This is the first of hopefully many posts about technical topics relating to SpaceChem. Is there anything in particular you’d like to know more about? Let me know in the comments!
Although the puzzles in SpaceChem are extremely open-ended with ridiculous numbers of possible solutions, there are a surprising number of similarities between solutions for the same puzzle. Exactly what are those similarities, and how do they vary between puzzles?
The images below were generated by taking up to 500 solution screenshots for each puzzle from our score server, averaging them together, and applying some post-processing magic* to better highlight the common features. Components that are brightly colored are more common in solutions, while components that are faded are less common.
* For those who are curious, the post-processing magic consisted of:
x’ = ColorBlend(Normalize(x^3), x*0.75)
6 Comments
WOW, this is so cool. Can believe how common I am.
i think it has to do with people using the solutions uploaded to youtube. im guilty of using some of them as inspiration as well.
What does the distribution look like for TOTAL symbols, cycles and reactors for players who have completed all levels?
I wonder what would happen if you did this for a level like Suspicious Behavior.
1) Guess wich levels are tutorials!
2) Grab/Drop v.s. Grab
3) Counter-clockwise preference for red, clockwise preference for blue.
I wonder what the difficulty curve for SpaceChem looks like…
Would probably be interesting.
One Trackback
[...] my previous statistics post I covered what the “average” SpaceChem solution looks like for each research assignment [...]