Removing Incorrect Voxel-Voxel Reflections — Handmade Hero — Episode Guide

0:01Recap and set the stage for the day

🗩

0:01Recap and set the stage for the day

🗩

0:01Recap and set the stage for the day

🗩

0:51Demo the current state of our grid-based ray cast lighting, noting the occasional feedback

🏃

0:51Demo the current state of our grid-based ray cast lighting, noting the occasional feedback

🏃

0:51Demo the current state of our grid-based ray cast lighting, noting the occasional feedback

🏃

3:30Determine to: 1) visualise lighting energy levels, and 2) interpolate the lighting more diffusely, to avoid the squared artifacting

🏃

3:30Determine to: 1) visualise lighting energy levels, and 2) interpolate the lighting more diffusely, to avoid the squared artifacting

🏃

3:30Determine to: 1) visualise lighting energy levels, and 2) interpolate the lighting more diffusely, to avoid the squared artifacting

🏃

6:41Prevent DebugDrawColorDir() from factoring the Intensity in to the line length

7:22Check out our balled octahedral map DiffuseMap lighting directions

🏃

7:22Check out our balled octahedral map DiffuseMap lighting directions

🏃

7:22Check out our balled octahedral map DiffuseMap lighting directions

🏃

7:37Let DebugDrawColorDir() factor the 0–1 clamped Intensity in to the line length

8:15Check out our more tractable octahedral map DiffuseMap lighting directions

🏃

8:15Check out our more tractable octahedral map DiffuseMap lighting directions

🏃

8:15Check out our more tractable octahedral map DiffuseMap lighting directions

🏃

9:31Consider our bounce-lighting to be plausible

🏃

9:31Consider our bounce-lighting to be plausible

🏃

9:31Consider our bounce-lighting to be plausible

🏃

10:39Leave our lighting flicker for later

🏃

10:39Leave our lighting flicker for later

🏃

10:39Leave our lighting flicker for later

🏃

11:38Determine to establish the notion of lighting energy levels

🏃

11:38Determine to establish the notion of lighting energy levels

🏃

11:38Determine to establish the notion of lighting energy levels

🏃

12:55Maintaining Constant Energy

🖌

12:55Maintaining Constant Energy

🖌

12:55Maintaining Constant Energy

🖌

16:13Explain that BuildDiffuseLightMaps() pre-creates incoming lighting energy values

📖

16:13Explain that BuildDiffuseLightMaps() pre-creates incoming lighting energy values

📖

16:13Explain that BuildDiffuseLightMaps() pre-creates incoming lighting energy values

📖

17:18Specular to Diffuse Blend

🖌

17:18Specular to Diffuse Blend

🖌

17:18Specular to Diffuse Blend

🖌

19:39Cosine-falloff Scattering of Photons

🖌

19:39Cosine-falloff Scattering of Photons

🖌

19:39Cosine-falloff Scattering of Photons

🖌

22:21Make BuildDiffuseLightMaps() scatter out only as much light as comes in

29:54Our world contains a lot less light

🏃

29:54Our world contains a lot less light

🏃

29:54Our world contains a lot less light

🏃

30:15Let GenerateRoom() add lamps

30:44Our lamps don't bounce much light

🏃

30:44Our lamps don't bounce much light

🏃

30:44Our lamps don't bounce much light

🏃

31:13Increase the lamps' light intensity from 2 to 10 in GenerateRoom()

31:26Our lamps bounce more light

🏃

31:26Our lamps bounce more light

🏃

31:26Our lamps bounce more light

🏃

31:38Increase the lamps' light intensity from 10 to 100 in GenerateRoom()

31:57Our world ends up turning red

🏃

31:57Our world ends up turning red

🏃

31:57Our world ends up turning red

🏃

32:09Decrease the lamps' light intensity from 100 to 50 in GenerateRoom()

