Debug Overlay Cleanup and Render Group Performance Investigation — Handmade Hero — Episode Guide

0:30"Today's episode might be a little bit of a potpourri"^α

1:06Determine to address the "Fix for profiler clipping issue on threads/frames/clocks view" issue¹

2:07Run the game and demo the incorrect clipping of the profiler

3:54handmade_render_group.cpp: Fix GetClipRect() to use Rectangle.Min

5:00Run the game to see that the profiler clipping is fixed

6:00handmade_debug.cpp: Make DEBUGStart() offset the text Z

8:22Run the game to see that the text drop-shadow looks correct

9:08handmade_debug.cpp: Greatly increase the Z offset for the tooltip, so it draws above the profile bars

10:20Run the game to see the tooltip box above the profile bars, but not its text contents, and discuss possible reasons why

17:53handmade_debug_ui.cpp: Consider simplifying the text system

24:08handmade_debug_ui.cpp: Enable DrawTooltips() to draw its text above everything

25:02Run the game to see our tooltip text back

28:16handmade_debug_interface.h: Temporarily conditionally append the function name to the UniqueFileCounterString__

29:19Run the game to see those function names

30:42handmade_debug.cpp: Enable DrawProfileBars() and DrawFrameBars() to put the Element->Name in their tooltip

32:59Run the game to see our complete tooltips

34:16Investigate why UpdateAndRenderEntities() takes so much time

37:10Run the game to see that the EntityRender routine is the main culprit

39:45handmade_entity.cpp: Further Isolate EntityRender

41:01Run the game to see that RenderVolume and RenderPieces are both slow

41:24Investigate this slowness

43:43Step through the assembly for the RenderPieces block

47:16handmade_render_group.cpp: Rewrite GetBitmapDim() without function overloading

50:40Inspect the assembly for the RenderPieces block and note that GetBitmapDim() has now been inlined, as has PushQuad()

57:03Investigate why RoundReal32ToInt32() isn't being inlined^β

59:29Determine to excise roundf()

1:00:55Consult the Intel Intrinsics Guide for rounding instructions²

1:03:53handmade_intrinsics.h: Make RoundReal32ToInt32() and RoundReal32ToUInt32() use _mm_cvtss_si32()

1:05:25Run the game to see that RenderPieces now takes 0.87% of the time, and consider the importance of profiling and looking at the assembly

1:09:27"Death by a thousand function calls"^γ

1:10:05Investigate why the RenderVolume block takes a lot of time

1:10:55Inspect the assembly for PushLineSegment()^δ

1:13:47handmade_intrinsics.h: Make SquareRoot() use _mm_sqrt_ss()³

1:17:49Run the game to see that that just works, but that RenderVolume is not faster

1:20:05handmade_opengl.cpp: Add some debug timing blocks in OpenGLRenderCommands()

1:22:30Run the game to see that OpenGL::TexturedQuads takes the most time, and investigate why

1:25:51Consider how to speed up passing the textures to the GPU

1:28:08handmade_opengl.cpp: Experiment with ways to speed up the OpenGL::QuadLoop block in OpenGLRenderCommands()

1:35:41Run the game to see garbage, note the speed up of the block, but back out those changes in favour of potentially using texture arrays

1:38:20Run the game to see that the cutscene does not work with the new 3D system

1:39:33Calculate how large of a texture array we would need for the cutscene

1:43:06handmade_cutscene.cpp: Reacquaint ourselves with the cutscene code

1:49:29Run the game and try to interpret how the cutscene is being drawn

1:50:14handmade_render_group.cpp: Make SetCameraTransform() take the NearClipPlane and FarClipPlane

1:53:19handmade_cutscene.cpp: Make RenderLayeredScene() specify the clip planes

1:53:47Run the game and note that our distance based fog is knocking out some of the cutscene

1:54:28handmade_render_group.cpp: Make SetCameraTransform() take Fog

1:56:34Run the game to see all the pieces of the cutscene

1:58:03handmade_cutscene.cpp: Make RenderLayeredScene() set the FocalLength and run the game to see it

2:01:45"Where Santa Claus went, I don't know"^ε

2:01:57Run the game and investigate where Santa Claus went

2:08:45Discover that the w values passed to PushQuad() are not all 0

2:09:57handmade_render_group.cpp: Make PushBitmap() only set the ZBias for upright sprites

2:10:24Run the game to see that Santa Claus is back

2:10:30Q&A

🗩

2:10:30Q&A

🗩

2:10:30Q&A

🗩

2:11:05desuused What will be our way forward to reduce glDrawArray calls? Will we look into texture atlases, megatextures, or invent our own way? Is it better to separate cutscene rendering because it's so different?

🗪

2:11:05desuused What will be our way forward to reduce glDrawArray calls? Will we look into texture atlases, megatextures, or invent our own way? Is it better to separate cutscene rendering because it's so different?

🗪

2:11:05desuused What will be our way forward to reduce glDrawArray calls? Will we look into texture atlases, megatextures, or invent our own way? Is it better to separate cutscene rendering because it's so different?

🗪

2:12:03kilo_pasztetowej Are we forced to use SIMD registers to use those intrinsics or are there some ways to do that without those, like inline assembly or something else?

🗪

2:12:03kilo_pasztetowej Are we forced to use SIMD registers to use those intrinsics or are there some ways to do that without those, like inline assembly or something else?

🗪

2:12:03kilo_pasztetowej Are we forced to use SIMD registers to use those intrinsics or are there some ways to do that without those, like inline assembly or something else?

🗪

2:16:48desuused Have you seen musl standard library? It has very good code quality. sqrtf on x86_64 is one instruction on it. It's for Linux, but parts of it are very portable⁴

🗪

2:16:48desuused Have you seen musl standard library? It has very good code quality. sqrtf on x86_64 is one instruction on it. It's for Linux, but parts of it are very portable⁴

🗪

2:16:48desuused Have you seen musl standard library? It has very good code quality. sqrtf on x86_64 is one instruction on it. It's for Linux, but parts of it are very portable⁴

🗪

2:21:17botondar Do you know whether the sqrt function in math.h uses processor instructions or calculates the results via some software algorithm?⁵

🗪

2:21:17botondar Do you know whether the sqrt function in math.h uses processor instructions or calculates the results via some software algorithm?⁵

🗪

2:21:17botondar Do you know whether the sqrt function in math.h uses processor instructions or calculates the results via some software algorithm?⁵

🗪

2:23:42Close it down

🗩

2:23:42Close it down

🗩

2:23:42Close it down

🗩

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