fglNormalPointer(3G)
fglNormalPointer - define an array of normals
As shipped in IRIX 6.5.19. Last changed in IRIX 6.5.15.
NAME fglNormalPointer - define an array of normals FORTRAN SPECIFICATION SUBROUTINE fglNormalPointer( INTEGER*4 type, INTEGER*4 stride, CHARACTER*8 pointer ) PARAMETERS type Specifies the data type of each coordinate in the array. Symbolic constants GL_BYTE, GL_SHORT, GL_INT, GL_FLOAT, and GL_DOUBLE are accepted. The initial value is GL_FLOAT. stride Specifies the byte offset between consecutive normals. If stride is 0- the initial value-the normals are understood to be tightly packed in the array. The initial value is 0. pointer Specifies a pointer to the first coordinate of the first normal in the array. The initial value is 0. DESCRIPTION fglNormalPointer specifies the location and data format of an array of normals to use when rendering. type specifies the data type of the normal coordinates and stride gives the byte stride from one normal to the next, allowing vertices and attributes to be packed into a single array or stored in separate arrays. (Single-array storage may be more efficient on some implementations; see fglInterleavedArrays.) When a normal array is specified, type, stride, and pointer are saved as client-side state. To enable and disable the normal array, call fglEnableClientState and fglDisableClientState with the argument GL_NORMAL_ARRAY. If enabled, the normal array is used when fglDrawArrays, fglDrawElements, or fglArrayElement is called. Use fglDrawArrays to construct a sequence of primitives (all of the same type) from prespecified vertex and vertex attribute arrays. Use fglArrayElement to specify primitives by indexing vertices and vertex attributes and fglDrawElements to construct a sequence of primitives by indexing vertices and vertex attributes. NOTES fglNormalPointer is available only if the GL version is 1.1 or greater. The normal array is initially disabled and isn't accessed when fglArrayElement, fglDrawElements, or fglDrawArrays is called. Execution of fglNormalPointer is not allowed between fglBegin and the corresponding fglEnd, but an error may or may not be generated. If an error is not generated, the operation is undefined. fglNormalPointer is typically implemented on the client side. Since the normal array parameters are client-side state, they are not saved or restored by fglPushAttrib and fglPopAttrib. Use fglPushClientAttrib and fglPopClientAttrib instead. ERRORS GL_INVALID_ENUM is generated if type is not an accepted value. GL_INVALID_VALUE is generated if stride is negative. ASSOCIATED GETS fglIsEnabled with argument GL_NORMAL_ARRAY fglGet with argument GL_NORMAL_ARRAY_TYPE fglGet with argument GL_NORMAL_ARRAY_STRIDE fglGetPointerv with argument GL_NORMAL_ARRAY_POINTER MACHINE DEPENDENCIES On RealityEngine, RealityEngine2, and VTX systems, do not enable or disable GL_VERTEX_ARRAY, GL_VERTEX_ARRAY_EXT, GL_NORMAL_ARRAY, GL_NORMAL_ARRAY_EXT, GL_COLOR_ARRAY, GL_COLOR_ARRAY_EXT, GL_INDEX_ARRAY,GL_INDEX_ARRAY_EXT, GL_TEXTURE_COORD_ARRAY, GL_TEXTURE_COORD_ARRAY_EXT, GL_EDGE_FLAG_ARRAY or GL_EDGE_FLAG_ARRAY_EXT between a call to fglNewList and the corresponding call to fglEndList. Instead, enable or disable before the call to fglNewList. On InfiniteReality systems it is particularly important to minimize the amount of data transferred from the application to the graphics pipe, since the host-to-pipe bandwidth limit can cause a performance bottleneck. One way to reduce the amount of data transferred per vertex is to use properly-aligned byte and short data types whenever possible. Accordingly, the EXT_vertex_array extension on InfiniteReality systems has been optimized for vertex information packed into the following data structures. (Note: v represents vertex coordinates, c represents color components, n represents normal coordinates, and t represents texture coordinates. Normals must have unit length.) struct {GLfloat v[3];} struct {GLubyte c[4]; GLfloat v[3];} struct {GLshort n[3]; GLfloat v[3];} struct {GLubyte c[4]; GLshort n[3]; GLfloat v[3];} struct {GLshort t[2]; GLfloat v[3];} struct {GLshort t[2]; GLubyte c[4]; GLfloat v[3];} struct {GLshort t[2]; GLshort n[3]; GLfloat v[3];} struct {GLshort t[2]; GLubyte c[4]; GLshort n[3]; GLfloat v[3];} struct {GLfloat t[2]; GLfloat v[3];} struct {GLfloat t[2]; GLubyte c[4]; GLfloat v[3];} struct {GLfloat t[2]; GLshort n[3]; GLfloat v[3];} struct {GLfloat t[2]; GLubyte c[4]; GLshort n[3]; GLfloat v[3];} Application-specific fields may be added to these structures, provided that all the fields described above retain their relative order and word alignment. An additional constraint applies when fglTexGen is being used. The implementation normally generates all four texture coordinates in parallel, and must take special action to generate just a subset of the four coordinates. Therefore performance is best when none of the texture coordinates are being generated, or when all of them are being generated. For example, when using 2D texturing (generating s and t coordinates) it will be faster to enable texture coordinate generation for the r and q coordinates as well as s and t. Choose a texture generation mode of GL_OBJECT_LINEAR and use the plane equations (0,0,0,0) and (0,0,0,1) for r and q, respectively. Using these structures on InfiniteReality systems can improve performance considerably, compared to structures in which all values are single- precision floating point. SEE ALSO fglArrayElement, fglColorPointer, fglDrawArrays, fglDrawElements, fglEdgeFlagPointer, fglEnable, fglGetPointerv, fglIndexPointer, fglInterleavedArrays, fglPopClientAttrib, fglPushClientAttrib, fglTexCoordPointer, fglVertexPointer