All Generators now have support for generating Interlaced images.

Field COP now does Pulldown and Pushup on arbitrary film and video rates.

Field COP now does Pulldown and Pushup on arbitrary film and video rates.

Field COP now does Pulldown and Pushup on arbitrary film and video rates.

Field COP now does Pulldown and Pushup on arbitrary film and video rates.

New COPs:

• Geometry (rasterizes geometry)
• Lookup (from LUT file and CHOPs)
• System (renamed from UNIX, similar functionality)
• Rename (renames planes)
• Fetch
• SubNet
• Blend (blends between 2 sequences)
• Bump (creates a bump map)
• TimeScale (stretchs or changes the frame rate of a sequence)
• Function (assorted mathematical functions)

New COPs:

• Geometry (rasterizes geometry)
• Lookup (from LUT file and CHOPs)
• System (renamed from UNIX, similar functionality)
• Rename (renames planes)
• Fetch
• SubNet
• Blend (blends between 2 sequences)
• Bump (creates a bump map)
• TimeScale (stretchs or changes the frame rate of a sequence)
• Function (assorted mathematical functions)

New COPs:

• Geometry (rasterizes geometry)
• Lookup (from LUT file and CHOPs)
• System (renamed from UNIX, similar functionality)
• Rename (renames planes)
• Fetch
• SubNet
• Blend (blends between 2 sequences)
• Bump (creates a bump map)
• TimeScale (stretchs or changes the frame rate of a sequence)
• Function (assorted mathematical functions)

New COPs:

• Geometry (rasterizes geometry)
• Lookup (from LUT file and CHOPs)
• System (renamed from UNIX, similar functionality)
• Rename (renames planes)
• Fetch
• SubNet
• Blend (blends between 2 sequences)
• Bump (creates a bump map)
• TimeScale (stretchs or changes the frame rate of a sequence)
• Function (assorted mathematical functions)

A preliminary version of a texture baker has been added to vmantra. The -u option can be used to render the 0-1 texture space for an object. Instead of rendering using the point positions, the attribute specified (defaulting to "uv") is projected into NDC screen space and rendered. All shading for the object specified still occurs in world space (with proper object/shader transforms).

The result of a texture baked render will be a texture map which captures the VEX procedural shaders and diffuse lighting information (including shadows if they exist) which can be re-mapped onto the rendered object.

Ideally, each polygon should map to a unique space in the 0-1 texture grid. Overlapping polygons are not handled properly and will result in bad texture map generation.

The texture baker rendering has the following limitations:

• The VEX function frontface(vector, vector) is a no-op. Thisrequires models to be built with consistent normals. Theform frontface(vector, vector, vector) is still enabled, andcan be used as a work-around if required: frontface(N, I, Ng);
• The VEX functions reflectlight(), refractlight() aredisabled since both of these are view dependent. Thetrace() function is still active (as are all otherray-tracing functions) so if required, this can be workedaround.
• The specular(), phong(), blinn() and Oren-Nayar diffuse()functions which are view dependent are disabled.Illuminance loops can still compute specular light which maycause problems since specular light is largely viewdependent and may result in bad textures (i.e. the brushedaluminum shader).
• Only one image can be rendered in a single IFD. Usersshould turn off shadow map and reflection map generation.
• The attribute specified must be a vertex attribute and bethree floats large.

It is possible to render deep rasters outputting arbitrary VEX variables when texture baking. This will generate multiple texture maps simultaneously (or a single deep raster Houdini .pic file).

For example: vmantra < foo.ifd -u logo:uv x.pic

A preliminary version of a texture baker has been added to vmantra. The -u option can be used to render the 0-1 texture space for an object. Instead of rendering using the point positions, the attribute specified (defaulting to "uv") is projected into NDC screen space and rendered. All shading for the object specified still occurs in world space (with proper object/shader transforms).

The result of a texture baked render will be a texture map which captures the VEX procedural shaders and diffuse lighting information (including shadows if they exist) which can be re-mapped onto the rendered object.

Ideally, each polygon should map to a unique space in the 0-1 texture grid. Overlapping polygons are not handled properly and will result in bad texture map generation.

The texture baker rendering has the following limitations:

• The VEX function frontface(vector, vector) is a no-op. Thisrequires models to be built with consistent normals. Theform frontface(vector, vector, vector) is still enabled, andcan be used as a work-around if required: frontface(N, I, Ng);
• The VEX functions reflectlight(), refractlight() aredisabled since both of these are view dependent. Thetrace() function is still active (as are all otherray-tracing functions) so if required, this can be workedaround.
• The specular(), phong(), blinn() and Oren-Nayar diffuse()functions which are view dependent are disabled.Illuminance loops can still compute specular light which maycause problems since specular light is largely viewdependent and may result in bad textures (i.e. the brushedaluminum shader).
• Only one image can be rendered in a single IFD. Usersshould turn off shadow map and reflection map generation.
• The attribute specified must be a vertex attribute and bethree floats large.

It is possible to render deep rasters outputting arbitrary VEX variables when texture baking. This will generate multiple texture maps simultaneously (or a single deep raster Houdini .pic file).

