vftdan
/
cc-stuff
forked from hkc/cc-stuff
1
0
Fork 0
Code Pull Requests Activity
183 Commits 6 Branches 0 Tags 10 MiB
Commit Graph

26 Commits

Author SHA1 Message Date
Vftdan fd810cafea Merge remote-tracking branch 'origin/master' into optimized-lookup 2024-10-03 23:08:04 +02:00
Vftdan 30b6534fc1 Rename character_mode to conversion_mode 2024-10-03 22:40:17 +02:00
Vftdan 53c9ba3db5 Implemented char-fast conversion mode 2024-10-03 22:37:32 +02:00
Vftdan bedc6d9166 Add --mode argument 
Deprecates --fast
 2024-10-03 22:01:34 +02:00
Vftdan 135c3642b0 Fix factored out font_atlas linear search 2024-10-03 21:19:56 +02:00
Vftdan 6ebe98e94c [WIP] attempt at factoring out font_atlas linear search 
For an unknown reason the result is different
 2024-10-03 19:36:56 +02:00
Vftdan 46f8ecd68e GlyphBitmap typedef 2024-10-03 19:35:31 +02:00
Casey 8589b2a730
Build stb_* stuff into a separate blob 
Speeds up build process a little bit because we don't have to build it
every single time
 2024-10-03 19:25:34 +03:00
Casey e2e5d57e1f
Moved things to a separate file 
Font atlas and palettes could be reused by other programs for CC, so
that makes sense IMO.

Also removed debug image saving in cc-pic.py that was left here for ALL
THAT TIME for some fucking reason.
 2024-10-03 18:57:44 +03:00
Vftdan 6028bb419e Avoid duplicate palette entries 
Don't make an item an empty cluster warp candidate, if it is already
warp candidate for another cluster
 2024-10-03 11:26:45 +02:00
Vftdan 5f95f895d2 Try to avoid having unused palette items when the source image only uses a limited gamut region 
Warp each empty cluster's centroid onto the closest dataset item
 2024-10-03 10:20:58 +02:00
Vftdan c91f4d79bd Allocate the correct amount of intermediate centroid data 2024-10-03 10:16:57 +02:00
Vftdan c5af524dac Implement k-means palette generation 2024-10-03 01:27:30 +02:00
Vftdan 849563285c Global resizeable palette 2024-10-02 23:57:12 +02:00
Vftdan d9f4362615 Dedicated palette type that also stores own length 2024-10-02 23:17:29 +02:00
Vftdan be21d42fa0 Now only calculate color difference once for each palette entry pair 
Not clear whether it is more optimal
 2024-10-02 19:45:27 +02:00
Vftdan d765e88679 Only calculate difference between each pixel × each palette entry once 2024-10-02 19:11:57 +02:00
Casey 1eccdde3a3
Moved stuff into separate functions, minor cleanup 2024-10-02 15:31:24 +03:00
Casey a0450a7d59
Removed debug stuff 2024-10-02 07:32:28 +03:00
Casey d511cc407e
Fixed it! Now it works fine 2024-10-02 07:22:53 +03:00
Casey 1e364fe3ea
WIP old converter port, gives funky results 2024-10-02 06:51:51 +03:00
Casey 33e3c1ad8c
Minor changes, unfinished palette selection 2024-09-30 14:19:28 +03:00
Casey a665f9498d
One can only hope 2024-09-30 13:58:22 +03:00
Casey 094308ce0b
There's no fucking way it just worked lol 2024-09-30 13:41:31 +03:00
Casey 0afb88e147
First semi-working prototype of converter in C 
It gives some funky results tho
 2024-09-30 13:32:59 +03:00
Casey bf3c5e6b1c
C version of cc-pic (in progress) 2024-01-29 09:41:42 +03:00