Skip to content

takes an image as input and uses psychovisual prediction for generating an X-scii representation. short: png2petscii

Notifications You must be signed in to change notification settings

bittorf/image-to-petscii-unscii-ascii-convert

Repository files navigation

=== idea ===

  • 2018 or 2017 breakpoint/revision
  • C64-X-Party 2018 erster code

=== screen === 320x200 = 40 x 25 chars = 1000 bytes (vermutlich aber mit weniger bits moeglich)

=== titten+links kommen hier hin ===

http://widerscreen.fi/numerot/2017-1-2/unplanned-blocky-puzzles-creating-petscii-for-the-commodore-64/ http://kofler.dot.at/c64/ http://www.ponomarenko.info/tid2013.htm https://images.guide/

http://metalvotze.de/content/videomodes.php http://c64.superdefault.com/ http://vectorpoem.com/playscii/howto_main.html https://kornel.ski/en/faircomparison http://www.kameli.net/marq/?page_id=2717 https://www.raquelmeyers.com/c64-petscii/ https://nurpax.github.io/petmate/

https://www.c64-wiki.de/wiki/PETSCII http://pelulamu.net/unscii/

https://stackoverflow.com/questions/5582778/writing-a-binary-file-in-shell-shell-awk

=== overview === use a fixed set of chars and convert an animation to it. with simple compression we should make it to 80.000 bytes = 80 frames = 3.5 secs.

=== steps ===

  • choose a frame and strip it to 320 x 200 monochrome.

    • convert image.jpg +dither -colors 1 -depth 1 image-mono.png && cp image-mono.png big.png
    • convert big.png -crop 8x8 parts-%03d.png
  • generate x,y.png with 8x8

  • have also all petscii chars in a raw format 8x8 -> png

    • convert big.png -crop 8x8 parts-%03d.png
  • loop {

    • choose a "random" set?
    • construct new image from random set:
      • for x in 1...40; do for y in 1...25; do compare dithered with all petscii-chars and get score ->dssim|butteraugli remember to best 3(?) chars or choose a random out of 5 best? done done
    • compare to whole dithered picture with all variations from above 40*25 !5 = ???
    • re-construct full picture
    • start (dssim + butteraugli) and get score for all permutations }

About

takes an image as input and uses psychovisual prediction for generating an X-scii representation. short: png2petscii

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages