pipeline.h File Reference

#include "fl/gfx/color_profile.h"
#include "fl/gfx/colorimetric_response.h"
#include "fl/gfx/flux_scalar.h"
#include "fl/gfx/gamut_map.h"
#include "fl/gfx/source_xyz.h"
#include "fl/stl/int.h"
#include "fl/stl/noexcept.h"
#include "fl/stl/shared_ptr.h"
#include "fl/stl/vector.h"

Include dependency graph for pipeline.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes
struct	fl::ResponseLutsQ16
	Owned code-to-light samples for the three physical emitters. More...
struct	fl::StreamingPipelineQ16
	Everything the per-pixel path needs, derived once when a profile binds. More...
struct	fl::WidePipelineQ16
	Allocated only for a bound four- or five-emitter profile. More...

Namespaces
namespace	fl
	Base definition for an LED controller.

Functions
bool	fl::buildStreamingPipelineQ16 (const SourceProfile &source, const colorimetric_response::EmitterProfile &device, GamutPolicy policy, StreamingPipelineQ16 *out, const Chromaticity *target_white=nullptr) FL_NO_EXCEPT
	Bind a source declaration and a device profile into a pipeline.
i32	fl::encodeLinearEmitterQ16 (const StreamingPipelineQ16 &pipeline, u8 emitter, i32 light, FluxScalar flux) FL_NO_EXCEPT
	Scale one linear emitter and invert its physical response, using the identical final stage as processPixel[Wide]Q16.
void	fl::processPixelLinearQ16 (const StreamingPipelineQ16 &pipeline, u8 r, u8 g, u8 b, i32(&light)[3]) FL_NO_EXCEPT
	Pre-response linear emitter light for a power prepass.
void	fl::processPixelQ16 (const StreamingPipelineQ16 &pipeline, u8 r, u8 g, u8 b, i32(&drives)[3]) FL_NO_EXCEPT
	One pixel: an RGB8 code triple to three physical emitter drives in s16.16.
void	fl::processPixelWideLinearQ16 (const StreamingPipelineQ16 &pipeline, u8 r, u8 g, u8 b, i32(&drives)[5]) FL_NO_EXCEPT
void	fl::processPixelWideQ16 (const StreamingPipelineQ16 &pipeline, u8 r, u8 g, u8 b, i32(&drives)[5]) FL_NO_EXCEPT
	Like processPixelQ16, but solves the physical 4/5-emitter hull.
void	fl::setPipelineFluxQ16 (StreamingPipelineQ16 *pipeline, FluxScalar flux) FL_NO_EXCEPT
	Set the composed brightness-and-power scalar for the frames that follow.

---

Class Documentation

fl::ResponseLutsQ16

struct fl::ResponseLutsQ16

Collaboration diagram for fl::ResponseLutsQ16:

Class Members
vector< u16 >	blue
vector< u16 >	green
vector< u16 >	red

fl::StreamingPipelineQ16

struct fl::StreamingPipelineQ16

Collaboration diagram for fl::StreamingPipelineQ16:

Class Members
FluxScalar	flux = FluxScalar::unity()	Brightness times power limiting, as one scalar (C4). Initialized here because FluxScalar has no default constructor – there is no sensible "unset" amplitude, and unity is what a pipeline that has never been told a brightness should do.
GamutMapQ16	gamut	The device hull, its solve, and the lightness bound.
GamutPolicy	gamut_policy = GamutPolicy::ChromaCompress	What to do with a target the device cannot reproduce. ChromaCompress is the default and the one the P7 study selected; Clamp is the caller saying they would rather have the cheap answer. The study measured what that costs: clipping drives into range lands about 20 dE2000 from the reference against a budget of 0.5, because no amount of clamping substitutes for the objective. It is offered because ColorProfileBinding offers it, not because it is close.
shared_ptr< const ResponseLutsQ16 >	response	Absent means linear code-to-light response. With a table, the inverse is applied after flux and before the final wire quantization.
SourceMatrixQ16	source	Source primaries to XYZ, with adaptation to D65 already folded in.
TransferFunction	transfer	The source's transfer function, applied per code.
shared_ptr< const WidePipelineQ16 >	wide

fl::WidePipelineQ16

struct fl::WidePipelineQ16

Collaboration diagram for fl::WidePipelineQ16:

Class Members
GamutMapRgbwQ16	rgbw
GamutMapRgbwwQ16	rgbww
EmitterTopology	topology
vector< u16 >	white1_response
vector< u16 >	white2_response