Unfavorable-Semicircle-Wiki

Investigation into color data in SN visuals

summarized from Discord conversation 2017-05-18

technical work by extra; ideas + transcription by unfavorablist

Concept

Can we extract data from the shifting “histogram” images in Stabilitory newing videos?

The images have fifty “pillars” or columns that can have a height from 0-35.

Method

extra assigned the heights in the “pillars” alphanumerically, A-Z and 0-9. (36 unique values)

Here’s each unique frame from BELT 1 with each line being one frame: https://pastebin.com/raw/X82VLNA9

This data can be saved (in a text editor) as a .data file, and that file can be opened in GIMP (or other image editor).

The image rendered by GIMP reveals a certain pattern:

This pattern exists in a dataset derived from RATE as well.

Adjusting the width to 572 gives this:

Is this random or coincidental?

To compare, I grabbed a random book (in plain txt format) from the Gutenberg Project, and saved that to a .data file. When I opened that in GIMP, it looks like this:

Follow-ups / next steps

Doing this with all of SN’s BROTHER seems like a reasonable step.

extra’s script

here: https://hastebin.com/ficubivoze.py

• uses python 3 and all you have to do is “python a.py "
• it’s probably the best idea to download in webm format. also duplicate images are handled so you don’t have to worry about that
• shl is the y position of the always-there horizontal line, I dont think it ever changes