Preparing for Octahedral Indirect Lighting — Handmade Hero — Episode Guide

0:04Recap our new radial interpolation of lighting samples and set the stage for the day

🗩

0:04Recap our new radial interpolation of lighting samples and set the stage for the day

🗩

0:04Recap our new radial interpolation of lighting samples and set the stage for the day

🗩

3:43This is Molly Bean

🗩

3:43This is Molly Bean

🗩

3:43This is Molly Bean

🗩

4:58Determine to: 1) Use the ray caster in the octahedral light maps, and 2) Streamline the lighting voxels intermediate storage out of the equation

🗩

4:58Determine to: 1) Use the ray caster in the octahedral light maps, and 2) Streamline the lighting voxels intermediate storage out of the equation

🗩

4:58Determine to: 1) Use the ray caster in the octahedral light maps, and 2) Streamline the lighting voxels intermediate storage out of the equation

🗩

8:55Demo the current state of the lighting, noting our artifacts when a light passes through a probe, and the persistent flower shape due to texel interpolation

🏃

8:55Demo the current state of the lighting, noting our artifacts when a light passes through a probe, and the persistent flower shape due to texel interpolation

🏃

8:55Demo the current state of the lighting, noting our artifacts when a light passes through a probe, and the persistent flower shape due to texel interpolation

🏃

11:14Toggle off the light probe drawing in EndLightingComputation()

11:31Traverse the garden without visible light probes

🏃

11:31Traverse the garden without visible light probes

🏃

11:31Traverse the garden without visible light probes

🏃

12:06Toggle off EDGE_COLORING in EndLightingComputation() and the light and depth map viewers in OpenGLEndFrame()

12:51Traverse the clean garden, with the determination to bounce light around

🏃

12:51Traverse the clean garden, with the determination to bounce light around

🏃

12:51Traverse the clean garden, with the determination to bounce light around

🏃

13:19Set up to use the ray caster in the octahedral light maps

📖

13:19Set up to use the ray caster in the octahedral light maps

📖

13:19Set up to use the ray caster in the octahedral light maps

📖

14:36Remove the accumulation data from light_voxel_cell, for EndLightingComputation() to update the light directly, enabling TEST_LIGHT_TRANSFER

17:26See a data-less world

🏃

17:26See a data-less world

🏃

17:26See a data-less world

🏃

17:37Remove SpamVoxel() and trim down SpamVoxelSlice()

20:36Consider casting rays from locations other than at light probes

🗩

20:36Consider casting rays from locations other than at light probes

🗩

20:36Consider casting rays from locations other than at light probes

🗩

22:23Rename SpamVoxelSlice() to SpamVoxel() and make it blend in the lighting directly, augmenting lighting_solution with tUpdateBlend

28:36Make TestCastFromProbes() pass an octahedral light map location to SpamVoxel(), introducing GetOctahedralOffset()

39:30See some lighting

🏃

39:30See some lighting

🏃

39:30See some lighting

🏃

41:40Investigate our apparent voxel spamming bug

📖

41:40Investigate our apparent voxel spamming bug

📖

41:40Investigate our apparent voxel spamming bug

📖

44:05Make TestCastFromProbes() always call SpamVoxel(), whether or not we hit a light

44:25See that we mostly write out a tile's lighting only once we stand on it

🏃

44:25See that we mostly write out a tile's lighting only once we stand on it

🏃

44:25See that we mostly write out a tile's lighting only once we stand on it

🏃

45:03Investigate our light writing bug

📖

45:03Investigate our light writing bug

📖

45:03Investigate our light writing bug

📖

46:31Scrutinise OctahedralFromUnitVector() and GetOctahedralOffset() for bugs

📖

46:31Scrutinise OctahedralFromUnitVector() and GetOctahedralOffset() for bugs

📖

46:31Scrutinise OctahedralFromUnitVector() and GetOctahedralOffset() for bugs

📖

49:45Break in to GetOctahedralOffset() and inspect its values

🏃

49:45Break in to GetOctahedralOffset() and inspect its values

🏃

49:45Break in to GetOctahedralOffset() and inspect its values

🏃

54:25Make SpamVoxel() write structured art into the voxel

56:28See no structured art in the voxels

🏃

56:28See no structured art in the voxels

🏃

56:28See no structured art in the voxels

🏃

57:05Scrutinise EndLightingComputation() for bugs

📖

57:05Scrutinise EndLightingComputation() for bugs

📖

57:05Scrutinise EndLightingComputation() for bugs

📖

58:47Make SpamVoxel() write red everywhere

59:02Only see red in voxels we land on

🏃

59:02Only see red in voxels we land on

🏃

59:02Only see red in voxels we land on

🏃

59:15Wonder why we're apparently not spamming the voxels

📖

59:15Wonder why we're apparently not spamming the voxels

📖

59:15Wonder why we're apparently not spamming the voxels

📖

1:00:41Break in to SpamVoxel() and watch it write to all the voxels

🏃

1:00:41Break in to SpamVoxel() and watch it write to all the voxels

🏃

1:00:41Break in to SpamVoxel() and watch it write to all the voxels

🏃

1:02:25Try making TestCastFromProbes() pass an upward-pointing OctOffset vector to SpamVoxel()

