Rebuilding Light Tables — Handmade Hero — Episode Guide

0:01Recap and set the stage for the day building light tables

🗩

0:01Recap and set the stage for the day building light tables

🗩

0:01Recap and set the stage for the day building light tables

🗩

1:16Demo our lighting colour bleed

🏃

1:16Demo our lighting colour bleed

🏃

1:16Demo our lighting colour bleed

🏃

2:07Our SampleDirectionTable in handmade_sampling_spheres.inl samples direction [0, 0, 1] multiple times

📖

2:07Our SampleDirectionTable in handmade_sampling_spheres.inl samples direction [0, 0, 1] multiple times

📖

2:07Our SampleDirectionTable in handmade_sampling_spheres.inl samples direction [0, 0, 1] multiple times

📖

3:35Plan to rebuild our light table in an evenly distributed blue noise pattern

📖

3:35Plan to rebuild our light table in an evenly distributed blue noise pattern

📖

3:35Plan to rebuild our light table in an evenly distributed blue noise pattern

📖

5:59Consider sampling the light at 16 samples per texel

📖

5:59Consider sampling the light at 16 samples per texel

📖

5:59Consider sampling the light at 16 samples per texel

📖

7:25bashtardis You also need it to be blue-noise temporally as well as spatially for best result

🗪

7:25bashtardis You also need it to be blue-noise temporally as well as spatially for best result

🗪

7:25bashtardis You also need it to be blue-noise temporally as well as spatially for best result

🗪

7:53Consider using a pseudo-blue noise sampling pattern, or tuning the distance parameter manually

📖

7:53Consider using a pseudo-blue noise sampling pattern, or tuning the distance parameter manually

📖

7:53Consider using a pseudo-blue noise sampling pattern, or tuning the distance parameter manually

📖

9:13Toggle on the sphere drawing in PushLightingRenderValues()

9:46Check out our sampling sphere

🏃

9:46Check out our sampling sphere

🏃

9:46Check out our sampling sphere

🏃

10:22Plan to rebuild the table then view our drawing

📖

10:22Plan to rebuild the table then view our drawing

📖

10:22Plan to rebuild the table then view our drawing

📖

10:46Reacquaint ourselves with hhsphere.cpp

📖

10:46Reacquaint ourselves with hhsphere.cpp

📖

10:46Reacquaint ourselves with hhsphere.cpp

📖

14:08Remove SamplingSpheres from lighting_solution, the LightSamplingSphereFloatTable from handmade_sampling_spheres.inl with its generator in OutputSphereINL(), InterleaveDirections(), cube_store and sphere_store

18:17Plan to make OutputSphereINL() output the SampleDirectionTable with all the octahedral rays for a given texel, then all those for the next texel, etc.

📖

18:17Plan to make OutputSphereINL() output the SampleDirectionTable with all the octahedral rays for a given texel, then all those for the next texel, etc.

📖

18:17Plan to make OutputSphereINL() output the SampleDirectionTable with all the octahedral rays for a given texel, then all those for the next texel, etc.

📖

20:20Make OutputSphereINL() output the SampleDirectionTable with all the octahedral rays for a given texel, then all those for the next texel, etc.

24:40Reacquaint ourselves with the light sample sieving in GeneratePoissonDistribution()

📖

24:40Reacquaint ourselves with the light sample sieving in GeneratePoissonDistribution()

📖

24:40Reacquaint ourselves with the light sample sieving in GeneratePoissonDistribution()

📖

26:41Propose to draw the sampling sphere coloured by texel, then fix hhsphere.cpp

📖

26:41Propose to draw the sampling sphere coloured by texel, then fix hhsphere.cpp

📖

26:41Propose to draw the sampling sphere coloured by texel, then fix hhsphere.cpp

📖

27:32Make PushLightingRenderValues() draw the sampling sphere coloured by texel

33:18Rework GenerateOctahedralLightingPattern() to structure the sampling directions with all the octahedral rays for a given texel, then all those for the next texel, etc.

35:58The bug: Our call to GeneratePoissonDistribution(), with its randomness, using TotalDirectionCount, may yield fewer samples on a texel than we need

📖

35:58The bug: Our call to GeneratePoissonDistribution(), with its randomness, using TotalDirectionCount, may yield fewer samples on a texel than we need

📖

35:58The bug: Our call to GeneratePoissonDistribution(), with its randomness, using TotalDirectionCount, may yield fewer samples on a texel than we need

📖

37:38Consider asking GeneratePoissonDistribution() for more sampling points

📖

37:38Consider asking GeneratePoissonDistribution() for more sampling points

📖

37:38Consider asking GeneratePoissonDistribution() for more sampling points

📖

38:40Introduce direction_gen for GeneratePoissonDistribution() to return, and remove GeneratePoissonSamples(), TestFunc() and GeneratePoissonLightingPattern()

42:48Hunt for a v2u + operator overload

📖

42:48Hunt for a v2u + operator overload

📖

42:48Hunt for a v2u + operator overload

📖

44:08Introduce a v2u + operator overload

44:37Update GenerateOctahedralLightingPattern() to use our direction_gen, and ask GeneratePoissonDistribution() for 20% more samples than we'll use, introducing AllocDirArray()

51:34Make GenerateOctahedralLightingPattern() print an error on "Direction count underflow for texel"

52:47Finish up the usage code of hhsphere.cpp

57:38Run hhsphere, and hit a number of "Direction count underflow for texel" errors

🗹

57:38Run hhsphere, and hit a number of "Direction count underflow for texel" errors

🗹

57:38Run hhsphere, and hit a number of "Direction count underflow for texel" errors

🗹

59:21Inspect handmade_sampling_spheres.inl, try to compile with it and hit a syntax error

📖

59:21Inspect handmade_sampling_spheres.inl, try to compile with it and hit a syntax error

📖

59:21Inspect handmade_sampling_spheres.inl, try to compile with it and hit a syntax error

📖

1:00:25Manually append a ")" to TOTAL_LIGHT_SAMPLE_DIRECTION_COUNT in handmade_sampling_spheres.inl

1:01:29Remove stale #if 0 from GridRayCast()

1:01:42Append the missing ")" to TOTAL_LIGHT_SAMPLE_DIRECTION_COUNT OutputSphereINL()

1:01:52Update GridRayCast() to use our newly structured SampleDirectionTable

1:18:09Hit a read access violation in GridRayCast()

🏃

1:18:09Hit a read access violation in GridRayCast()

🏃

1:18:09Hit a read access violation in GridRayCast()

🏃

1:18:43Investigate our read access violation in GridRayCast()

🏃

1:18:43Investigate our read access violation in GridRayCast()

🏃

1:18:43Investigate our read access violation in GridRayCast()

🏃

1:21:02Reacquaint ourselves with tTerminate in ComputeWalkTable()

📖

1:21:02Reacquaint ourselves with tTerminate in ComputeWalkTable()

📖

1:21:02Reacquaint ourselves with tTerminate in ComputeWalkTable()

📖

1:25:24Make ComputeWalkTable() set tTerminate to 0 if greater than or equal to 1000000

1:27:20Plan to update remedybg from 0.3.5.0 to 0.3.6.3

🗹

1:27:20Plan to update remedybg from 0.3.5.0 to 0.3.6.3

🗹

1:27:20Plan to update remedybg from 0.3.5.0 to 0.3.6.3

🗹

1:28:05Again hit a read access violation in GridRayCast()

🏃

1:28:05Again hit a read access violation in GridRayCast()

🏃

1:28:05Again hit a read access violation in GridRayCast()

🏃

1:29:39Let ComputeWalkTable() use tTerminate as is, and instead make GenerateOctahedralLightingPattern() output a valid [0, 0, 1] ray direction on underflow

1:30:58Run hhsphere, and hit a number of "Direction count underflow for texel" errors

🗹

1:30:58Run hhsphere, and hit a number of "Direction count underflow for texel" errors

🗹

1:30:58Run hhsphere, and hit a number of "Direction count underflow for texel" errors

🗹

1:31:49Inspect handmade_sampling_spheres.inl and compile just fine

📖

1:31:49Inspect handmade_sampling_spheres.inl and compile just fine

📖

1:31:49Inspect handmade_sampling_spheres.inl and compile just fine

📖

1:32:29Run the game without crashing, but with the sampling sphere possibly being drawn wrong

🏃

1:32:29Run the game without crashing, but with the sampling sphere possibly being drawn wrong

🏃

1:32:29Run the game without crashing, but with the sampling sphere possibly being drawn wrong

🏃

1:33:12Check RenderDiagrams() for participation in lighting

📖

1:33:12Check RenderDiagrams() for participation in lighting

📖

1:33:12Check RenderDiagrams() for participation in lighting

📖

1:34:12Make PushLightingRenderValues() draw the sampling sphere lower down

1:34:31Our sampling sphere is lit, and contains gaps

🏃

1:34:31Our sampling sphere is lit, and contains gaps

🏃

1:34:31Our sampling sphere is lit, and contains gaps

🏃

1:35:55Fix GridRayCast() to set TransferPPS[Tx] to the already accumulated TransferPPSAccum

1:37:11Our lighting is no longer overblown, but still bleeds, and dims while moving

🏃

1:37:11Our lighting is no longer overblown, but still bleeds, and dims while moving

🏃

1:37:11Our lighting is no longer overblown, but still bleeds, and dims while moving

🏃

1:39:19Try making GenerateOctahedralLightingPattern() ask GeneratePoissonDistribution() for 100% more samples than we'll use, and tighten the convergence criteria from 0.025 to 0.0125

1:41:39Run hhsphere without completing

🗹

1:41:39Run hhsphere without completing

🗹

1:41:39Run hhsphere without completing

🗹

1:42:41Revert the convergence criteria to 0.025 in GenerateOctahedralLightingPattern()

1:43:13Run hhsphere, and still hit a number of "Direction count underflow for texel" errors

🗹

1:43:13Run hhsphere, and still hit a number of "Direction count underflow for texel" errors

🗹

1:43:13Run hhsphere, and still hit a number of "Direction count underflow for texel" errors

🗹

1:43:46Investigate why GenerateOctahedralLightingPattern() is not receiving enough sample directions

📖

1:43:46Investigate why GenerateOctahedralLightingPattern() is not receiving enough sample directions

📖

1:43:46Investigate why GenerateOctahedralLightingPattern() is not receiving enough sample directions

📖

1:47:03Our lighting sphere looks the same

🏃

1:47:03Our lighting sphere looks the same

🏃

1:47:03Our lighting sphere looks the same

🏃

1:47:17We're missing something

📖

1:47:17We're missing something

📖

1:47:17We're missing something

📖

1:48:11Break for snacks

🗹

1:48:11Break for snacks

🗹

1:48:11Break for snacks

🗹

1:48:38afk

1:51:56Return with snacks

🗹

1:51:56Return with snacks

🗹

1:51:56Return with snacks

🗹

1:52:12Consider abandoning vs debugging the sampling point picking scheme

📖

1:52:12Consider abandoning vs debugging the sampling point picking scheme

📖

1:52:12Consider abandoning vs debugging the sampling point picking scheme

📖

1:56:40tomisqi What snacks?

🗪

1:56:40tomisqi What snacks?

🗪

1:56:40tomisqi What snacks?

🗪

1:57:21Need more chicharrones

🗩

1:57:21Need more chicharrones

🗩

1:57:21Need more chicharrones

🗩

1:57:48Consider evenly distributing sampling points in texel space

📖

1:57:48Consider evenly distributing sampling points in texel space

📖

1:57:48Consider evenly distributing sampling points in texel space

📖

1:59:01Propose seeding 16 rays per texel, using a white noise distribution, repulsed to increase evenness, and prevented from leaving their containing texel

📖

1:59:01Propose seeding 16 rays per texel, using a white noise distribution, repulsed to increase evenness, and prevented from leaving their containing texel

📖

1:59:01Propose seeding 16 rays per texel, using a white noise distribution, repulsed to increase evenness, and prevented from leaving their containing texel

📖

2:01:02spacenaming Won't you get clamping on the edges?

🗪

2:01:02spacenaming Won't you get clamping on the edges?

🗪

2:01:02spacenaming Won't you get clamping on the edges?

