Sergey Sharybin
5eefcb95df
Setting tracking object scale shall not depend on active object
2013-05-09 16:38:50 +00:00
Sergey Sharybin
2e96e41da3
Made bundles in 3D viewport have constant size
...
This means bundles' size is not affected by camera scale.
This way it's more useful to work with -- bundles never
becomes too small or too large (depending on reconstructed
scene scale).
2013-05-09 16:38:47 +00:00
Sergey Sharybin
4f5f97254c
Reconstructed scene scale ambiguity improvement
...
Made it so reconstructed scene always scaled in a way
that variance of camera centers is unity.
This solves "issues" when different keyframes will
give the same reprojection error but will give scenes
with different.scale, which could easily have been
considered as a bad keyframe combination.
This change is essential for automatic keyframe
selection algorithm to work reliable for user.
2013-05-09 16:38:43 +00:00
Thomas Dinges
522eeaa6a0
Cycles / OpenCL:
...
* Remove old comment for sm_13 cards and really check for OpenCL 1.1.
2013-05-09 16:16:41 +00:00
Brecht Van Lommel
168bcfb46b
Cycles OpenCL: fix other build issues when enabling more features.
2013-05-09 15:28:38 +00:00
Thomas Dinges
ba55f2da64
Cycles:
...
* Change some if / else if conditions to switch / case.
2013-05-09 15:20:00 +00:00
Sergey Sharybin
3451baf141
Show dash instead of filename when displaying frame above image sequence length.
2013-05-09 15:04:32 +00:00
Sergey Sharybin
2ad8ec5eaf
Changes to footage information panel
...
- Display additional information about channels
and buffer type (float/byte).
- Don't show frame number beyong sequence length.
- Also fixed issues with footage length calculation,
so it's pronbably will be needed to reload some
of existing footages.
2013-05-09 14:57:20 +00:00
Thomas Dinges
602a7f77c6
Cycles / Filter functions:
...
* Avoid some variable castings here to save some calculations.
2013-05-09 14:37:32 +00:00
Thomas Dinges
13bcf0c104
Cycles:
...
* Remove unused film_response table code.
2013-05-09 14:21:36 +00:00
Thomas Dinges
ce40e7fe27
Cycles / OSL:
...
* Remove custom template from r52885, not needed anymore as OSL 1.3 is a requirement and we do not use a custom branch anymore.
2013-05-09 14:13:43 +00:00
Brecht Van Lommel
d0ffbeec73
Cycles OpenCL: a few fixes to get things compiling after kernel changes,
...
for Apple OpenCL on OS X 10.8 and simple AO render.
Also environment variable CYCLES_OPENCL_TEST can now be set to CPU, GPU,
ACCELERATOR, DEFAULT or ALL values to test particuler devices.
2013-05-09 14:05:40 +00:00
Brecht Van Lommel
d236b4d60f
Cycles bump node: change the Strength value to work better, previously it would
...
give results that were either too weak or too strong, this makes it give more
predictable results. The downside is that it breaks backwards compatibility but
the previous behavior was almost broken.
2013-05-09 14:05:37 +00:00
Campbell Barton
d326d92b2f
fix [ #35280 ] blender crashes when setting Environment Map imagefile
2013-05-09 14:01:42 +00:00
Campbell Barton
675f845116
use standard name for operator properties - 'props'.
2013-05-09 13:05:36 +00:00
Campbell Barton
4c042f2145
bmesh: optimize bmesh_vert_separate, redice allocs (best cast it wont do any allocs).
...
gives approx 16% overall speedup to edgesplit modifier.
also reduce size of smallhash stack, was 521, which got doubled and was quite large on the stack. reduce to 64.
2013-05-09 12:46:35 +00:00
Lukas Toenne
56485b6562
Upgrade for the add_search node operator. This now uses the same basic system as the regular add_node operator, with enum items generated from the common node categories system (nodeitems_utils module). This means that any node listed in the regular node Add menu can now also be added via searching, including node groups and the like. The search operator also uses the subsequent transform to make insertion a bit more streamlined.
2013-05-09 11:43:48 +00:00
Campbell Barton
2e0f741d01
can't use alloca in inline functions (fills up stack - I thought compiler would be smart here).
