diff options
Diffstat (limited to 'renderproto.txt')
| -rw-r--r-- | renderproto.txt | 120 |
1 files changed, 68 insertions, 52 deletions
diff --git a/renderproto.txt b/renderproto.txt index e6bae96..3c8e492 100644 --- a/renderproto.txt +++ b/renderproto.txt @@ -1,6 +1,6 @@ The X Rendering Extension - Version 0.8 - 2003-5-3 + Version 0.9 + 2004-7-23 Keith Packard keithp@keithp.com @@ -29,7 +29,7 @@ This extension was the work of many people, in particular: how OpenGL works. + Carl Worth for providing the sample implementation of - trapezoid rendering + trapezoid rendering and showing how broken the spec was + Sam Pottle and Jamey Sharp for helping demonstrate the correctness of the trapezoid specification. @@ -208,9 +208,13 @@ LINEFIX [ p1, p2: POINTFIX ] TRAP [ + top, bottom: SPANFIX + ] +TRAPEZOID [ top, bottom: FIXED left, right: LINEFIX ] +(TRAPEZOID is deprecated) COLORTRIANGLE [ p1, p2, p3: COLORPOINT ] @@ -372,22 +376,13 @@ required filter alias which must map to one of the available filters. 10. Polygon Rasterization -All polygons must be convex. Rendering of concave polygons is unspecified -except that the result must obey the clipping rules. - -Each polygon request fills the region closed by the specified path. The -path is automatically closed if the last point does not coincide with the -first point. +Render provides only two kinds of polygons, trapezoids and triangles. To +improve efficiency, several different wire encodings exist for each. -A point is infinitely small and the path is an infinitely thin line. A -pixel is inside if the center point of the pixel is inside and the center -point is not on the boundary. If the center point is on the boundary, the -pixel is inside if and only if the polygon interior is immediately to its -right (x increasing direction). Pixels with centers along a horizontal edge -are a special case and are inside if and only if the polygon interior is -immediately below (y increasing direction). A polygon contains a pixel if -the pixel is inside the polygon. +All trapezoids must be convex. Rendering of concave trapezoids is unspecified +except that the result must obey the clipping rules. +Composite Polygons are rasterized by implicit generating an alpha mask and using that in the general compositing operator along with a supplied source image: @@ -404,21 +399,20 @@ definition which is close to the right answer. This value is truncated to the alpha width in the fallback format before application of the compositing operator. ---- +Rasterization +Alpha values are generated by point sampling the coverage of a square +surrounding the center of each pixel by the polygon. -This needs rewriting to match current trapezoid specification and -base other polygons on that. I suspect imprecise polygons will need -to have a relaxed specification as well; hardware is unlikely to -meet the "sum to one" constraint. +In Precise poly mode, the sample points are located in a regular grid. When +alpha depth 'e' is even, the regular grid is 2**(e/2) + 1 samples wide and +2**(e/2) -1 samples high. For odd alpha depth 'o', the sample grid is 2**o +- 1 samples wide and 1 sample high. Note that odd alpha depth usually +occurs only at depth 1, so this misshapen sample grid has no ill effects. +The sample grid is centered within the pixel and then each sample point is +rounded down to a point on the sub-pixel coordinate grid. ---- - -When rasterized in Precise mode, the pixelization will match this -specification exactly. - -When rasterized in Imprecise mode, the pixelization may deviate from this -specification by up to 1/2 pixel along any edge subject to the following -constraints: +In Imprecise mode, the location of the sample points is not specified, but +the implementation must conform to the following constraints: + Abutting edges must match precisely. When specifying two polygons abutting along a common edge, if that edge is specified with the @@ -436,16 +430,6 @@ constraints: + Order independent. Two identical polygons specified with vertices in different orders must generate identical results. -Polygons can also be specified with colors for each vertex. These color -values are interpolated along the edges and across each scanline. - -When rasterized in Precise mode, the interpolated colors are exact. - -When rasterized in Imprecise mode, the color of each pixel may optionally be -interpolated from a triangle containing the pixel which is formed from any -three polygon vertices. Any interpolated color value can err up to 1 lsb in -each channel. - 11. Image Filtering When computing pixels from source and mask images, a filter may be applied @@ -460,8 +444,8 @@ provided: Filter Name Description - Nearest Nearest neighbor filtering - Bilinear Linear interpolation in two dimensions + nearest Nearest neighbor filtering + bilinear Linear interpolation in two dimensions Additional names may be provided for any filter as aliases. A set of standard alias names are required to be mapped to a provided filter so that @@ -469,13 +453,29 @@ applications can use the alias names without checking for availability. Alias name Intended interpretation - Fast High performance, quality similar to Nearest - Good Reasonable performance, quality similar to Bilinear - Best Highest quality available, performance may not + fast High performance, quality similar to Nearest + good Reasonable performance, quality similar to Bilinear + best Highest quality available, performance may not be suitable for interactive use Aliases must map directly to a non-aliased filter name. +There is also a set of standard filters which are not required but may be +provided. If they are provided, using the standard name, they must match +the definition specified here. + + Filter Name Description + + convolution MxN convolution filter. The values specified + in SetPictureFilter are M, N and then M * N + filter parameters. M and N must be integers + represented as fixed point numbers. + gaussian Gaussian blur. The value specified is a radius + in pixels (which can be fractional). A standard + Gaussian 2D convolution filter will be applied. + binomial Binomial blur. An approximation of a Gaussian + blur using binomial coefficients + 12. Glyph Rendering Glyphs are small alpha masks which can be stored in the X server and @@ -816,14 +816,15 @@ Trapezoids src-x, src-y: INT16 dst: PICTURE mask-format: PICTFORMAT or None - traps: LISTofTRAP + traps: LISTofTRAPEZOID - This request rasterizes the list of trapezoids. For each trap, the - area between the left and right edges is filled from the top to the - bottom. src-x and src-y register the pattern to the floor of the - top x and y coordinate of the left edge of the first trapezoid, they - are adjusted for subsequent trapezoids so that the pattern remains - globally aligned within the destination. + This request rasterizes the list of trapezoids. + + For each trap, the area between the left and right edges is filled + from the top to the bottom. src-x and src-y register the pattern to + the floor of the top x and y coordinate of the left edge of the + first trapezoid, they are adjusted for subsequent trapezoids so that + the pattern remains globally aligned within the destination. When mask-format is not None, trapezoids are rendered in the following way with the effective mask computed in mask-format: @@ -1110,6 +1111,19 @@ CreateAnimCursor 'cursors', showing each cursor in the element for the number of milliseconds indicated by the 'delay' member of that element. +AddTraps + picture: PICTURE + off-x, off-y: INT16 + trapezoids: LISTofTRAP + + Errors: Match + + Each trap is PictOpAdd'ed to 'picture'. 'off-x', 'off-y' + are added to each coordinate. + + 'picture' must be an alpha-only picture else a 'Match' error is + returned. + 15. Extension Versioning The Render extension was developed in parallel with the implementation to @@ -1158,3 +1172,5 @@ what each version before 1.0 implemented: QueryPictIndexValues 0.8: CreateAnimCursor + 0.9: + AddTrapezoids |