🗪

2:01:37Consider starting with the 16 rays per texel white noise distribution, without repulsion

📖

2:01:37Consider starting with the 16 rays per texel white noise distribution, without repulsion

📖

2:01:37Consider starting with the 16 rays per texel white noise distribution, without repulsion

📖

2:02:09Make GeneratePoissonDistribution() seed the requested rays per texel in a white noise distribution, and remove GenerateOctahedralLightingPattern()

2:09:23Run hhsphere, and find a more coherent SampleDirectionTable

🗹

2:09:23Run hhsphere, and find a more coherent SampleDirectionTable

🗹

2:09:23Run hhsphere, and find a more coherent SampleDirectionTable

🗹

2:09:55Our lighting sphere is only half filled in

🏃

2:09:55Our lighting sphere is only half filled in

🏃

2:09:55Our lighting sphere is only half filled in

🏃

2:10:32Fix GeneratePoissonDistribution() to produce a -1 to 1 distribution

2:11:58Run hhsphere

🗹

2:11:58Run hhsphere

🗹

2:11:58Run hhsphere

🗹

2:12:09Our lighting sphere is complete, if clumpy

🏃

2:12:09Our lighting sphere is complete, if clumpy

🏃

2:12:09Our lighting sphere is complete, if clumpy

🏃

2:13:05Make GeneratePoissonDistribution() repulse the rays, keeping them in the same quadrant

2:18:16Run hhsphere and quickly lock up

🗹

2:18:16Run hhsphere and quickly lock up

🗹

2:18:16Run hhsphere and quickly lock up

🗹

2:19:08Loosen the convergence criteria from 0.025 to 0.14 in GeneratePoissonDistribution()

2:19:35Run hhsphere to completion

🗹

2:19:35Run hhsphere to completion

🗹

2:19:35Run hhsphere to completion

🗹

2:19:52Our lighting sphere is not horrible, but good enough for now

🏃

2:19:52Our lighting sphere is not horrible, but good enough for now

🏃

2:19:52Our lighting sphere is not horrible, but good enough for now

🏃

2:21:19Our rays now look more sane

🏃

2:21:19Our rays now look more sane

🏃

2:21:19Our rays now look more sane

🏃

2:22:26Make FullCast() clamp the RayIndex to the total count minus 1

2:22:34Determine to investigate our green light leakage

🏃

2:22:34Determine to investigate our green light leakage

🏃

2:22:34Determine to investigate our green light leakage

🏃

2:23:19Investigate our green light leakage

📖

2:23:19Investigate our green light leakage

📖

2:23:19Investigate our green light leakage

📖

2:24:47Toggle on the light atlas drawing in OpenGLEndFrame()

2:24:56Our light atlases clearly contain energy bleed

🏃

2:24:56Our light atlases clearly contain energy bleed

🏃

2:24:56Our light atlases clearly contain energy bleed

🏃

2:25:08Make OpenGLChangeToSettings() draw our light atlases in their entirety

2:26:35Check out our full light atlases

🏃

2:26:35Check out our full light atlases

🏃

2:26:35Check out our full light atlases

🏃

2:26:49Make OpenGLChangeToSettings() draw 0.25 of our light atlases in both X and Y

2:27:02Glimpse our light atlases

🏃

2:27:02Glimpse our light atlases

🏃

2:27:02Glimpse our light atlases

🏃

2:27:11Make OpenGLChangeToSettings() draw 0.025 of our light atlases in both X and Y

2:27:34Glimpse our light atlases

🏃

2:27:34Glimpse our light atlases

🏃

2:27:34Glimpse our light atlases

🏃

2:27:37Make OpenGLChangeToSettings() draw 0.025 of our light atlases in X

2:28:05Check out our light atlases

🏃

2:28:05Check out our light atlases

🏃

2:28:05Check out our light atlases

🏃

2:28:35Make OpenGLChangeToSettings() draw 0.0125 of our light atlases in X

2:28:46Our light atlases have a consistent light bleed

🏃

2:28:46Our light atlases have a consistent light bleed

🏃

2:28:46Our light atlases have a consistent light bleed

🏃

2:28:55Let OpenGLChangeToSettings() draw the original portion of our light atlases