1:03:16See red spammed as expected

🏃

1:03:16See red spammed as expected

🏃

1:03:16See red spammed as expected

🏃

1:03:43Revert SpamVoxel() to use the computed lighting values

1:04:02Find that moving the light around does not change the intensity

🏃

1:04:02Find that moving the light around does not change the intensity

🏃

1:04:02Find that moving the light around does not change the intensity

🏃

1:04:40Make TestCastFromProbes() fall-off the light

1:05:44See our lighting fall-off work correctly

🏃

1:05:44See our lighting fall-off work correctly

🏃

1:05:44See our lighting fall-off work correctly

🏃

1:06:28Revert TestCastFromProbes() to only call SpamVoxel() if we hit a light

1:06:38Find that we fail to light into newly visible corners

🏃

1:06:38Find that we fail to light into newly visible corners

🏃

1:06:38Find that we fail to light into newly visible corners

🏃

1:06:59Make TestCastFromProbes() always call SpamVoxel(), but blank out the LightC it we didn't hit a light

1:07:44Find that our light interpolation is pretty fast

🏃

1:07:44Find that our light interpolation is pretty fast

🏃

1:07:44Find that our light interpolation is pretty fast

🏃

1:08:24Try reducing the tUpdateBlend in EndLightingComputation()

1:08:38Find that our light fades out slower than it fades in

🏃

1:08:38Find that our light fades out slower than it fades in

🏃

1:08:38Find that our light fades out slower than it fades in

🏃

1:09:38Try further reducing the tUpdateBlend in EndLightingComputation()

1:09:46Watch the speed of light fading

🏃

1:09:46Watch the speed of light fading

🏃

1:09:46Watch the speed of light fading

🏃

1:11:09Prevent TestCastFromProbes() from boosting the light brightness

1:11:19Briefly watch our light fade

🏃

1:11:19Briefly watch our light fade

🏃

1:11:19Briefly watch our light fade

🏃

1:11:34Increase the tUpdateBlend in TestCastFromProbes()

1:11:50Consider our light fade in and out to be more uniform

🏃

1:11:50Consider our light fade in and out to be more uniform

🏃

1:11:50Consider our light fade in and out to be more uniform

🏃

1:12:08Reduce the tUpdateBlend in TestCastFromProbes()

1:12:10Find that our light fades out slower than it fades in

🏃

1:12:10Find that our light fades out slower than it fades in

🏃

1:12:10Find that our light fades out slower than it fades in

🏃

1:13:45Hunt the code for other pieces of non-linearity in the light fade in / out

📖

1:13:45Hunt the code for other pieces of non-linearity in the light fade in / out

📖

1:13:45Hunt the code for other pieces of non-linearity in the light fade in / out

📖

1:16:54Make a note in SpamVoxel() to hunt down the attack / decay non-linearity

1:17:37Our faster attack may happen if two out of three probes switch from 0 to 1

🗩

1:17:37Our faster attack may happen if two out of three probes switch from 0 to 1

🗩

1:17:37Our faster attack may happen if two out of three probes switch from 0 to 1

🗩

1:18:36Make GenerateRoom() generate one, rather than two, light probes per tile

1:19:17Find that our light still fades out slower than it fades in

🏃

1:19:17Find that our light still fades out slower than it fades in

🏃

1:19:17Find that our light still fades out slower than it fades in

🏃

1:20:11Make GenerateRoom() generate zero light probes per tile

1:20:24See no lighting

🏃

1:20:24See no lighting

🏃

1:20:24See no lighting

🏃

1:20:27Make GenerateRoom() generate one light probe per tile

1:20:44Find that our light still fades out slower than it fades in

🏃

1:20:44Find that our light still fades out slower than it fades in

🏃

1:20:44Find that our light still fades out slower than it fades in

🏃

1:21:02Revert GenerateRoom() to generate two light probes per tile

1:21:06Consider how to investigate our fading speed discrepancy

🗩

1:21:06Consider how to investigate our fading speed discrepancy

🗩

1:21:06Consider how to investigate our fading speed discrepancy

🗩

1:21:48Begin to plot our lighting update equation in Desmos¹

📖

1:21:48Begin to plot our lighting update equation in Desmos¹

📖

1:21:48Begin to plot our lighting update equation in Desmos¹

📖

1:24:29Set the day

🗹

1:24:29Set the day

🗹

1:24:29Set the day

🗹

1:25:26Lighting Linear Blend Per Frame: In vs Out

🖌

1:25:26Lighting Linear Blend Per Frame: In vs Out

🖌

1:25:26Lighting Linear Blend Per Frame: In vs Out

🖌

1:33:46Further study: Perceptual rate of fading in vs out

🖌

1:33:46Further study: Perceptual rate of fading in vs out

🖌

1:33:46Further study: Perceptual rate of fading in vs out

🖌

1:35:25Remove Frumbledygok(), and toggle from TestCastFromProbes() to ComputeLightPropagation() in EndLightingComputation()

1:37:26Make ComputeLightPropagationWork() call SpamVoxel() and cast our real rays into the octahedral light map

1:42:34Begin to update ComputeVoxelIrradianceAt() to sample the octahedral light map

1:45:47Diffuse Sampling

🖌

1:45:47Diffuse Sampling

🖌

1:45:47Diffuse Sampling

🖌

1:49:47Consult 'Dynamic Diffuse Global Illumination with Ray-Traced Irradiance Fields'² on diffuse sampling

📖

1:49:47Consult 'Dynamic Diffuse Global Illumination with Ray-Traced Irradiance Fields'² on diffuse sampling

📖

1:49:47Consult 'Dynamic Diffuse Global Illumination with Ray-Traced Irradiance Fields'² on diffuse sampling

📖

1:52:22Consult 'Real-Time Global Illumination using Precomputed Light Field Probes'³ on diffuse sampling

📖

1:52:22Consult 'Real-Time Global Illumination using Precomputed Light Field Probes'³ on diffuse sampling

📖

1:52:22Consult 'Real-Time Global Illumination using Precomputed Light Field Probes'³ on diffuse sampling

📖

2:03:13Spec out specular sampling in ComputeVoxelIrradianceAt()

2:05:24Glimpse the (broken) specular lighting

🏃

2:05:24Glimpse the (broken) specular lighting

🏃

2:05:24Glimpse the (broken) specular lighting

🏃

2:05:32Defer figuring out diffuse sampling of the octahedral map

🗩

2:05:32Defer figuring out diffuse sampling of the octahedral map

🗩

2:05:32Defer figuring out diffuse sampling of the octahedral map

🗩

2:06:24Q&A

🗩

2:06:24Q&A

🗩

2:06:24Q&A

🗩

2:07:50xxthebigfoxx Q: Is the diffuse component going to be computed on the CPU or the GPU then?

🗪

2:07:50xxthebigfoxx Q: Is the diffuse component going to be computed on the CPU or the GPU then?

🗪

2:07:50xxthebigfoxx Q: Is the diffuse component going to be computed on the CPU or the GPU then?

🗪

2:08:33relvet Q: Since hardware standards move quickly, have you considered doing a pass on the codebase to move to newer versions of OpenGL and SSE? The optimization implications alone should be worth it, or?

🗪

2:08:33relvet Q: Since hardware standards move quickly, have you considered doing a pass on the codebase to move to newer versions of OpenGL and SSE? The optimization implications alone should be worth it, or?

🗪

2:08:33relvet Q: Since hardware standards move quickly, have you considered doing a pass on the codebase to move to newer versions of OpenGL and SSE? The optimization implications alone should be worth it, or?

🗪

2:09:56cerdus What is the current OpenGL version?

🗪

2:09:56cerdus What is the current OpenGL version?

🗪

2:09:56cerdus What is the current OpenGL version?

🗪

2:10:38philliptrudeau Q: Did the paper⁴ say that they compute a separate cosine-weighted radiance for every probe? Where do they store that information?

🗪

2:10:38philliptrudeau Q: Did the paper⁴ say that they compute a separate cosine-weighted radiance for every probe? Where do they store that information?

🗪

2:10:38philliptrudeau Q: Did the paper⁴ say that they compute a separate cosine-weighted radiance for every probe? Where do they store that information?

🗪

2:12:10erdomina Q: Any pointers on reading research papers effectively?

🗪

2:12:10erdomina Q: Any pointers on reading research papers effectively?

🗪

2:12:10erdomina Q: Any pointers on reading research papers effectively?

🗪

2:12:35philliptrudeau Is this a "research paper"?

🗪

2:12:35philliptrudeau Is this a "research paper"?

🗪

2:12:35philliptrudeau Is this a "research paper"?

🗪

2:13:25philliptrudeau Journal of Fluffy Opinion Papers

🗪

2:13:25philliptrudeau Journal of Fluffy Opinion Papers

🗪

2:13:25philliptrudeau Journal of Fluffy Opinion Papers

🗪

2:13:58Wrap it up

🗩

2:13:58Wrap it up

🗩

2:13:58Wrap it up

🗩

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