32:23Our light is not propagating stably

🏃

32:23Our light is not propagating stably

🏃

32:23Our light is not propagating stably

🏃

34:16Scrutinise BuildDiffuseLightMaps() for bugs

📖

34:16Scrutinise BuildDiffuseLightMaps() for bugs

📖

34:16Scrutinise BuildDiffuseLightMaps() for bugs

📖

35:21Prevent GridRayCast() from applying the cosine-falloff to ReflectColorFoo

36:02Our light is still not propagating stably

🏃

36:02Our light is still not propagating stably

🏃

36:02Our light is still not propagating stably

🏃

37:11Let GridRayCast() sample the light when our ray also does not hit

37:49Our light now propagates

🏃

37:49Our light now propagates

🏃

37:49Our light now propagates

🏃

37:59Decrease the lamps' light intensity from 50 to 10 in GenerateRoom()

38:15Our light still propagates

🏃

38:15Our light still propagates

🏃

38:15Our light still propagates

🏃

38:23Increase the lamps' light intensity from 10 to 15 in GenerateRoom()

38:26Our light looks great

🏃

38:26Our light looks great

🏃

38:26Our light looks great

🏃

38:49Increase the lamps' light intensity from 15 to 20 in GenerateRoom()

39:04Our light does seem to leak again great

🏃

39:04Our light does seem to leak again great

🏃

39:04Our light does seem to leak again great

🏃

40:11Prevent GenerateRoom() from adding lamps, and increase the intensity of the debug light by 10× in UpdateAndRenderWorld()

40:43Our light is powerful enough to bounce

🏃

40:43Our light is powerful enough to bounce

🏃

40:43Our light is powerful enough to bounce

🏃

41:03Increase the intensity of the debug light from 10 to 20 in UpdateAndRenderWorld()

41:20Our debug light bounces and leaks

🏃

41:20Our debug light bounces and leaks

🏃

41:20Our debug light bounces and leaks

🏃

41:47Scrutinise the termination sampling in GridRayCast() for bugs

📖

41:47Scrutinise the termination sampling in GridRayCast() for bugs

📖

41:47Scrutinise the termination sampling in GridRayCast() for bugs

📖

43:24Test preventing terminator hits from transferring the light in GridRayCast()

44:29Our light still leaks

🏃

44:29Our light still leaks

🏃

44:29Our light still leaks

🏃

44:59Test preventing early-outs from transferring the light in GridRayCast()

45:19Our light no longer leaks

🏃

45:19Our light no longer leaks

🏃

45:19Our light no longer leaks

🏃

45:57Scrutinise the terminator sampling code in GridRayCast()

📖

45:57Scrutinise the terminator sampling code in GridRayCast()

📖

45:57Scrutinise the terminator sampling code in GridRayCast()

📖

48:01Make ComputeWalkTable() terminate within the current cell, not sample from the next one

48:42Our light still leaks

🏃

48:42Our light still leaks

🏃

48:42Our light still leaks

🏃

49:34Add a PieceType_Occluder to entity_visible_piece_flag for UpdateAndRenderEntities() to use

52:22Our light spreads, not occluded

🏃

52:22Our light spreads, not occluded

🏃

52:22Our light spreads, not occluded

🏃

52:37Make GenerateRoom() add occluders, introducing AddPieceOccluder()

55:43Check out our occluders

🏃

55:43Check out our occluders

🏃

55:43Check out our occluders

🏃

56:04Make GenerateRoom() extend the wall occluders to the top of the tile

58:35Our occluders are not aligned properly

🏃

58:35Our occluders are not aligned properly

🏃

58:35Our occluders are not aligned properly

🏃

58:47Make GenerateRoom() size the occluders relative to their position

59:22Our light leaks less

🏃

59:22Our light leaks less

🏃

59:22Our light leaks less

🏃

1:01:20Our light does leak below from the start

🏃

1:01:20Our light does leak below from the start

🏃

1:01:20Our light does leak below from the start

🏃

1:01:30Test preventing apron hits from transferring the light in GridRayCast()

