A converter for multicolor-mode on the TIC-80.
The TIC-80 normaly supports a max. resolution of 240 pixels x 136 pixels with a palette of 16 RGB-colors. But it is possible to change the palette every scanline with some tweaks, resulting in max. 2176 or even 4216 colors on screen.
mulTIColor converts an image (240 x 136 pixels or smaller) to a version with max. 16 or 31 colors per line. The result will be saved as an image and as a Lua-script with all the palettes, pixels and a display-routine (when using the "o"-option). Images with 16 or less colors (indexed) will be converted too, but with a single palette.
To reduce the colors to 16/31 per line, there are 3 options by default to achieve this:
- PIL (Python Image Library, built-in)
- IrfanView (external executable)
- ImageMagick (external executable)
You might have to adjust the paths for IrfanView and ImageMagick in the configfile (mtc.cfg). You can also add other converters or paths for a different operating system to the config aswell. Make sure to include the three keywords {IN}, {OUT} and {RANGE} to your entries. They ensure, that the current line of the image which is written to a temporary file can be converted and imported again afterwards. There is an optional keyword {TEMPDIR}, which can be used to store additional files for a converter like IrfanView, if needed.
Based on the source image, all available options should be tried and the best result can be used directly. But it is recommended to manually edit/optimize the best result by the creator of the original image.
When using the -o option, a .lua-script will be generated (too), which contains the palettes and pixel data. The pal-variable will contain 136 palettes with 16 or 31 RGB-colors, each in hex (from 00 to ff). The gfx-variable will contain 240 x 136 pixels (color-index in hex, from 0 to f with 16 colors / from 00 to 1e with 31 colors). When using RLE (Run-length encoding) the palettes and pixels are stored in the rlp- and rlg-variable.
- Python (3.6, or greater) - https://www.python.org/
- Pillow (Python Imaging Library) - https://pypi.org/project/Pillow/
If missing, Pillow can be installed using pip (for Python3).
Linux:
$ pip install pillow
Windows:
$ python -m pip install pillow
mulTIColor is easy to use and its really simple to convert an image to multicolor-mode.
Just run multicolor using the provided example images:
$ multicolor.py piggie_power_without_the_price.png
Generates a preview of the converted image, will be shown by the OS's default image viewer.
Specify an output name to generate the lua-script and save the preview:
$ multicolor.py kittens.png -o smallcats
The preview is saved as "smallcats-mtc.png" and the lua-script as "smallcats.lua".
Use a different converter to reduce the colors of the image per line:
$ multicolor.py dryad.png -c iview -o resultsmyvary
Have a look at the provided config file "mtc.cfg" for the paths and binaries.
Switch the range to 31 colors per line to get a, hopefully, better result:
$ multicolor.py kittens.png -c iview -r 31 -o somemorecolors
Encode the pixel data as rle (run-length encoded) to save some space:
$ multicolor.py kittens.png -o smallerscript -m rle
Set a bordercolor to prevent streaking (will reduce colors per line to 15/30):
$ multicolor.py gameboys.png -o blackborder -b 0 0 0
$ multicolor.py --help
Usage: multicolor imagefile [OPTION...]
Generate TIC-80 multicolor images
mandatory arguments:
imagefile imagefile with graphicsdata (e.g.: .png, .gif, etc.)
optional arguments:
-c, --converter converter: pil (default), see mtc.cfg for more
-o, --output outputfile for multicolor values (.lua)
-f, --force force overwrite of outputfile when it already exist
-r, --range range of colors per line (16 or 31)
-b, --bordercolor bordercolor (R G B) to prevent streaking
-m, --mode mode to encode values: raw (default) or rle
-v, --version show version info
-h, --help show this help
The optional arguments are only needed if the default setting does not meet the
required needs. A specific name for the outputfile can be set (-o / --output).
The maximum range (-r / --range) of colors per line, can be set to 16 or 31.
To prevent colorstreaking in the border, a specific color (-b / --bordercolor)
can be set. WARNING: this will reduce the overall colors to 15 or 30 per line!
Mode (-m / --mode) to encode the pixel data via rle (run-length encoding) or as
raw, which is the default. To reduce the colors of the image per line, various
converters (-c / --converter) can be used. These can be configured in "mtc.cfg".
Additional defined are: "iview" for IrfanView, "magick" for ImageMagick.
examples:
multicolor imagefile.png
multicolor graphic.gif -o multicolor.lua
multicolor pixels.png -c iview -o mydata.lua
multicolor colorful.gif -r 16 -o only16.lua
multicolor border.jpg -r 31 -b 255 127 64 -o frame.lua
multicolor truecol.png -m rle -o compress.lua
multicolor logo.gif -o overwriteme.lua -f
- multicolor.py (the converter itself)
- images/piggie_power_without_the_price.png (example image (23960 colors), original by Bossman/Rift)
- images/kittens.png (example image (23704 colors), original by Evil/Accession)
- images/dryad.png (example image (17056 colors), original by Lycan/LNX)
- images/gameboys.png (example image (18538 colors), original by Lucius Kwok)
- images/replicaanxiety.png (example image (16 indexed colors), original by Decca/Lego)
- components/display16.lua (the display-routine for 16 colors per line)
- components/display31.lua (the display-routine for 31 colors per line)
- components/display.lua (the display-routine for 16 or less indexed colors)
- components/rle-decoder16.lua (the decoder when using rle-mode for 16 colors)
- components/rle-decoder31.lua (the decoder when using rle-mode for 31 colors)
- components/rle-decoder.lua (the decoder when using rle-mode for 16 or less indexed colors)
There is no warranty for the scripts or it's functionality, use it at your own risk.
The icon/logo consists of the Sweetie 16-palette © by GrafxKid & a Shredder clipart © by Clipartmax.
If you have any suggestions, bug reports or annoyances please report them to the issue tracker at https://github.com/RiftTeam/multicolor/issues
Development of mulTIColor happens at GitHub: https://github.com/RiftTeam/multicolor