paint. Was own regression when optimizing colour operations. I will not
use an alpha bit mask since it may run into portability issues with byte
order.
Zooming in on images in Image window now shows pixels again (was filtered).
Now the glaDrawPixelsTex() and glaDrawPixelsAuto() have an argument to
define if images should zoom in with linear filter, or draw pixels.
Image Editor and 3D view background image now use new automatic switching for
drawing GPU texture or OpenGL DrawPixels too. For large zoomed images
it gives massive speedup.
Automatic switching for drawing pixel buffers via glDrawPixels or using GPU textures
It works with a User Preference limit, in megapixels, to define whether to use
GPU or direct pixel drawing. Default is now initialized to 10 MP (4k buffers).
Especially for zooming out (draw smaller) texture drawing is much smaller. Also
Nvidia cards typically draw much faster with textures in general.
Added to node backdrop first now, the other editors follow in a next commit.
For coders: added new DNA function to initialize new struct variables, so you
don't have to sub-version files anymore.
DNA_struct_elem_find(fd->filesdna, "structname", "typename", "varname")
"filesdna" is the sdna description of the current file being versioned.
too nice but it frees the Ctrl-LClick shortcut and allows us to set
invert mode for other paint tools, such as sharpen vs blur or invert
colour for draw brush. This conflict has existed on GSOC branch, better
resolve now before merging invert functionality.
- without python builds without warnings.
- replace MAXFLOAT -> FLT_MAX in some areas, MAXFLOAT overflows (lager then float range).
- add cmake option WITH_GCC_MUDFLAP to enable libmudflap use.
Fixes [#34348] Collapse short edges affects masked geometry
/projects.blender.org/tracker/?func=detail&aid=34348&group_id=9&atid=498
Fixed by disallowing topology updates on edges with a vertex masked at
50% or greater. This is a necessarily arbitrary choice; can't do
halfway updates for topology like we can with displacement.
* Click-select operators don't have "register" flag anymore
* Graph Editor click select can be undone, just like everything else
* "Current frame" -> "Current Frame" for consistency with rest of menu
intended
Finally, I've figured out a way to get the ID Selector Template to only show the
icon for the type of ID block (i.e. the intended design), instead of icon+text.
This improves the layout a bit, as previously the type selector and ID-block
pointer and label would all be the same sizes (roughly 1/3rd of the overall
width each), which meant that there wasn't much room left to actually see which
datablock had been selected.
Additionally, I've fixed a few more layout types which still didn't correctly
propagate the red-alert flags.
- assert if BLI_buffer_at() is called with an out of bounds value.
- add BLI_buffer_resize_data() macro which resizes and returns a pointer to the new array.
- warn if missing call to BLI_buffer_free().
When running in non-interactive mode, dont add a area draw callback or adjust the cursor, use a fixed snap value and don't use opengl lookups on faces.
Minor optimization - compare squared distances where possible.
brushes, due to issues with color coded drawing or slow/buggy reading from such
a buffer on some systems.
In case multisample is enabled now, it uses an offscreen buffer for such drawing,
which is not multisampled and so should not cause issues. This does mean there is
some extra GPU memory usage when multisample is enabled, and we could optimize
triple buffer to work together here somehow to share buffers, but it's better than
having selection not working.
- use more accurate method to find if a face is inside projected lines.
- check faces intersect while expanding selection - in some cases this could result in the entire model being selected.
Handle sequences in a special case for dealing with
sequence sources.
Namely handle separate frames in separate threads,
but do disk read from a critical section since HDD
is not so friendly with lots threads requesting for
data from it.
Makes proxy building much faster than it was before.
This commit implements multi-threaded calculation of frames
when building proxies. Both scaling and undistortion steps
are now threaded.
Frames and proxy resolution are still handled one-by-one,
saving files after every single step. So if HDD is not so
fast, this commit could have not so much benefit.
Internal changes:
- Added IMB_scaleImBuf_threaded which scales given image
buffer in multiple threads and uses bilinear filtering.
- libmv's camera intrinsics now have SetThreads() method
which is used to specify how many OpenMP threads to use
for buffer distortion/undistortion.
And yeah, this code is using OpenMP for threading.
- Reshuffled a bit libmv-capi calls and added function
BKE_tracking_distortion_set_threads to specify number
of threads used by intrinscis.