1:02:45Our light still leaks

🏃

1:02:45Our light still leaks

🏃

1:02:45Our light still leaks

🏃

1:03:01Test preventing terminator hits from transferring the light in GridRayCast()

1:03:15Our light no longer leaks

🏃

1:03:15Our light no longer leaks

🏃

1:03:15Our light no longer leaks

🏃

1:03:25Scrutinise the terminator related code in GridRayCast() and ComputeWalkTable()

📖

1:03:25Scrutinise the terminator related code in GridRayCast() and ComputeWalkTable()

📖

1:03:25Scrutinise the terminator related code in GridRayCast() and ComputeWalkTable()

📖

1:07:44See our non-leaky light

🏃

1:07:44See our non-leaky light

🏃

1:07:44See our non-leaky light

🏃

1:08:02Let terminator hits transfer the light in GridRayCast()

1:08:10Our light leaks

🏃

1:08:10Our light leaks

🏃

1:08:10Our light leaks

🏃

1:08:59Scrutinise the termination code ComputeWalkTable()

📖

1:08:59Scrutinise the termination code ComputeWalkTable()

📖

1:08:59Scrutinise the termination code ComputeWalkTable()

📖

1:10:27Remove the DEBUG_ZERO code from GridRayCast()

1:10:50Edit our DebugGridIndex towards a grid index below our orphanage

🏃

1:10:50Edit our DebugGridIndex towards a grid index below our orphanage

🏃

1:10:50Edit our DebugGridIndex towards a grid index below our orphanage

🏃

1:11:46Try setting DebugGridIndex to 10788 in UpdateLighting()

1:12:14Edit our DebugGridIndex towards a grid index below our orphanage

🏃

1:12:14Edit our DebugGridIndex towards a grid index below our orphanage

🏃

1:12:14Edit our DebugGridIndex towards a grid index below our orphanage

🏃

1:12:35Try setting DebugGridIndex to 11788 in UpdateLighting()

1:12:47Edit our DebugGridIndex towards a grid index below our orphanage

🏃

1:12:47Edit our DebugGridIndex towards a grid index below our orphanage

🏃

1:12:47Edit our DebugGridIndex towards a grid index below our orphanage

🏃

1:13:04Try setting DebugGridIndex to 13788 in UpdateLighting()

1:13:23Edit our DebugGridIndex and DebugRayIndex to 13586 and 335

🏃

1:13:23Edit our DebugGridIndex and DebugRayIndex to 13586 and 335

🏃

1:13:23Edit our DebugGridIndex and DebugRayIndex to 13586 and 335

🏃

1:14:40Make UpdateLighting() set DebugGridIndex and DebugRayIndex to 13586 and 335, and increase the height of our octahedron drawing region in DebugDrawOctahedralValues()

1:15:45Hit a LIGHT_ATLAS_ASSERT in GetTileUnclamped()

🏃

1:15:45Hit a LIGHT_ATLAS_ASSERT in GetTileUnclamped()

🏃

1:15:45Hit a LIGHT_ATLAS_ASSERT in GetTileUnclamped()

🏃

1:15:57Fix the VoxStart centering computation in DebugDrawOctahedralValues()

1:16:35See our taller octahedron drawing stack

🏃

1:16:35See our taller octahedron drawing stack

🏃

1:16:35See our taller octahedron drawing stack

🏃

1:16:49Try making DebugDrawOctahedralValues() draw the entire height's worth of octahedral values

1:17:06Hit an assertion in PushQuad()

🏃

1:17:06Hit an assertion in PushQuad()

🏃

1:17:06Hit an assertion in PushQuad()

🏃

1:17:09Make DebugDrawOctahedralValues() originate the octahedral drawing region at 0 in z

