Minor bug fixes for the parameter box.

Minor bug fixes for the parameter box.

Minor bug fixes for the parameter box.

As 16-bit floating point numbers are becoming more common in Houdini, here are some statistics about what a 16 bit float can store:

• Largest value is 65504.0
• Smallest non-zero value is around .001

In general a half precision floating point number will yield around 4 digits of precision. For many cases (i.e. texture maps), the advantage of requiring half the memory is a clear advantage.

As 16-bit floating point numbers are becoming more common in Houdini, here are some statistics about what a 16 bit float can store:

• Largest value is 65504.0
• Smallest non-zero value is around .001

In general a half precision floating point number will yield around 4 digits of precision. For many cases (i.e. texture maps), the advantage of requiring half the memory is a clear advantage.

As 16-bit floating point numbers are becoming more common in Houdini, here are some statistics about what a 16 bit float can store:

• Largest value is 65504.0
• Smallest non-zero value is around .001

In general a half precision floating point number will yield around 4 digits of precision. For many cases (i.e. texture maps), the advantage of requiring half the memory is a clear advantage.

As 16-bit floating point numbers are becoming more common in Houdini, here are some statistics about what a 16 bit float can store:

• Largest value is 65504.0
• Smallest non-zero value is around .001

In general a half precision floating point number will yield around 4 digits of precision. For many cases (i.e. texture maps), the advantage of requiring half the memory is a clear advantage.

As 16-bit floating point numbers are becoming more common in Houdini, here are some statistics about what a 16 bit float can store:

• Largest value is 65504.0
• Smallest non-zero value is around .001

In general a half precision floating point number will yield around 4 digits of precision. For many cases (i.e. texture maps), the advantage of requiring half the memory is a clear advantage.

It's now possible to specify more than just 8 and 16 bit depths for image conversion using iconvert. The -d option takes: -d byte,char,8 = 8 bit unsigned int -d short,16 = 16 bit unsigned int -d half = 16 bit floating point -d float = 32 bit floating point This only works for formats which support the bit depth specified. For example, SGI images don't support half-precision floating point images.

It's now possible to specify more than just 8 and 16 bit depths for image conversion using iconvert. The -d option takes: -d byte,char,8 = 8 bit unsigned int -d short,16 = 16 bit unsigned int -d half = 16 bit floating point -d float = 32 bit floating point This only works for formats which support the bit depth specified. For example, SGI images don't support half-precision floating point images.

It's now possible to specify more than just 8 and 16 bit depths for image conversion using iconvert. The -d option takes: -d byte,char,8 = 8 bit unsigned int -d short,16 = 16 bit unsigned int -d half = 16 bit floating point -d float = 32 bit floating point This only works for formats which support the bit depth specified. For example, SGI images don't support half-precision floating point images.

It's now possible to specify more than just 8 and 16 bit depths for image conversion using iconvert. The -d option takes: -d byte,char,8 = 8 bit unsigned int -d short,16 = 16 bit unsigned int -d half = 16 bit floating point -d float = 32 bit floating point This only works for formats which support the bit depth specified. For example, SGI images don't support half-precision floating point images.

It's now possible to specify more than just 8 and 16 bit depths for image conversion using iconvert. The -d option takes: -d byte,char,8 = 8 bit unsigned int -d short,16 = 16 bit unsigned int -d half = 16 bit floating point -d float = 32 bit floating point This only works for formats which support the bit depth specified. For example, SGI images don't support half-precision floating point images.

Added "Enforce Consistent Topology" parameter to the Subdivide SOP

Added "Enforce Consistent Topology" parameter to the Subdivide SOP

Added "Enforce Consistent Topology" parameter to the Subdivide SOP

Added "Enforce Consistent Topology" parameter to the Subdivide SOP

Added "Enforce Consistent Topology" parameter to the Subdivide SOP

In all shading contexts, there are now functions to evaluate the BRDF's for the built-in lighting models

• specularBRDF
• phongBRDF
• blinnBRDF
• diffuseBRDF (both Lambertian and Oren-Nayar)

In all shading contexts, there are now functions to evaluate the BRDF's for the built-in lighting models

• specularBRDF
• phongBRDF
• blinnBRDF
• diffuseBRDF (both Lambertian and Oren-Nayar)

In all shading contexts, there are now functions to evaluate the BRDF's for the built-in lighting models

• specularBRDF
• phongBRDF
• blinnBRDF
• diffuseBRDF (both Lambertian and Oren-Nayar)

In all shading contexts, there are now functions to evaluate the BRDF's for the built-in lighting models

• specularBRDF
• phongBRDF
• blinnBRDF
• diffuseBRDF (both Lambertian and Oren-Nayar)

In all shading contexts, there are now functions to evaluate the BRDF's for the built-in lighting models

• specularBRDF
• phongBRDF
• blinnBRDF
• diffuseBRDF (both Lambertian and Oren-Nayar)

The texture/colormap functions in houdini now support access of images stored in half-float (16 bit float) format.

The texture/colormap functions in houdini now support access of images stored in half-float (16 bit float) format.