Confirming No Total Ordering — Handmade Hero — Episode Guide

1:15handmade_render_group.cpp: Correctly compute the SpriteBound values in GetBoundFor()

2:23Recap and set the stage for the day

6:42handmade_render_group.cpp: Enable SortEntries() to do a total ordering check

12:14Run the game, try doing the new check and hit our Assertion

12:46Debugger: Inspect the SortKey values for EntryA and EntryB

15:28handmade_sort.cpp: Temporarily change the Assert in SortEntries() to a BreakHere, and count the SortErrors

17:22Debugger: Break into SortEntries() to see how many sort errors there were

18:13Blackboard: Sort Partial Ordering Problem

20:43Blackboard: Turning our Partial Ordering into a Total Ordering

24:00handmade_sort.cpp: Consider making IsInFrontOf() sort by distance from the camera

26:20Blackboard: Sorting by distance from the camera

29:43Blackboard: Orthographic camera

32:55Blackboard: Will taking the topmost point of cards be sufficient?

35:25handmade_sort.cpp: Introduce BuildSpriteGraph()

40:50Blackboard: Graph Theory

43:14Blackboard: Directed Graph

44:27Blackboard: Directed Acyclic Graph

46:43handmade_sort.cpp: Continue implementing BuildSpriteGraph()

49:45handmade_math.h: Introduce a version of RectanglesIntersect() that takes rectangle2

52:31handmade_sort.cpp: Consider partitioning the screen in order to build this graph and potentially using the Z-buffer

55:40handmade_sort.cpp: Consider leveraging the screen partitioning in order to optimise the software renderer

56:15Q&A

🗩

56:15Q&A

🗩

56:15Q&A

🗩

57:03roam00010011 How much more work would you have to do to get a full 3D sorting?

🗪

57:03roam00010011 How much more work would you have to do to get a full 3D sorting?

🗪

57:03roam00010011 How much more work would you have to do to get a full 3D sorting?

🗪

58:47roam00010011 I mean, use fake 3D data, since our cards are essentially flat, we can give an epsilon width to it for the sorting

🗪

58:47roam00010011 I mean, use fake 3D data, since our cards are essentially flat, we can give an epsilon width to it for the sorting

🗪

58:47roam00010011 I mean, use fake 3D data, since our cards are essentially flat, we can give an epsilon width to it for the sorting

🗪

59:15Blackboard: Sorting 3D objects

1:02:38Blackboard: Sorting all entities in a linear total ordering

1:10:55Blackboard: Common problematic cases in Handmade Hero

1:14:23insofaras Sorry if this was already answered, but what was the issue with a "return a_z != b_z ? a_z - b_z : a_y - b_y" style total ordering?

🗪

1:14:23insofaras Sorry if this was already answered, but what was the issue with a "return a_z != b_z ? a_z - b_z : a_y - b_y" style total ordering?

🗪

1:14:23insofaras Sorry if this was already answered, but what was the issue with a "return a_z != b_z ? a_z - b_z : a_y - b_y" style total ordering?

🗪

1:14:41Blackboard: The problem with sorting heroes behind walls on rugs on tiles

1:19:02insofaras Would the Y then Z sort work if Z is quantized to room layers, and sprites are split where they cross layers (and maybe the ground is special cased)?

🗪

1:19:02insofaras Would the Y then Z sort work if Z is quantized to room layers, and sprites are split where they cross layers (and maybe the ground is special cased)?

🗪

1:19:02insofaras Would the Y then Z sort work if Z is quantized to room layers, and sprites are split where they cross layers (and maybe the ground is special cased)?

🗪

1:20:10Wind it down

🗩

1:20:10Wind it down

🗩

1:20:10Wind 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