1:17:42Check out our nonsensical octahedral values down below the orphanage

🏃

1:17:42Check out our nonsensical octahedral values down below the orphanage

🏃

1:17:42Check out our nonsensical octahedral values down below the orphanage

🏃

1:18:06Note that, while we can only sample from cells inside our lighting voxel, we may cast outside it

📖

1:18:06Note that, while we can only sample from cells inside our lighting voxel, we may cast outside it

📖

1:18:06Note that, while we can only sample from cells inside our lighting voxel, we may cast outside it

📖

1:18:31Pick a grid cell to closely observe

🏃

1:18:31Pick a grid cell to closely observe

🏃

1:18:31Pick a grid cell to closely observe

🏃

1:20:12Make UpdateLighting() set DebugGridIndex to 0, and remove the AtlasToSpatialGrid expansion (beyond the apron)

1:21:04Find the DebugGridIndex (717) and DebugRayIndex (531) for our close observation grid cell and ray

🏃

1:21:04Find the DebugGridIndex (717) and DebugRayIndex (531) for our close observation grid cell and ray

🏃

1:21:04Find the DebugGridIndex (717) and DebugRayIndex (531) for our close observation grid cell and ray

🏃

1:25:08Make UpdateLighting() set DebugGridIndex and DebugRayIndex to 717 and 531

1:25:19Take a look at our debug ray

🏃

1:25:19Take a look at our debug ray

🏃

1:25:19Take a look at our debug ray

🏃

1:25:35Make DebugDrawOctahedralValues() position the lighting voxel to encompass our ray's target hit

1:26:28Our target is indeed receiving light was somewhere

🏃

1:26:28Our target is indeed receiving light was somewhere

🏃

1:26:28Our target is indeed receiving light was somewhere

🏃

1:28:48Make ComputeVoxelIrradianceAt() take an OcclusionN which opposes our sampling direction, until we hit

1:31:22Our light no longer leaks

🏃

1:31:22Our light no longer leaks

🏃

1:31:22Our light no longer leaks

🏃

1:32:13Increase the intensity of the debug light from 20 to 50 in UpdateAndRenderWorld()

1:32:57Our light leaks a little bit

🏃

1:32:57Our light leaks a little bit

🏃

1:32:57Our light leaks a little bit

🏃

1:33:10Decrease the intensity of the debug light from 50 to 30 in UpdateAndRenderWorld()

1:33:18Our light seems to leak

🏃

1:33:18Our light seems to leak

🏃

1:33:18Our light seems to leak

🏃

1:33:53Decrease the intensity of the debug light from 30 to 10 in UpdateAndRenderWorld(), let UpdateAndRenderEntities() colour occluders like normal and GenerateRoom() add lamps

1:34:29Check out our lighting

🏃

1:34:29Check out our lighting

🏃

1:34:29Check out our lighting

🏃

1:35:21Make GenerateRoom() add occluders for trees

1:36:49Check out our occluding trees

🏃

1:36:49Check out our occluding trees

🏃

1:36:49Check out our occluding trees

🏃

1:37:10Make GenerateRoom() add shrunken occluders for trees

1:39:10Hop towards the outside area

🏃

1:39:10Hop towards the outside area

🏃

1:39:10Hop towards the outside area

🏃

1:39:21Make UpdateAndRenderEntities() draw occluder outlines

1:39:35Our trees lack occluders

🏃

1:39:35Our trees lack occluders

🏃

1:39:35Our trees lack occluders

🏃

1:39:51Fix GenerateRoom() to use the expanded TreeVol for the occluder

1:40:02Check out our thinned tree occluders

🏃

1:40:02Check out our thinned tree occluders

🏃

1:40:02Check out our thinned tree occluders

🏃

1:40:09Make GenerateRoom() further thin the tree occluder, and shrink its height

1:42:43Check out our thinned and shrunken tree occluders

🏃

1:42:43Check out our thinned and shrunken tree occluders

🏃

1:42:43Check out our thinned and shrunken tree occluders

🏃

1:43:37Decrease the LIGHT_FLOOR_VALUE from 0.2 to 0.0

1:44:40Admire our more lit world

🏃

1:44:40Admire our more lit world

🏃

1:44:40Admire our more lit world

🏃

1:45:37Prevent UpdateAndRenderEntities() from drawing occluder outlines, and UpdateLighting() calling DebugDrawOctahedralValues()

