the distance checks could get into a feedback loop so that the result depended on the order of verts/edges.
now you can randomize vert/edge/faces and get exactly the same results.
also made some internal improvements,
- used fixed sized arrays (no need to realloc).
- use vertex tag flags rather then a visit-hash.
- remove 'tots' array that did nothing (not sure why it was added).
* Enable the Non-Progressive integrator on GPU (CUDA) for testing.
In order to compile the CUDA kernel with it, you need at least 6GB of system memory and CUDA Toolkit 5.0 or 5.5.
It should also work with CUDA Toolkit 4.2, but in this case you should have 12GB of RAM.
In case any problems arise, just change line 65 of kernel_types.h to disable Non-Progressive again.
-- #define __NON_PROGRESSIVE__
++ //#define __NON_PROGRESSIVE__
whether preview or render settings need to be used in curves
and mballs tessellation.
This is not noticeable for artists yet, but this is silly
to use G.is_rendering anyway and using argument will help
making rendering and viewport working simultaneously in
the future.
so special case in BKE_displist_make_mball is not needed anymore.
The same could be done in trunk as well, but rather run some
more intensive tests first.
Moved all the static variables into a PROCESS
structure which is now passing all over to where
static variables used to be used.
There's still one static variable which is
cubetable, but it's being initialized once
and then used read-only. Maybe we'll need
to move cubetable initialization to blender
startup, but that could wait a bit for now.
For users it means BI rendered viewport
wouldn't crash when using metaballs.
--
svn merge -r57515:57516 ^/branches/soc-2013-depsgraph_mt
Moved all the static variables into a PROCESS
structure which is now passing all over to where
static variables used to be used.
There's still one static variable which is
cubetable, but it's being initialized once
and then used read-only. Maybe we'll need
to move cubetable initialization to blender
startup, but that could wait a bit for now.
For users it means BI rendered viewport
wouldn't crash when using metaballs.
As OSL headers mix #include <OpenEXR/foo.h> and #include <foo.h>, we need both /path/to/openexr/include and /path/to/openexr/include/OpenEXR include paths... :/
Issue investigated by doug65536 on IRC, thanks.
* Replaced the Brute Force version with a nice lookup table, this speeds it up a lot.
Patch by Philipp Oeser (lichtwerk) with some cleanup and changes by myself. Thanks!
ToDo:
* Temperature values between 800 and 804 Kelvin are wrong in SVM, check on this.
Simply recompute texture context on every redraw, like already done for sbuts->texuser (also called texture context :/ ).
Also allows simplification, buttons_check_texture_context can be merged back into set_texture_context...
Adding only the .svg file, the PNG can make once the multiview branch is merged.
I wanted to commit this to avoid having conflicts in the .svg file if
someone works on new icons too.
The new icon is at X-23
Now also check previous button context, when switching. This way, if the previous one was a texture one, and become valid after the last switch (e.g. a material added to a new object), relevant texture context can still be set.
Note this commit also partially fixes [#35769] The “show texture in texture tab” button in a modifier does not show up until the user manually shows the texture in the Texture tab (full fix of this one will probably implies to always have a valid ButsContextTexture [sbuts->texuser]...).
* First (brute force) implementation for SVM. This works and delivers the same result as OSL, but it's slow.
* Code inside svm_blackbody.h inspired by a patch by Philipp Oeser (#35698), thanks.
Ideas:
* Use a lookup table to perform the calculations on render/ level.
* Implement it as a RNA property only, and do the calculation like Sun/Sky precompute.
draw mode open. OpenGL texture free needs to happen in the main thread, but it was
freeing a copy of the image datablock. I can't understand how this code ever worked,
probably it never did.
