Comments on: Understanding Half-Pixel and Half-Texel Offsets

By: Edge Detector Antialiasing Shader Stalker Tabula Rasa | Ploobs

Edge Detector Antialiasing Shader Stalker Tabula Rasa | Ploobs — Sat, 25 Jun 2011 12:05:03 +0000

[...] The pixel_size variable (a float2 with 1/target width in x and 1/target height in y) is used in DirectX 9 Shaders (Opengl and DirectX > 9 does not need this). The need for this “correction” is very well explained here. [...]

By: Drawing FullScreen Quads with XNA 4.0 | Ploobs

Drawing FullScreen Quads with XNA 4.0 | Ploobs — Thu, 23 Jun 2011 14:05:10 +0000

[...] XNA 4.0 works under DirectX9.0. In this environment, we have to make a “somehow strange correction” to the position of our vertices in the Vertex Shader (subtract the halfPixel size) when processing quads. The reason for this is the algorithm used in the rasterization unit (More info about this is Here). [...]

By: kdmiller3

kdmiller3 — Sun, 13 Feb 2011 06:49:48 +0000

The half-pixel shift is particularly ingenious. I solved a text corruption issue with MSAA by turning off MSAA for those polygons, but forcing MSAA in the driver bypassed that fix. The half-pixel shift is completely immune.

By: jischnieder

jischnieder — Sun, 09 Jan 2011 18:48:01 +0000

The best explanation of this subject. Thank you very much!!

By: Bhenhur

Bhenhur — Mon, 26 Apr 2010 00:05:28 +0000

Cheers for this. Best explanation of this DirectX9 oddity I’ve come across so far.