1:46:01Admire our lighting

🏃

1:46:01Admire our lighting

🏃

1:46:01Admire our lighting

🏃

1:46:49Prevent PushLight() from drawing light outlines, and make AddSnake() increase the intensity of the lamps

1:47:40Admire our lighting, and plan for: 1) adapting the power curve; 2) better spreading; and 3) reducing the noise

🏃

1:47:40Admire our lighting, and plan for: 1) adapting the power curve; 2) better spreading; and 3) reducing the noise

🏃

1:47:40Admire our lighting, and plan for: 1) adapting the power curve; 2) better spreading; and 3) reducing the noise

🏃

1:49:20Q&A

🗩

1:49:20Q&A

🗩

1:49:20Q&A

🗩

1:49:45mindmark42 Q: In the diffuse map weighting divide, could you get divide by zero?

🗪

1:49:45mindmark42 Q: In the diffuse map weighting divide, could you get divide by zero?

🗪

1:49:45mindmark42 Q: In the diffuse map weighting divide, could you get divide by zero?

🗪

1:51:05vaualbus Q: With this system, how could we give a more physically-based rendering look to the light?

🗪

1:51:05vaualbus Q: With this system, how could we give a more physically-based rendering look to the light?

🗪

1:51:05vaualbus Q: With this system, how could we give a more physically-based rendering look to the light?

🗪

1:52:36pragmascrypt Q: Does the diffuse bounce use the albedo of the materials yet to color the bounce light?

🗪

1:52:36pragmascrypt Q: Does the diffuse bounce use the albedo of the materials yet to color the bounce light?

🗪

1:52:36pragmascrypt Q: Does the diffuse bounce use the albedo of the materials yet to color the bounce light?

🗪

1:53:15Make trees bounce green light, and everything else bounce blue

1:54:00See our tinted light

🏃

1:54:00See our tinted light

🏃

1:54:00See our tinted light

🏃

1:54:34Make trees bounce yellow light

1:54:48See our tinted light

🏃

1:54:48See our tinted light

🏃

1:54:48See our tinted light

🏃

1:55:35Make trees bounce red light

1:55:46See our red-tinted trees

🏃

1:55:46See our red-tinted trees

🏃

1:55:46See our red-tinted trees

🏃

1:56:06Revert the light tinting

1:56:40uplinkcoder Q: Can you compare with the older raycast?

🗪

1:56:40uplinkcoder Q: Can you compare with the older raycast?

🗪

1:56:40uplinkcoder Q: Can you compare with the older raycast?

🗪

1:56:57Disable LIGHTING_USE_GRID

1:57:03Check the old AABB ray cast lighting

🏃

1:57:03Check the old AABB ray cast lighting

🏃

1:57:03Check the old AABB ray cast lighting

🏃

1:58:35miltondts Q: After the shader modification, it appears the light is still leaking. Why is that?

🗪

1:58:35miltondts Q: After the shader modification, it appears the light is still leaking. Why is that?

🗪

1:58:35miltondts Q: After the shader modification, it appears the light is still leaking. Why is that?

