Blender Git Loki

Kaikki Blender Git kommitit.

Page: 5029 / 8462

February 26, 2015, 10:46 (GMT)
Grmll, sorry for the noise, fix stupi cpoypaste typo in previous commit.
February 26, 2015, 10:42 (GMT)
Attempt to fix BGE build on windows.

Caused by rBb5b359b48f7f35a79b3eec, better to try a full build before pushing that type of commit. :/
February 26, 2015, 10:27 (GMT)
Fix T43817: issue with non-utf8 encodings in Windows paths, again...

We need to use 'W' widechar variants of win funcs and convert wchar to utf8-encoded bytes
in those cases, sigh...

Note: theoritical fix only, need org reporter to test it...
February 26, 2015, 10:27 (GMT)
Fix potential buffer overflow in `BLI_strncpy_wchar_as_utf8()`.
February 26, 2015, 09:30 (GMT)
Cleanup: Simplify camera sample motion blur code.
February 26, 2015, 09:26 (GMT)
Merge branch 'master' into depsgraph_refactor
February 26, 2015, 08:43 (GMT)
Replaced point cache in cloth modifiers (also hair) with the new caching
system.
February 26, 2015, 07:48 (GMT)
Fix T43643: Solidify crashes with 'only-rim'

When verts were shared by 2+ face-islands, the number of allocated edges was wrong.

Cleanup the logic for new verts/edges.
February 26, 2015, 06:49 (GMT)
mesh_validate: don't add invalid edges into the hash
February 26, 2015, 04:53 (GMT)
RNA: correct limits
February 26, 2015, 03:50 (GMT)
use BKE_mball_tessellate.h include

also remove unused includes
February 26, 2015, 03:48 (GMT)
BKE_mball: split tessellation into its own file

this has a lot of its own local structs, functions,
better to keep isolated from general metaball selection/library logic.
February 25, 2015, 22:30 (GMT)
Change samples variable for SSAO - now it is actual samples, not squared
- spiral scheme is much more intuitive here. Also use float number for
spirals to avoid having samples on the same radial direction when sample
number is divisible by spiral number.
February 25, 2015, 22:10 (GMT)
Use spiral mapping for SSAO - it reduces banding a lot, especially in
higher sample counts. Probably a blurring pass might be a good addition
here as well.
February 25, 2015, 21:43 (GMT)
possible fix for loading subobject group objects with textures
February 25, 2015, 21:05 (GMT)
SSAO: sampling coordinates go out of screen, reject the sample

Previous behaviour would get occlusion at borders which could create
over occlusion at edges of the screen.
February 25, 2015, 19:42 (GMT)
Depsgraph: Code cleanup, comments
February 25, 2015, 19:39 (GMT)
Depsgraph: Fix update not happening correct in certain circumstances

Because of the lack of real incremental updates it was possible that incorrect
object matrix was used by the constraint stack.

Now flush update will tag all operations within component for update, solving
the issue with lack of incremental updates.

It's not totally optimal implementation atm, but good enough for further tests
and improvements.
February 25, 2015, 19:34 (GMT)
Sequencer: Implement 'Use Multi-View' for sequencer strips as well.

Note: Sequencer still needs the same cleanup we did for Image/Movie
datablocks regarding the load functions.

The Sequencer code flow is very differnt than regular image I/O though. So
I'm leaving the sequencer cleanup for after I get the other suggested
changes implemented in the Image code (i.e., more separation be image
and Scene/RenderData).
February 25, 2015, 19:32 (GMT)
From review: Add Use Multi-View UI option

New flag to determine if the image is or not a potential multiview
The flag is to be used in the UI while IMA_IS_MULTIIEW and IMA_IS_STEREO
are used internally when the image actually is a Multi-View image.

It is working now for Image Editor, Compositor, and even Image textures
(though the latter may be misleading since the rendering always take the
1st ibuf when it comes to textures - not that we can't shouldn't
implement that later).
Tehnyt: Miika HämäläinenViimeksi päivitetty: 07.11.2014 14:18MiikaH:n Sivut a.k.a. MiikaHweb | 2003-2021