For 1.30 we are changing some of the mappings for the different texture projection modes, to make things more consistent. So I will cover the changes here for all the projection types.
The cube, cylinder and sphere shown have a bounding box of (-1, -1, -1)–(1, 1, 1), and use object space coordinates.
Use the mesh UV coordinates. Nothing changed here.
For environment textures and IES Light distributions this will apply a spherical mapping, but for more control over the projection (mainly rotation) you may switch to a spherical projection node in these cases.
XYZ to UVW
also known as: Planar, Flat mapping
Just take the coordinates in world or object space and use them as UVW coordinates. For images only the X and Y coordinates are relevant, which are mapped to U and V. In other words the images use flat mapping projected along the Z axis.
In 1.30 the projection will map image textures to the (-1, -1)–(1, 1) range. So rotating the mapping around the Z axis will rotate the image around the center, as the UVW rotation would do.
also known as: Cube mapping (although some applications may have a different mapping called Cube mapping)
This is an extension of XYZ to UVW mapping, it will pick a different projection axis depending on the normal. This gives a quick way to map a texture on any object without too much distortion, but possibly with a lot of seams. The same change in mapping applies here: image textures are mapped to the (-1, -1)–(1, 1) range.
Perform latitude-longitude mapping for the U and V coordinates, and for procedural textures the W coordinate is the distance from the origin. This is mostly used for environment textures and IES Light distributions. Nothing changed here.
This performs cylindrical mapping: the U coordinate is the longitude, the Y coordinate is the world space Y coordinate. For images the mapping on the Y axis has changed to map the image to the [-1, 1] interval. For procedural textures the W coordinate is the distance from the Y axis. For points on the ground plane (Y = 0) cylindrical and spherical mapping now map to the same points on the images (what would be the equator on spherical mapping).
This mapping takes the world space coordinates, and divides the X and Y coordinates by the Z coordinate. This can be useful if you want to model a projector (you use a texture with this projection as the distribution, with black border mode). It can also be used for camera mapping.
The same change as the other projections applies here: the image is mapped to (-1, -1)–(1, 1), so by default you don't need any offset to use this mapping for projectors or camera mapping.