🗪

1:59:08Enable LIGHTING_USE_GRID

1:59:18Our light doesn't seem to leak

🏃

1:59:18Our light doesn't seem to leak

🏃

1:59:18Our light doesn't seem to leak

🏃

2:01:12Increase the intensity of the debug light from 10 to 50 in UpdateAndRenderWorld()

2:01:39Our light does leak

🏃

2:01:39Our light does leak

🏃

2:01:39Our light does leak

🏃

2:01:50Decrease the intensity of the debug light from 50 to 10 in UpdateAndRenderWorld()

2:02:15Make GenerateRoom() add ceiling occluders

2:04:29Check out our ceiling

🏃

2:04:29Check out our ceiling

🏃

2:04:29Check out our ceiling

🏃

2:04:52Increase the intensity of the debug light from 10 to 50 in UpdateAndRenderWorld()

2:05:10Our light does leak

🏃

2:05:10Our light does leak

🏃

2:05:10Our light does leak

🏃

2:05:25Decrease the intensity of the debug light from 50 to 10 in UpdateAndRenderWorld()

2:05:56Check our regularly non-leaking light

🏃

2:05:56Check our regularly non-leaking light

🏃

2:05:56Check our regularly non-leaking light

🏃

2:06:36i_am_seabass Q: Is it possible to smooth out the lighting so it's not flickering as much?

🗪

2:06:36i_am_seabass Q: Is it possible to smooth out the lighting so it's not flickering as much?

🗪

2:06:36i_am_seabass Q: Is it possible to smooth out the lighting so it's not flickering as much?

🗪

2:07:14sharlock93 Q: From what I have "perceived" watching Handmade Hero is that lighting has taken up a pretty sizeable chunk of the episodes. Is lighting this inherently complex or am I missing something?

🗪

2:07:14sharlock93 Q: From what I have "perceived" watching Handmade Hero is that lighting has taken up a pretty sizeable chunk of the episodes. Is lighting this inherently complex or am I missing something?

🗪

2:07:14sharlock93 Q: From what I have "perceived" watching Handmade Hero is that lighting has taken up a pretty sizeable chunk of the episodes. Is lighting this inherently complex or am I missing something?

🗪

2:07:57toocanzs Q: What are your plans to fix the square looking interpolation?

🗪

2:07:57toocanzs Q: What are your plans to fix the square looking interpolation?

🗪

2:07:57toocanzs Q: What are your plans to fix the square looking interpolation?

🗪

2:08:24internationalizationist Q: Do you think, will there be such topics with which we (you) will spend as much time as with light?

🗪

2:08:24internationalizationist Q: Do you think, will there be such topics with which we (you) will spend as much time as with light?

🗪

2:08:24internationalizationist Q: Do you think, will there be such topics with which we (you) will spend as much time as with light?

🗪

2:08:47billdstrong Q: Even fluid simulation, such as plasma?

🗪

2:08:47billdstrong Q: Even fluid simulation, such as plasma?

🗪

2:08:47billdstrong Q: Even fluid simulation, such as plasma?

🗪

2:11:27ravarix handmade_hero I'd argue networking which has the same issue. Trying to simulate two things which are apart as temporally the same

🗪

2:11:27ravarix handmade_hero I'd argue networking which has the same issue. Trying to simulate two things which are apart as temporally the same

🗪

2:11:27ravarix handmade_hero I'd argue networking which has the same issue. Trying to simulate two things which are apart as temporally the same

🗪

2:12:24andrevon Q: Yet it seems like when you walk you drag the light behind you with inertia, more like a liquid. Is that on purpose or an effect of computation per frame?

🗪

2:12:24andrevon Q: Yet it seems like when you walk you drag the light behind you with inertia, more like a liquid. Is that on purpose or an effect of computation per frame?

🗪

2:12:24andrevon Q: Yet it seems like when you walk you drag the light behind you with inertia, more like a liquid. Is that on purpose or an effect of computation per frame?

🗪

2:13:53ravarix I'm sorry, but that just sounds ignorant to the complexities of game networking

🗪

2:13:53ravarix I'm sorry, but that just sounds ignorant to the complexities of game networking

🗪

2:13:53ravarix I'm sorry, but that just sounds ignorant to the complexities of game networking

🗪

2:16:14internationalizationist Q: What is so special in RTX?

