProperties()

[SVN revision 19805]
ENH (#2733): increased quality of GPUVolumeMapper3D

[SVN revision 19765]
ENH (#2733): lod now again useable with GPUVolumeMapper3D / toggleable with "volumerendering.uselod" bool property

[SVN revision 19754]
ENH (#2733): removed more space in ui

[SVN revision 19767]
ENH (#2733): opacity/grayvalue widgets need less space

[SVN revision 19775]
ENH (#2733): crosswidget now locks mouse position / added first tooltip

[SVN revision 19799]
ENH (#2733): increased quality of GPUVolumeMapper3D

[SVN revision 19751]
ENH (#2733): removed imageselector / using blueberry selection service / removed space-costly groupboxes

[SVN revision 19774]
ENH (#2733): levelwindow and bell generate smoother transferfunction

[SVN revision 19665]
ENH (#2733): commit in detail in previous bug comment

[SVN revision 19668]
COMP (#2733): added missing file

[SVN revision 19677]
ENH (#2733): automatic generation of transferfunction for binary images

[SVN revision 20424]
FIX (#2727): brighter gradient background

[SVN revision 20408]
FIX (#2727): dark gradient background for better results with volumerendering

[SVN revision 20542]
ENH (#2727): better CT Generic default TransferFunction

[SVN revision 19766]
COMP (#2727): removed debug message

Changerequest under
http://mitk.org/wiki/ChangeRequests/2727

[SVN revision 19764]
CHG (#2727): reverted renderingmanager use 2 lod levels instead of 1 (means no), removed legacy localizer code

[SVN revision 20136]
FIX (#2727): fixed legacy VolumeDataVtkMapper3D to not use lod level 2

[SVN revision 19662]
CHG (#2727): explained in detail in changerequest 2727

Fixing this consists of two things:

One line of code was mistakenly removed and did not add the default shader properties. Is another core request needed for this?

[SVN revision 21412]
FIX (#2709): updated surfaceVtkMapper3D doxygen documentation to match the new property scheme for materials

[SVN revision 21408]
COMP (#2709): false commit

[SVN revision 21407]
COMP (#2709): false commit

[SVN revision 21405]
FIX (#2709): removal of MaterialProperty

[SVN revision 21406]
COMP (#2709): false commit

has to be still done,

[SVN revision 20610]
FIX (#2681): added missing */

this problem is not related to the surface material editor himself.
its caused trough the global reinit, which doesn't affect only the stdmultiwidget,
also the preview render window of the surface material editor.
we need to have multiple renderwindow manager instances to divide the preview window from the stdmultiwidget (see #3342)

[SVN revision 21566]
FIX (#2638): fixed as in the proposed solution

this bug is no longer feasable to finish for the MITK 1.0 release.

MacOS support for GPGPU is not 1.0 relevant,
cause this feature is barely used on any platform,
(except for an internal filter)
and we have a new OpenCL framework for advanced GPGPU programming.

[SVN revision 19191]
ADD (#2528): ADD (#2528): File backend for mitk / mbi logging

[SVN revision 19204]
CHG (#2462): reduced further memory requirements of volume mapper

After "researching" what is claimed to be too slow:
(i.e. when MITK is used in Radiology)

Slicing an Image with Wheel in a 2D Window appears to be slow when also an
segmentation (a Surface) is loaded/generated

mitkGPGPU printfs are replaced, closing bug

merge with trunk on commit #19064

[SVN revision 20444]
FIX (#2292): cleaned up logmessages

due to the sharing problem of opengl renderwindows, we had need to enable global immediate mode rendering for all vtk rendering.

[SVN revision 19740]
FIX (#2292): moved material editor to right place in sourcetree

merge with trunk on commit #19064

[SVN revision 19112]
FIX (#2292): fixed material editor exception, when no surface is available

[SVN revision 19152]
FIX (#2292): replaced printf by LOG_INFO

GLSL support seems stable, closing bug

bug was fixed in another commit, gl code was unneccesary

changing layout works now in windows and linux, by using a container widget encapsulating the renderWindow widget.
widget menu graphical artifacts fixed.

(In reply to Jasmin Metzger from comment #5)

Markus, can you have a look at this bug (it's a very old one :-) )?

outdated, volume rendering was in many parts rewritten, and this bug was fixed with it.

[SVN revision 21422]
ENH (#1717): imagemapper2d now uses the new iil4mitkImage::RGBA and pixeltype supports now itk::RGBAPixel

commiting solution 1:

[SVN revision 21348]
ENH (#1717): updated IIL4MITK to support RGBA

[SVN revision 25860]
FIX (#1692): added color lookuptable

(NEXT COMMIT) Added possibility to color points with individual colors. It can be used with the new property "colorLookupTable". This mitk::LookupTable associates the point index to point color.

This may be related to your platform project, Christoph?