Blender Git Commits

Blender Git "render25" branch commits.

Page: 16 / 29

April 19, 2010, 13:30 (GMT)
April 19, 2010, 10:01 (GMT)
April 19, 2010, 09:50 (GMT)
Render Branch: Fix missing AO pass with SSS materials, forgot to check for
this case in optimizations.

April 18, 2010, 10:49 (GMT)
April 16, 2010, 15:25 (GMT)
Render Branch: Fix some alpha premul problems with mipmap files by moving
the premul code to imbuf, still a small difference that needs to be fixed,
but no very bright pixels anymore hopefully.

April 16, 2010, 15:24 (GMT)
April 16, 2010, 15:03 (GMT)
Render Branch: Attemp to fix sss + irr cache render crash with displaying
cache samples.

April 16, 2010, 14:27 (GMT)
Render Branch:
* Fix crash with NULL image texture.
* Add test for render cancel in deep shadow buffer compression.

April 16, 2010, 14:09 (GMT)
Render Branch: SSS now works together with irradiance caching.

April 16, 2010, 12:00 (GMT)
Render Branch: Fix mistake in imbuf changes, forgot to replace one free()
by MEM_freeN().

April 16, 2010, 11:19 (GMT)
Render Branch: Fix merge full sample not working after cancelling render,
proper fix would be to get rid of G.afbreek entirely..

April 16, 2010, 10:47 (GMT)
Render Branch: Fix for broken raytracing neighbour intersection test.

April 16, 2010, 08:35 (GMT)
April 15, 2010, 21:33 (GMT)
fix for missing null pointer check.
April 15, 2010, 17:10 (GMT)
April 15, 2010, 16:59 (GMT)
Render Branch:
* Forgot to mention this in last commit. Currently image loading will look
for an .tx file next to the specified image file, and if it exists, use
that instead automatically. It compares the dates and always used the newest
of the two.
* Fix uv y derivative being scaled wrong.
* Fix for possible use of invalid mipmap level -1.

April 15, 2010, 15:49 (GMT)
Render Branch:

Imbuf module: some small refactoring and removing a lot of unused or old code
(about 6.5k lines), it's a bit drastic but can add back if it's still useful.

* Added a ImFileType struct with callbacks to make adding an file format type,
or making changes to the API easier.
* Move imbuf init/exit code into IMB_init()/IMB_exit() functions.
* Increased mipmap levels from 10 to 20, you run into this limit already with
a 2k image.
* Removed hamx, amiga, anim5 format support.
* Removed colormap saving, only simple colormap code now for reading tga.
* Removed gen_dynlibtiff.py, editing this is almost as much work as just
editing the code directly.
* Functions removed that were only used for sequencer plugin API:
IMB_anim_nextpic, IMB_clever_double, IMB_antialias, IMB_gamwarp,
IMB_scalefieldImBuf, IMB_scalefastfieldImBuf, IMB_onethird, IMB_halflace,
IMB_dit0, IMB_dit2, IMB_cspace

Image textures can now lazely load mipmap levels from files, so if an object
is far away only a lower level is loaded. This is supported for the .tx file
format as used by prman, which is basically a tiff image with all the mipmap
levels saved in it. I tested generating them using OpenImageIO maketx utility,
(there is no way to save these from Blender currently).

Added mipmap levels simplify option (couldn't think of a better name). This
makes it possible to force using lower resolution mipmap levels, which helps
reducing memory usage a lot when using .tx files.

Next: tile loading support, then swapping them in and out with a cache.

April 15, 2010, 15:36 (GMT)
Render Branch: three utility functions:

* mul_v4_fl
* BLI_replace_extension: replace extension in a filename
* BLI_file_newer: test if one file is newer than the other

April 15, 2010, 13:51 (GMT)
disable an assert, use a guess assignment instead.
April 15, 2010, 11:04 (GMT)
Render Branch: fix for particle duplis not rendering from inside a dupligroup,
not sure why this check was here, will need to find out if something goes wrong.

By: Miika HämäläinenLast update: Nov-07-2014 14:18MiikaHweb | 2003-2021