For example: vmantra < foo.ifd -u logo:uv x.pic

A preliminary version of a texture baker has been added to vmantra. The -u option can be used to render the 0-1 texture space for an object. Instead of rendering using the point positions, the attribute specified (defaulting to "uv") is projected into NDC screen space and rendered. All shading for the object specified still occurs in world space (with proper object/shader transforms).

The result of a texture baked render will be a texture map which captures the VEX procedural shaders and diffuse lighting information (including shadows if they exist) which can be re-mapped onto the rendered object.

Ideally, each polygon should map to a unique space in the 0-1 texture grid. Overlapping polygons are not handled properly and will result in bad texture map generation.

The texture baker rendering has the following limitations:

• The VEX function frontface(vector, vector) is a no-op. Thisrequires models to be built with consistent normals. Theform frontface(vector, vector, vector) is still enabled, andcan be used as a work-around if required: frontface(N, I, Ng);
• The VEX functions reflectlight(), refractlight() aredisabled since both of these are view dependent. Thetrace() function is still active (as are all otherray-tracing functions) so if required, this can be workedaround.
• The specular(), phong(), blinn() and Oren-Nayar diffuse()functions which are view dependent are disabled.Illuminance loops can still compute specular light which maycause problems since specular light is largely viewdependent and may result in bad textures (i.e. the brushedaluminum shader).
• Only one image can be rendered in a single IFD. Usersshould turn off shadow map and reflection map generation.
• The attribute specified must be a vertex attribute and bethree floats large.

It is possible to render deep rasters outputting arbitrary VEX variables when texture baking. This will generate multiple texture maps simultaneously (or a single deep raster Houdini .pic file).

For example: vmantra < foo.ifd -u logo:uv x.pic

A preliminary version of a texture baker has been added to vmantra. The -u option can be used to render the 0-1 texture space for an object. Instead of rendering using the point positions, the attribute specified (defaulting to "uv") is projected into NDC screen space and rendered. All shading for the object specified still occurs in world space (with proper object/shader transforms).

The result of a texture baked render will be a texture map which captures the VEX procedural shaders and diffuse lighting information (including shadows if they exist) which can be re-mapped onto the rendered object.

Ideally, each polygon should map to a unique space in the 0-1 texture grid. Overlapping polygons are not handled properly and will result in bad texture map generation.

The texture baker rendering has the following limitations:

• The VEX function frontface(vector, vector) is a no-op. Thisrequires models to be built with consistent normals. Theform frontface(vector, vector, vector) is still enabled, andcan be used as a work-around if required: frontface(N, I, Ng);
• The VEX functions reflectlight(), refractlight() aredisabled since both of these are view dependent. Thetrace() function is still active (as are all otherray-tracing functions) so if required, this can be workedaround.
• The specular(), phong(), blinn() and Oren-Nayar diffuse()functions which are view dependent are disabled.Illuminance loops can still compute specular light which maycause problems since specular light is largely viewdependent and may result in bad textures (i.e. the brushedaluminum shader).
• Only one image can be rendered in a single IFD. Usersshould turn off shadow map and reflection map generation.
• The attribute specified must be a vertex attribute and bethree floats large.

It is possible to render deep rasters outputting arbitrary VEX variables when texture baking. This will generate multiple texture maps simultaneously (or a single deep raster Houdini .pic file).

For example: vmantra < foo.ifd -u logo:uv x.pic

There are new pragmas to bind default handles/selectors for VEX Ops. Please see online VEX docs for further details.

There are new pragmas to bind default handles/selectors for VEX Ops. Please see online VEX docs for further details.

There are new pragmas to bind default handles/selectors for VEX Ops. Please see online VEX docs for further details.

There are new pragmas to bind default handles/selectors for VEX Ops. Please see online VEX docs for further details.

UV Project PI bound to the UV Project SOP which also shows what in the past would have been the guide geometry.

UV Project PI bound to the UV Project SOP which also shows what in the past would have been the guide geometry.

UV Project PI bound to the UV Project SOP which also shows what in the past would have been the guide geometry.

UV Project PI bound to the UV Project SOP which also shows what in the past would have been the guide geometry.

The smooth sop now allows the user to specify Texture UV or Color as target. In this way we apply the smoothing to the current layer. The user still has the option to specify the attribute name. Also we now use vertex attributes by default, and if not found, then we try for a point attribute.

The smooth sop now allows the user to specify Texture UV or Color as target. In this way we apply the smoothing to the current layer. The user still has the option to specify the attribute name. Also we now use vertex attributes by default, and if not found, then we try for a point attribute.

The smooth sop now allows the user to specify Texture UV or Color as target. In this way we apply the smoothing to the current layer. The user still has the option to specify the attribute name. Also we now use vertex attributes by default, and if not found, then we try for a point attribute.

The smooth sop now allows the user to specify Texture UV or Color as target. In this way we apply the smoothing to the current layer. The user still has the option to specify the attribute name. Also we now use vertex attributes by default, and if not found, then we try for a point attribute.