2:29:04Traverse the orphanage and watch the light bleed in certain areas

🏃

2:29:04Traverse the orphanage and watch the light bleed in certain areas

🏃

2:29:04Traverse the orphanage and watch the light bleed in certain areas

🏃

2:30:54Start by investigating BlockCopyAtlas()

📖

2:30:54Start by investigating BlockCopyAtlas()

📖

2:30:54Start by investigating BlockCopyAtlas()

📖

2:32:05Assert in UpdateLighting() that dVoxel.z is 0

2:32:23Correctly fail to hit the dVoxel.z assertion

🏃

2:32:23Correctly fail to hit the dVoxel.z assertion

🏃

2:32:23Correctly fail to hit the dVoxel.z assertion

🏃

2:32:36Scour UpdateLighting() for errors in other dVoxel values

📖

2:32:36Scour UpdateLighting() for errors in other dVoxel values

📖

2:32:36Scour UpdateLighting() for errors in other dVoxel values

📖

2:35:21Assert in UpdateLighting() that dVoxel.x and dVoxel.y are <= 1

2:36:18Hit our dVoxel.x assertion

🏃

2:36:18Hit our dVoxel.x assertion

🏃

2:36:18Hit our dVoxel.x assertion

🏃

2:36:30Make UpdateLighting() skip the dVoxel.x and dVoxel.y assertions first time round

2:37:05Correctly fail to hit the dVoxel.x and dVoxel.y assertions

🏃

2:37:05Correctly fail to hit the dVoxel.x and dVoxel.y assertions

🏃

2:37:05Correctly fail to hit the dVoxel.x and dVoxel.y assertions

🏃

2:37:15Remove the dVoxel.x and dVoxel.y assertions from UpdateLighting()

🏃

2:37:15Remove the dVoxel.x and dVoxel.y assertions from UpdateLighting()

🏃

2:37:15Remove the dVoxel.x and dVoxel.y assertions from UpdateLighting()

🏃

2:37:36Scour BlockCopyAtlas() for bugs

📖

2:37:36Scour BlockCopyAtlas() for bugs

📖

2:37:36Scour BlockCopyAtlas() for bugs

📖

2:39:52Traverse the orphanage and watch the light bleed in certain areas

🏃

2:39:52Traverse the orphanage and watch the light bleed in certain areas

🏃

2:39:52Traverse the orphanage and watch the light bleed in certain areas

🏃

2:40:03Toggle off the light atlas drawing in OpenGLEndFrame()

2:40:15Traverse the orphanage and carefully watch the lighting at a slow 0.04 timestep

🏃

2:40:15Traverse the orphanage and carefully watch the lighting at a slow 0.04 timestep

🏃

2:40:15Traverse the orphanage and carefully watch the lighting at a slow 0.04 timestep

🏃

2:41:50Set tUpdateBlend to 1 in UpdateLighting()

2:42:07The lighting clearly blinks when copying the atlases

🏃

2:42:07The lighting clearly blinks when copying the atlases

🏃

2:42:07The lighting clearly blinks when copying the atlases

🏃

2:43:29Scour BlockCopyAtlas() for bugs

📖

2:43:29Scour BlockCopyAtlas() for bugs

📖

2:43:29Scour BlockCopyAtlas() for bugs

📖

2:44:32Toggle on the LIGHT_ATLAS_ASSERT for VALIDATE_TEXEL_ATLAS() to fire

2:44:43Trap on VALIDATE_TEXEL_ATLAS() of the DiffuseAtlas

🏃

2:44:43Trap on VALIDATE_TEXEL_ATLAS() of the DiffuseAtlas

🏃

2:44:43Trap on VALIDATE_TEXEL_ATLAS() of the DiffuseAtlas

🏃

2:45:11Make ValidateTexelComponent() consider a component of between -100 and 100 to be valid

2:45:30Immediately trap on VALIDATE_TEXEL_ATLAS() of the SpecAtlas

🏃

2:45:30Immediately trap on VALIDATE_TEXEL_ATLAS() of the SpecAtlas

🏃

2:45:30Immediately trap on VALIDATE_TEXEL_ATLAS() of the SpecAtlas

🏃

2:45:44Consider deferring until next time the investigation of our out-of-bounds lighting values