...
also reserve the exact number of vert/face/edge/loops when creating a bmesh during undo.
2013-05-09 11:42:24 +00:00
Campbell Barton
278240f4b0
bmesh: avoid using BLI_array macros for every face when converting to bmesh, replace with alloca
2013-05-09 10:44:38 +00:00
Campbell Barton
1e784c54cd
bmesh speedup: skip free-realloc while running CustomData_bmesh_merge() when nothing is changed (happens quite often that there is nothing to do).
2013-05-09 10:41:05 +00:00
Lukas Toenne
efcfe88adc
Fix for add_node operator: needs to have SKIP_SAVE set for the settings collection to prevent it from re-using settings from a previous call. This would otherwise lead to python exceptions when those properties don't exist for the other node type (point in case: node_tree only is defined for group nodes).
2013-05-09 10:13:13 +00:00
Thomas Dinges
c1f408c058
Partial revert of own commits r56604 and r56603:
...
* Reverted the changes to code comments, as suggested by Campbell. It makes it more hard to follow.
* Only keep changes to actual UI messages.
2013-05-09 10:03:38 +00:00
Lukas Toenne
3fb67ac16d
Fix for do_versions bug with node groups: The special case of direct input-to-output connections was not handled correctly. In this case both the tonode and fromnode pointers in old node groups are NULL.
2013-05-09 08:29:59 +00:00
Lukas Toenne
89d77c538a
Fix suggested by Campbell Barton: use %r instead of %s when building the python expression for the node operator settings to ensure correct escaping.
2013-05-09 08:05:02 +00:00
Campbell Barton
6f8c29ab0b
fix for 2 errors introduced since release
...
- shape key NULL pointer dereference.
- use uninitialized variable for bmesh free.
also update credits and merge dissolve flag assignment.
2013-05-09 07:02:51 +00:00
Campbell Barton
279a2a1916
fix rna_info, python method to C function wasn't being tested for. (broke changelog generator)
2013-05-09 02:50:59 +00:00
Campbell Barton
8771a9f861
update autogenerated changelog. also correct typos in the generator.
2013-05-09 01:58:49 +00:00
Campbell Barton
3d95873cf6
fix [ #35257 ] Brige > Merge sometimes flips result
2013-05-08 23:14:27 +00:00
Dalai Felinto
102c0d76e9
bugfix: [ #31757 ] setGLSLMaterialSetting has no effect
...
original patch by me, with contribution from HG1 and Florian Völker
bug introduced in rev. 40113
2013-05-08 21:43:35 +00:00
Thomas Dinges
e0edac4952
UI naming consistency:
...
* ShapeKey -> Shape Key. Was called "Shape Key" in most places already.
Pointed out by Dalai, thanks!
2013-05-08 21:41:47 +00:00
Thomas Dinges
b98550590b
UI naming consistency:
...
* DopeSheet -> Dope Sheet. No need to glue the words together.
Only changed comments and UI strings, no functional changes. Request by Dalai Felinto.
2013-05-08 21:05:52 +00:00
Thomas Dinges
01dc4b034f
Cycles / Tile Rendering:
...
* Code refactor of tile ordering to simplify the code and avoid some branching.
* Changed the Center method, so it really follows center -> corners, instead of the BI method, which was confusing sometimes.
2013-05-08 19:49:09 +00:00
Brecht Van Lommel
360a86d658
Attempt to fix OS X build with 10.6 SDK, was not working due to recent fullscreen bug fixes.
2013-05-08 19:37:36 +00:00
Thomas Dinges
872a8ed1bf
Cycles / Hair rendering:
...
* Enable hair rendering on the GPU.
Patch by Stuart Broadfoot, with small tweaks by me, to only enable it on sm_20 and above.
2013-05-08 17:33:25 +00:00
Thomas Dinges
9037a2d3ab
Blender 2.68 release cycle begin:
...
* BCon1, Alpha. Let's have an awesome one!
2013-05-08 17:23:56 +00:00
Lukas Toenne
a3c82d4cde
Updated the custom_nodes.py template script to reflect socket draw API changes in r56584 and r56585.
