diff options
Diffstat (limited to 'doc/testspecs')
| -rw-r--r-- | doc/testspecs/GLES2/functional.scissors.txt | 23 |
1 files changed, 9 insertions, 14 deletions
diff --git a/doc/testspecs/GLES2/functional.scissors.txt b/doc/testspecs/GLES2/functional.scissors.txt index ec50a34b5..db1c81d7c 100644 --- a/doc/testspecs/GLES2/functional.scissors.txt +++ b/doc/testspecs/GLES2/functional.scissors.txt @@ -42,19 +42,14 @@ Excludes: Description: -This test set compares images rendered by a reference software -rasterizer to images rendered by the target. +This test set compares images rendered by the target with the scissor test +enabled to references generated with the scissor test disabled but with +areas outside the scissor rectangle cleared by the test logic. -Primitives are tested one at a time with the scissor area in one of the -above configurations. +All primitive types are tested in all reasonable scissor configurations +listed above. Points are tested in a somewhat limited fashion since +support for point sizes above one is not required. -Color clears are tested by clearing to a color with the scissor test disabled -and then clearing to a different color with the scissor test enabled in each -of the above scissor area configurations. - -Depth & stencil clears are tested by clearing only the scissor area to a value -that will permit a subsequent primitive to be rendered in the presence of -depth/stencil testing. - -Framebuffer blitting is tested like clearing but with the last clear replaced -by a blit. +After clearing with scissoring enabled, depth and stencil clears disable +scissoring and render a colored fullscreen quad with depth & stencil +tests requiring values set by the previous clears to produce the final output. |