The description of various cl_image_format data types in the 1.1 man pages and the 1.1 spec appears to be inconsistent.
From the 1.1 man page table (http://www.khronos.org/registry/cl/sdk/ … ormat.html)
headed
Image Channel Data Type Description
CL_SNORM_INT8 Each channel component is a normalized signed 8-bit integer value.
CL_SNORM_INT16 Each channel component is a normalized signed 16-bit integer value.
CL_UNORM_INT8 Each channel component is a normalized unsigned 8-bit integer value.
CL_UNORM_INT16 Each channel component is a normalized unsigned 16-bit integer value.
This is also present in the 1.1 spec. http://www.khronos.org/registry/cl/spec … df#page=78
Please can someone clarify and confirm that
CL_UNORM_xxx are normalized values
If they are normalized, how does CL_SNORM_x differ from CL_UNORM_x ?
Or is this a typo in the standard and the spec?
Regards
Tim