2013-05-08 16:17:04 +00:00
Lukas Toenne
8deba33497
Fix for node group user count: since node group pointer uses an explicit setter callback it has to do user count increment/decrement manually.
2013-05-08 16:00:08 +00:00
Lukas Toenne
a092611f67
Fix for node item polling: recursion check was the wrong way around, needs to test is the parent tree is inside the group.
2013-05-08 15:41:09 +00:00
Lukas Toenne
d7c74acab1
Added filter to the node group items callback to remove recursive node groups right from the start. These tree pointers would be polled out internally as well, but this way they don't show up in the menus in the first place.
2013-05-08 15:41:07 +00:00
Lukas Toenne
8863222a90
A bit more pythonic way of using the items callback in node categories. The category.items attribute is now a function taking a context parameter instead of a property. This can be used for checking validity, e.g. for doing node group recursion checks, and filter out unusable items.
2013-05-08 15:41:05 +00:00
Lukas Toenne
8542d97f73
2 fixes for node group node_tree pointer property: Make sure the nodeGroupPoll function (which checks for recursion) is used both in the poll callback as well as the actual pointer assignment (set). The poll callback doesn't seem to be used when directly setting the node_tree pointer from the API, so to make sure no dangerous recursion situation can happen this needs a second check.
2013-05-08 15:41:01 +00:00
Lukas Toenne
3234f7e497
Expose the data type property of node group interfaces for easier comparison in python.
2013-05-08 15:40:58 +00:00
Lukas Toenne
3ebe7d970e
Moved a couple of common properties into the NodeAddOperator base class to avoid repetitive code. A new operator node_add_and_link is another variant that first creates a node and them connects a specific socket to an existing one (defined by context pointer).
2013-05-08 15:40:53 +00:00
Lukas Toenne
451a32070c
Removed the unused socket template system from the bpy_types Node base class (it interferes with the input_templates/output_templates functions from C nodes). This can be reimplemented in a nicer way for pynode subclasses later on, but should not be part of the basic Node class.
2013-05-08 15:40:51 +00:00
Lukas Toenne
ee950c9dcb
Exposed the node socket templates of C nodes in RNA by means of a classmethod. This will allow the node link menu to determine feasible socket connections before an actual node instance is created.
2013-05-08 15:40:49 +00:00
Lukas Toenne
701aa8b59b
Removed the add_group_node operator, this is now integrated into the generic node_add operator. The group node tree pointer is then part of the settings dictionary.
2013-05-08 15:40:46 +00:00
Lukas Toenne
33e990109b
Expose the HIDE_VALUE flag of node sockets in RNA.
2013-05-08 15:40:44 +00:00
Lukas Toenne
bfa97b4710
Workaround for C nodes: In order to make registerable RNA methods of the standard C nodes (e.g. poll or draw_buttons) available in python scripts, they need a specialized Node subtype (called NodeInternal). This is necessary because bpy omits any registerable functions of RNA types in the generated python classes, relying instead on using the supposed native implementation in a registered python class. Since the standard shader/compositor/texture nodes in Blender are not registered but directly created in makesrna they lack all registerable function in the associated python types. The NodeInternal RNA subtype replaces the registerable functions of the base Node type to solve this issue.
2013-05-08 15:40:40 +00:00
Lukas Toenne
6fe753c11b
Extended the draw callback API for node sockets with an explicit text parameter, instead of always using the sock->name string. This can be useful for drawing the socket values separate from the label, e.g. in the node view template.
2013-05-08 14:58:41 +00:00
Lukas Toenne
672d393517
Change to socket draw functions: instead of always only drawing the socket label for connected sockets, leave this check up to the socket draw function itself. This allows future socket types to draw buttons or other info in all cases and handle connected/unconnected state more flexibly.
...
The drawinputfunc/drawoutputfunc callbacks in bNodeType are pretty much empty wrappers now and should be removed at some point. This per-node differentiation should rather be implemented as a specialized socket type if necessary. The only use case for this feature that remains is the file output node in compositor, which displays shortened file format info for each socket.
2013-05-08 14:58:37 +00:00