1 block placement/removal timers
3 removal timing depending on the tool/block combination
6 composite entity animations
8 complex entities are made up of parts which have their own local
9 transform that can be animated (like e.g. an arm or head)
13 with background now being a thing, a padding might be nice
14 that or maybe separate bg from fg rendering
16 it may also be feasible to get rid of SDL_ttf and use freetype
17 directly to eliminate the unneccessary surface creation
18 ftgl might also be worth looking at
23 and show me their output
27 merge IO counters, so number of operations per frame is kept
28 low, no matter what exactly is done
30 store some kind of byte order mark?
34 definitely needs throttling for the internets
36 players stats (who's connected, their ping, and game-relevant
37 things) should be sent to clients
41 (clientside) networking and disk IO are prime candidates for threading
45 select or create a world with configurable parameters
49 pathfinding, better turning behaviour
53 occlusion/neighbor light mixing is implemented, but still linear
54 this could be solved by using a pre-interpolated light texture and
55 mapping light levels to coordinates on that
57 there's a bug where a chunk's model is not updated if its neighbor
58 changes border light levels
59 I kinda mitigated it a little for direct neighbors during linking, but
60 it still can happen in (hopefully) rare corner cases
62 propagation through semi-filled blocks is wonky. I worked around it by
63 having the light propagate into solid blocks, but feels like this
64 could cause some weird behaviours
66 entity lighting is now derived from block light levels
67 it's not interpolated and the calculation is very basic, so it
68 has some unexpected effects
72 maybe like light levels? should also store a direction with it in
73 that case. also, global gravity may be a world option.
74 no, per-block gravity vector is most probably too expensive.
75 better have the chunks store a few point masses (maybe blocks that
76 emit gravitation?) and calculate from that
80 when blocks are not just a solid rock of color, attributes may
81 become interesting. like labels on signs and contents of
84 transparency (blocks and entities)
86 transparent blocks because awesome
88 world generator that is not boring
91 still needs way more block types and structure generation
92 a minimum distance from origin could be interesting as well, to ensure
93 the spawn vicinity doesn't contain bloks with would be useless at the
94 beginning (if there even is such a thing), also it would encourage
96 biomes seem too small, maybe that will become easier to tune when
97 there's a little more diversity between them
99 entity/world collision
101 I don't like the force/spring based collision response, maybe revert
106 need a way to find a suitable location to spawn new players in
107 I imagine a "random block" function of ChunkIndex could be nice
108 (also for use with the AI spawner)
109 also, finding a spawn position for a player must no fail. after a
110 certain number of tries, the world must change to safely accomodate
112 chunk generation could be adjusted to make a little more room near the
113 origin (since that's where the usual spawn point will be), but that's
114 not strictly necessary and might overcomplicate the generation
115 if all fails, the spawner has to modify the world
116 how much space has to be cleared and how to make sure the spawning
117 space connects to "open space" I don't know yet, it's all a little
120 sprite/particle system
122 these could help make the world seem more alive
126 items representing both blocks and non-blocks (such as tools, weapons,
127 armor), with a simpler physics simulation than entities, much like the
129 they can be picked up by entities, so those should have one or more parts
130 in their skeleton to render them when they're "held"
131 players' inventories have to be changed so they select an item rather
133 item IDs could be the block ID for blocks, and anything from 2^16 up for