📖

2:45:44Consider deferring until next time the investigation of our out-of-bounds lighting values

📖

2:45:44Consider deferring until next time the investigation of our out-of-bounds lighting values

📖

2:49:13Introduce VoxelIndexIsInDim() for GridRayCast() to assert that the sample fetch is in bounds

2:55:54Hit our IsInBounds assertion in GridRayCast()

🏃

2:55:54Hit our IsInBounds assertion in GridRayCast()

🏃

2:55:54Hit our IsInBounds assertion in GridRayCast()

🏃

2:58:36Get it

📖

2:58:36Get it

📖

2:58:36Get it

📖

3:00:34Toggle off the ray drawing in PushLightingRenderValues(), and toggle on DebugDrawSpatialGrid() in UpdateLighting()

3:01:36Immediately trap on VALIDATE_TEXEL_ATLAS() of the SpecAtlas

🏃

3:01:36Immediately trap on VALIDATE_TEXEL_ATLAS() of the SpecAtlas

🏃

3:01:36Immediately trap on VALIDATE_TEXEL_ATLAS() of the SpecAtlas

🏃

3:01:49Toggle off the LIGHT_ATLAS_ASSERT

3:01:57Our bug: We can sample garbage in Z when not hitting an occluder

🏃

3:01:57Our bug: We can sample garbage in Z when not hitting an occluder

🏃

3:01:57Our bug: We can sample garbage in Z when not hitting an occluder

🏃

3:06:26GridRayCast() is already clamping

📖

3:06:26GridRayCast() is already clamping

📖

3:06:26GridRayCast() is already clamping

📖

3:09:10Scour GridRayCast() for a subtle bug causing out-of-bounds sampling

📖

3:09:10Scour GridRayCast() for a subtle bug causing out-of-bounds sampling

📖

3:09:10Scour GridRayCast() for a subtle bug causing out-of-bounds sampling

📖

3:14:00Set the first entry of SampleDirectionTable to an invalid [0, 0, 0]

3:14:49Hit our read access violation in GridRayCast() on an invalid [0, 0, 0] sample direction

🏃

3:14:49Hit our read access violation in GridRayCast() on an invalid [0, 0, 0] sample direction

🏃

3:14:49Hit our read access violation in GridRayCast() on an invalid [0, 0, 0] sample direction

🏃

3:14:52Inspect the values in GridRayCast() on an invalid [0, 0, 0] sample direction

🏃

3:14:52Inspect the values in GridRayCast() on an invalid [0, 0, 0] sample direction

🏃

3:14:52Inspect the values in GridRayCast() on an invalid [0, 0, 0] sample direction

🏃

3:18:21GridRayCast() produces a bogus Txy value

🏃

3:18:21GridRayCast() produces a bogus Txy value

🏃

3:18:21GridRayCast() produces a bogus Txy value

🏃

3:19:17Revert the first entry of SampleDirectionTable to the generated value

3:19:25Consider calling it for today

🗩

3:19:25Consider calling it for today

🗩

3:19:25Consider calling it for today

🗩

3:20:04Take one last look at our light bleed, blinking on copy and light source occlusion

🏃

3:20:04Take one last look at our light bleed, blinking on copy and light source occlusion

🏃

3:20:04Take one last look at our light bleed, blinking on copy and light source occlusion

🏃

3:23:06midnightsun55 Brooo what is going on, Casey is about to do Unreal Engine 6 by himself at this point

🗪

3:23:06midnightsun55 Brooo what is going on, Casey is about to do Unreal Engine 6 by himself at this point

🗪

3:23:06midnightsun55 Brooo what is going on, Casey is about to do Unreal Engine 6 by himself at this point

🗪

3:23:18Call it for today

🗩

3:23:18Call it for today

🗩

3:23:18Call it for today

🗩

Keyboard Navigation

Global Keys

Menu toggling

In-Menu and Index Controls

Quotes and References Menus and Index

Quotes, References and Credits Menus

Filter Menu

Filter and Link Menus

Credits Menu