skia / external / github.com / KhronosGroup / OpenGL-Registry / 108-cleanup-xml / . / extensions / SGIX / SGIX_texture_supersample.txt

Name | |

SGIX_texture_supersample | |

Name Strings | |

GL_SGIX_texture_supersample | |

Version | |

$Date: 1997/07/17 03:04:37 $ $Revision: 1.1 $ | |

Number | |

XXX | |

Dependencies | |

SGIX_texture_lod_bias affects the definition of this extension | |

Overview | |

This extension introduces high quality texture minification | |

filters which do not require prefiltering or multiple level of detail | |

maps. The texture map is supersampled based on a pixel's projected | |

footprint in texture space. The weighted average of the samples is | |

used as the texture value for the pixel. | |

Issues | |

* Should 3D textures be supported? Not for now. | |

* Should the TEXTURE_SUPERSAMPLE_WIDTH_SGIX greater than 1 be supported? | |

Dignam would like to consider supporting them -- they will be quite slow. | |

New Procedures and Functions | |

None | |

New Tokens | |

Accepted by the <pname> parameter of TexParameteri and TexParameterf, | |

and by the <pname> parameter of TexParameteriv and TexParameterfv. | |

TEXTURE_SUPERSAMPLE_WIDTH_SGIX | |

Accepted by the <param> parameter of TexParameteri and TexParameterf, | |

and by the <params> parameter of TexParameteriv and TexParameterfv, when | |

their <pname> parameter is TEXTURE_MIN_FILTER. | |

LINEAR_SUPERSAMPLE_SGIX | |

FILTER4_SUPERSAMPLE_SGIX | |

Additions to Chapter 2 of the 1.1 Specification (OpenGL Operation) | |

None | |

Additions to Chapter 3 of the 1.1 Specification (Rasterization) | |

GL Specification Table 3.9 is updated with the addition of: | |

Name Type Legal Values | |

---- ---- ------------ | |

TEXTURE_MIN_FILTER integer LINEAR_SUPERSAMPLE_SGIX, | |

FILTER4_SUPERSAMPLE_SGIX | |

TEXTURE_SUPERSAMPLE_WIDTH_SGIX integer 1, 2, or 4 | |

Texture Minification | |

------------------------ | |

This extensions substantially changes Section 3.8.1. Previously | |

a single scale factor P was determined based on the pixel's projection | |

into texture space. Now two scale factors, Px and Py, are computed. | |

Px = sqrt(dudx^2 + dvdx^2) | |

Py = sqrt(dudy^2 + dvdy^2) | |

It is also acceptable for an implementation to approximate the ideal | |

functions Px and Py with functions Fx and Fy subject to the following | |

conditions: | |

1. Fx is continuous and monotonically increasing in |du/dx| and |dv/dx|. | |

Fy is continuous and monotonically increasing in |du/dy| and |dv/dy|. | |

2. max(|du/dx|,|dv/dx|} <= Fx <= |du/dx| + |dv/dx|. | |

max(|du/dy|,|dv/dy|} <= Fy <= |du/dy| + |dv/dy|. | |

LamdaX = log2(Px) | |

LamdaY = log2(Py) | |

Nx = round (2^LamdaX); | |

Ny = round (2^LamdaY); | |

Instead of a single sample, Tau, at (u,v), Nx x Ny locations of the | |

texture at LOD 0 are sampled and averaged as follows: | |

i=Nx,j=Ny | |

--- | |

Tau =1/(Nx*Ny) \ Tau( u_sub_ij, v_sub_ij); | |

/ | |

--- | |

i=1,j=1 | |

where: | |

u_sub_ij = u(x - 0.5 + i/(Nx+1), y - 0.5 + j/(Ny+1)); | |

v_sub_ij = v(x - 0.5 + i/(Nx+1), y - 0.5 + j/(Ny+1)); | |

It is acceptable to reduce the smaller of Nx and Ny as the pixel | |

footprint collapses to a line, so long as the sampling rate of the texture | |

coordinates with respect to each of the screen coordinates is maintained. | |

The sampling function for Tau(u,v) will be LINEAR if TEXTURE_MIN_FILTER is | |

LINEAR_SUPERSAMPLE_SGIX, and FILTER4_SGIS if TEXTURE_MIN_FILTER is | |

FILTER4_SUPERSAMPLE_SGIX. | |

The default resampling filter is an equal weighted box filter. A linear | |

or cubic filter may be specified by setting the | |

TEXTURE_SUPERSAMPLE_WIDTH_SGIX from 1 to 2 or 4. | |

When TEXTURE_SUPERSAMPLE_WIDTH_SGIX is 2, the width and height | |

of the pixel footprint in screen space will double, and a linear filter | |

centered at the center of the pixel will be used to weight each sample. | |

When TEXTURE_SUPERSAMPLE_WIDTH_SGIX is 4, the width and height | |

of the pixel footprint in screen space will quadruple, and the FILTER4_SGIS | |

texture filter function will be used to weight each sample. | |

Additions to Chapter 4 of the 1.1 Specification (Per-Fragment Operations | |

and the Frame Buffer) | |

None | |

Additions to Chapter 5 of the 1.1 Specification (Special Functions) | |

None | |

Additions to Chapter 6 of the 1.1 Specification (State and State Requests) | |

None | |

Additions to the GLX Specification | |

None | |

Dependencies on SGIX_texture_lod_bias | |

If SGIX_texture_lod_bias is supported, dudx, dvdx, dwdx, dudy, dvdy, dwdy | |

are replaced with dlodudx, dlodvdx, dlodwdx, dlodudy, dlodvdy, dlodwdy in | |

the Px and Py functions. | |

Dependencies on SGIX_texture_lod | |

LamdaX and LamdaY are clamped to the range [MIN_LOD, MAX_LOD] if the | |

SGIS_texture_lod extension is supported. | |

Errors | |

None | |

New State | |

Initial | |

Get Value Get Command Type Value Attrib | |

--------- ----------- ---- ------- ------ | |

TEXTURE_SUPERSAMPLE_WIDTH_SGIX GetTexParameteriv Z1* 1 texture | |

New Implementation Dependent State | |

Minimum | |

Get Value Get Command Type Value | |

--------- ----------- ---- ------- | |

MAX_TEXTURE_SUPERSAMPLE_WIDTH GetIntegerv Z1* 1 |