Package net.neoforged.neoforge.client.model.ao

Class EnhancedAoRenderStorage

java.lang.Object

net.minecraft.client.renderer.block.ModelBlockRenderer.CommonRenderStorage

net.minecraft.client.renderer.block.ModelBlockRenderer.AmbientOcclusionRenderStorage

net.neoforged.neoforge.client.model.ao.EnhancedAoRenderStorage

public class EnhancedAoRenderStorage
extends ModelBlockRenderer.AmbientOcclusionRenderStorage

Entrypoint and main class of our enhanced AO pipeline.

Vanilla's AO logic works well for faces that are axis-aligned. That computation is replicated in FullFaceCalculator, with some bug fixes. The job of the enhanced pipeline is to handle faces that are more complicated, by combining multiple full faces as needed using interpolation.

Compared to vanilla, we also remove any assumption about vertex order in the quad.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
private static final record	EnhancedAoRenderStorage.AoObjectCache	Cache these objects so that they don't need to be reallocated for every EnhancedAoRenderStorage.

Field Summary

Fields

Modifier and Type	Field	Description
private static final float	AO_EPS
private static final ThreadLocal<EnhancedAoRenderStorage.AoObjectCache>	AO_OBJECT_CACHE
private static final float	AVERAGE_WEIGHT
private final FullFaceCalculator	calculator	Calculator for full faces.
private static final boolean	COMPARE_WITH_VANILLA	Debug option to compare the emulated vanilla AO with the actual vanilla AO.
private BakedQuad	currentQuad
private static final org.slf4j.Logger	LOGGER
private static final float	MAX_WEIGHT
private final float[]	weights

Fields inherited from class net.minecraft.client.renderer.block.ModelBlockRenderer.AmbientOcclusionRenderStorage

faceShape

Fields inherited from class net.minecraft.client.renderer.block.ModelBlockRenderer.CommonRenderStorage

brightness, cache, faceCubic, facePartial, lightmap, scratchPos, tintCacheIndex, tintCacheValue

Constructor Summary

Constructors

Constructor	Description
EnhancedAoRenderStorage()

Method Summary

Modifier and Type	Method	Description
static void	applyFlatQuadBrightness(BlockAndTintGetter level, BakedQuad quad, ModelBlockRenderer.CommonRenderStorage storage)	"Enhanced" flat shading logic.
void	calculate(BlockAndTintGetter level, BlockState state, BlockPos pos, Direction direction, boolean shade)
private void	calculateAxisAligned(BlockAndTintGetter level, BlockState state, BlockPos pos, Direction direction, boolean shade)	Computes AO for an axis-aligned quad.
private void	calculateIrregular(BlockAndTintGetter level, BlockState state, BlockPos pos, boolean shade)	Computes AO for a general quad.
void	captureQuad(BakedQuad quad)
private static float	interpolateBrightness(AoCalculatedFace in, float[] weights)	Interpolates brightness from the 4 corners of a face.
private static int	interpolateLightmap(AoCalculatedFace in, float[] weights)	Interpolates lightmap from the 4 corners of a face.
private static int	lerpLightmap(int lightmap1, float w1, int lightmap2, float w2)	Interpolates two lightmaps linearly.
(package private) static int	maxLightmap(int lightmap1, int lightmap2)
static ModelBlockRenderer.AmbientOcclusionRenderStorage	newInstance()
private static float	normalComponent(int normal, int axis)	Unpacks a normal component.
private static float	vertexPos(int[] vertices, int vertex, int axis)	Extracts the position of a vertex from quad data.

Methods inherited from class net.minecraft.client.renderer.block.ModelBlockRenderer.AmbientOcclusionRenderStorage

blend

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

COMPARE_WITH_VANILLA

private static final boolean COMPARE_WITH_VANILLA

Debug option to compare the emulated vanilla AO with the actual vanilla AO. Only does something if emulated AO is enabled.

LOGGER

private static final org.slf4j.Logger LOGGER

AO_OBJECT_CACHE

private static final ThreadLocal<EnhancedAoRenderStorage.AoObjectCache> AO_OBJECT_CACHE

calculator

private final FullFaceCalculator calculator

Calculator for full faces.

weights

private final float[] weights

currentQuad

private BakedQuad currentQuad

AO_EPS

private static final float AO_EPS

See Also:

• Constant Field Values

AVERAGE_WEIGHT

private static final float AVERAGE_WEIGHT

See Also:

• Constant Field Values

MAX_WEIGHT

private static final float MAX_WEIGHT

See Also:

• Constant Field Values

Constructor Details

EnhancedAoRenderStorage

public EnhancedAoRenderStorage()

Method Details

newInstance

public static ModelBlockRenderer.AmbientOcclusionRenderStorage newInstance()

applyFlatQuadBrightness

public static void applyFlatQuadBrightness(BlockAndTintGetter level,
 BakedQuad quad,
 ModelBlockRenderer.CommonRenderStorage storage)

"Enhanced" flat shading logic.

captureQuad

public void captureQuad(BakedQuad quad)

Overrides:

captureQuad in class ModelBlockRenderer.AmbientOcclusionRenderStorage

calculate

public void calculate(BlockAndTintGetter level,
 BlockState state,
 BlockPos pos,
 Direction direction,
 boolean shade)

Overrides:

calculate in class ModelBlockRenderer.AmbientOcclusionRenderStorage

calculateAxisAligned

private void calculateAxisAligned(BlockAndTintGetter level,
 BlockState state,
 BlockPos pos,
 Direction direction,
 boolean shade)

Computes AO for an axis-aligned quad.

This is similar to vanilla in how we select whether to use the inside or outside light. However, we still use our own interpolation logic which does not make any assumption about vertex winding order.

calculateIrregular

private void calculateIrregular(BlockAndTintGetter level,
 BlockState state,
 BlockPos pos,
 boolean shade)

Computes AO for a general quad. Projects onto each axis, computes the AO, then combines proportionally to the square of each normal component.

vertexPos

private static float vertexPos(int[] vertices,
 int vertex,
 int axis)

Extracts the position of a vertex from quad data.

Parameters:

vertices - quad data

vertex - vertex index, from 0 to 3 included

axis - axis index, for 0 to 2 included

normalComponent

private static float normalComponent(int normal,
 int axis)

Unpacks a normal component.

interpolateBrightness

private static float interpolateBrightness(AoCalculatedFace in,
 float[] weights)

Interpolates brightness from the 4 corners of a face.

interpolateLightmap

private static int interpolateLightmap(AoCalculatedFace in,
 float[] weights)

Interpolates lightmap from the 4 corners of a face.

lerpLightmap

private static int lerpLightmap(int lightmap1,
 float w1,
 int lightmap2,
 float w2)

Interpolates two lightmaps linearly.

maxLightmap

static int maxLightmap(int lightmap1,
 int lightmap2)