🗪

2:16:14internationalizationist Q: What is so special in RTX?

🗪

2:16:14internationalizationist Q: What is so special in RTX?

🗪

2:17:04equivocatorrr Q: Is the difficulty of global illumination part of the reason why you're doing it?

🗪

2:17:04equivocatorrr Q: Is the difficulty of global illumination part of the reason why you're doing it?

🗪

2:17:04equivocatorrr Q: Is the difficulty of global illumination part of the reason why you're doing it?

🗪

2:18:03ttbjm Q: Even the RTX version of Minecraft has noticeable light lag

🗪

2:18:03ttbjm Q: Even the RTX version of Minecraft has noticeable light lag

🗪

2:18:03ttbjm Q: Even the RTX version of Minecraft has noticeable light lag

🗪

2:19:06philliptrudeau Q: If someone were making a different game with much sparser scenes containing more complex geometry, would you have any alternative prescriptions? Could the probes be arranged in a sparse grid? Or in camera-frustum-space rather than world-space?

🗪

2:19:06philliptrudeau Q: If someone were making a different game with much sparser scenes containing more complex geometry, would you have any alternative prescriptions? Could the probes be arranged in a sparse grid? Or in camera-frustum-space rather than world-space?

🗪

2:19:06philliptrudeau Q: If someone were making a different game with much sparser scenes containing more complex geometry, would you have any alternative prescriptions? Could the probes be arranged in a sparse grid? Or in camera-frustum-space rather than world-space?

🗪

2:21:36andrevon Q: When you tested the secondary bounces with the red light, it seemed like it only worked in left and down directions. Do you have an explanation for that?

🗪

2:21:36andrevon Q: When you tested the secondary bounces with the red light, it seemed like it only worked in left and down directions. Do you have an explanation for that?

🗪

2:21:36andrevon Q: When you tested the secondary bounces with the red light, it seemed like it only worked in left and down directions. Do you have an explanation for that?

🗪

2:23:13Scrutinise the obstruction Coefficient array computation code in ComputeVoxelIrradianceAt()

📖

2:23:13Scrutinise the obstruction Coefficient array computation code in ComputeVoxelIrradianceAt()

📖

2:23:13Scrutinise the obstruction Coefficient array computation code in ComputeVoxelIrradianceAt()

📖

2:24:08Masking out obstructed light cells

🖌

2:24:08Masking out obstructed light cells

🖌

2:24:08Masking out obstructed light cells

🖌

2:25:28brimonk Q: How does an RTX card actually perform ray tracing? Is it similar to what the PS5 can provide, which is hardware to work on a BVH structure to perform ray tracing, or is that different?

🗪

2:25:28brimonk Q: How does an RTX card actually perform ray tracing? Is it similar to what the PS5 can provide, which is hardware to work on a BVH structure to perform ray tracing, or is that different?

🗪

2:25:28brimonk Q: How does an RTX card actually perform ray tracing? Is it similar to what the PS5 can provide, which is hardware to work on a BVH structure to perform ray tracing, or is that different?

🗪

2:30:18mindmark42 Q: Are the probes in the wall still getting cleared to zero?

🗪

2:30:18mindmark42 Q: Are the probes in the wall still getting cleared to zero?

🗪

2:30:18mindmark42 Q: Are the probes in the wall still getting cleared to zero?

🗪

2:30:48mystran The "ray incoherence" is really why a regular GPU has trouble with it as well since thread divergence wreaks havoc with traditional SIMT

🗪

2:30:48mystran The "ray incoherence" is really why a regular GPU has trouble with it as well since thread divergence wreaks havoc with traditional SIMT

🗪

2:30:48mystran The "ray incoherence" is really why a regular GPU has trouble with it as well since thread divergence wreaks havoc with traditional SIMT

🗪

2:34:46Thank you, everyone

🗩

2:34:46Thank you, everyone

🗩

2:34:46Thank you, everyone

🗩

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