[Bf-docboard-svn] bf-manual: [8179] trunk/blender_docs/manual: Document new Copy Transforms and Target Space options from D9469, D9493.

[Alexander Gavrilov]

Revision: 8179
          https://developer.blender.org/rBM8179
Author:   angavrilov
Date:     2021-07-02 17:12:13 +0200 (Fri, 02 Jul 2021)
Log Message:
-----------
Document new Copy Transforms and Target Space options from D9469, D9493.

Only text; waiting for technical support to add example videos.

Ref rB5a693ce9e396 rBbc8ae58727bf

Modified Paths:
--------------
    trunk/blender_docs/manual/animation/constraints/interface/common.rst
    trunk/blender_docs/manual/animation/constraints/transform/copy_transforms.rst
    trunk/blender_docs/manual/images/animation_constraints_transform_copy-transforms_panel.png

Modified: trunk/blender_docs/manual/animation/constraints/interface/common.rst
===================================================================
--- trunk/blender_docs/manual/animation/constraints/interface/common.rst	2021-07-01 12:53:08 UTC (rev 8178)
+++ trunk/blender_docs/manual/animation/constraints/interface/common.rst	2021-07-02 15:12:13 UTC (rev 8179)
@@ -94,12 +94,12 @@
    and orientation of the bone itself. Only transformations applied to the object or bone itself
    are taken into account.
 
-Local with Parent (bones only)
+Local with Parent :guilabel:`Bones Only`
    The bone position and orientation is evaluated relative to its rest pose location and orientation,
    thus including both its own transformations and those caused by a possible parent relationship
    (i.e. the chain's transformations above the bone).
 
-Pose Space (bones only)
+Pose Space :guilabel:`Bones Only`
    The bone position and orientation is evaluated in the armature object local space
    (i.e. independently from the armature transformations in *Object Mode*).
    Hence, if the armature object has null transformations,
@@ -110,7 +110,21 @@
    an arbitrary object or bone that is specified via additional input fields that appear when this option is selected.
    This can be used to evaluate the constraint using an arbitrary coordinate system.
 
+Local Space (Owner Orientation) :guilabel:`Bone Targets Only`
+   This space works like *Local Space*, with an additional coordinate space transformation
+   that compensates for the difference in the rest pose orientations of the owner and target
+   bones. If applied as the *Local Space* of the owner, this will produce the same global
+   space movement as the target, provided parents are still at rest pose.
 
+   This option replaces the following setup with two additional bones:
+
+   1. An extra `child` bone of the `target`, rotated the same as `owner` in rest pose.
+   2. An extra `sibling` bone of the `target`, positioned same as `child` in rest pose and using
+      :doc:`Copy Transforms </animation/constraints/transform/copy_transforms>` in *World Space* from `child`.
+   3. The constraint uses *Local Space* of `sibling`.
+
+   .. TODO: add a demonstration video
+
 .. _bpy.types.constraint.influence:
 
 Influence

Modified: trunk/blender_docs/manual/animation/constraints/transform/copy_transforms.rst
===================================================================
--- trunk/blender_docs/manual/animation/constraints/transform/copy_transforms.rst	2021-07-01 12:53:08 UTC (rev 8178)
+++ trunk/blender_docs/manual/animation/constraints/transform/copy_transforms.rst	2021-07-02 15:12:13 UTC (rev 8179)
@@ -19,22 +19,45 @@
    :ref:`ui-data-id` used to select the constraints target, and is not functional (red state) when it has none.
    See :ref:`common constraint properties <rigging-constraints-interface-common-target>` for more information.
 
+Remove Target Shear
+   Removes shearing from the target transformation after the target space conversion, ensuring it consists purely
+   of translation, rotation and scale. Note that :doc:`Copy Rotation </animation/constraints/transform/copy_rotation>`
+   always does this.
+
 Mix
    Specifies how the copied transformation is combined with the existing transformation.
 
    Replace
       The new transformation replaces the existing transformation.
-   Before Original
-      The new transformation is added before the existing transformation, as if it was
-      applied to an imaginary parent of the constraint owner. Scale is handled like in
-      the :ref:`Aligned Inherit Scale <bpy.types.EditBone.inherit_scale>` mode of bones
-      to avoid creating shear.
-   After Original
-      The new transformation is added after the existing transformation, as if it was
-      applied locally to an imaginary child of the constraint owner. Scale is handled like
-      in the :ref:`Aligned Inherit Scale <bpy.types.EditBone.inherit_scale>` mode of bones
-      to avoid creating shear.
 
+   Before/After Original (Full)
+      The new transformation is added before/after the existing transformation, as if it was
+      applied to an imaginary parent/child of the constraint owner. Scale is handled like in
+      the most basic :ref:`Full Inherit Scale <bpy.types.EditBone.inherit_scale>` mode of bones,
+      so combining non-uniform scale and rotation will create shear.
+
+   Before/After Original (Aligned)
+      The new transformation is added before/after the existing transformation, as if it was
+      applied to an imaginary parent/child of the constraint owner. Scale is handled like in
+      the :ref:`Aligned Inherit Scale <bpy.types.EditBone.inherit_scale>` mode of bones to
+      avoid creating shear.
+
+      This is equivalent to using the *Split Channels* option, but replacing the location component with
+      the result of *Full*. If only uniform scale is used, the result is identical to *Full*.
+
+   Before/After Original (Split Channels)
+      Combines location, rotation and scale components of the transformation separately, similar
+      to a sequence of three :doc:`Copy Location </animation/constraints/transform/copy_location>`,
+      :doc:`Copy Rotation </animation/constraints/transform/copy_rotation>` and
+      :doc:`Copy Scale </animation/constraints/transform/copy_scale>` (with Offset)
+      constraints bundled together in one operation; the result may be slightly different
+      in case of sheared inputs.
+
+      Unlike *Aligned*, in this mode location channels are simply added together, so rotation
+      and scale components of the input transformations cannot affect the resulting location.
+
+   .. TODO: add video comparing mix modes
+
 Target/Owner
    Standard conversion between spaces.
    See :ref:`common constraint properties <rigging-constraints-interface-common-space>` for more information.

Modified: trunk/blender_docs/manual/images/animation_constraints_transform_copy-transforms_panel.png
===================================================================
(Binary files differ)