foldAdaptationIntoSourceMatrix()

void fl::foldAdaptationIntoSourceMatrix	(	const AdaptationMatrixQ16 &	adaptation,
		SourceMatrixQ16 *	source )

Pre-multiply the adaptation into a source matrix, in place.

This is how the streaming path should use it: primaries and adaptation become one matrix at bind time, so the per-pixel cost of adaptation is zero rather than a second matrix multiply per channel.

Definition at line 151 of file chromatic_adaptation.cpp.hpp.

                                                                          {
    if (source == nullptr) {
        return;
    }
    SourceMatrixQ16 result;
    for (int row = 0; row < 3; ++row) {
        for (int col = 0; col < 3; ++col) {
            const i64 acc =
                static_cast<i64>(adaptation.m[row][0]) * source->m[0][col]
              + static_cast<i64>(adaptation.m[row][1]) * source->m[1][col]
              + static_cast<i64>(adaptation.m[row][2]) * source->m[2][col];
            result.m[row][col] = static_cast<i32>(
                acc >= 0 ? (acc + 32768) >> 16 : -((-acc + 32768) >> 16));
        }
    }
    *source = result;
}

References FL_NO_EXCEPT.

Referenced by buildStreamingPipelineQ16().

Here is the caller graph for this function: