The following warnings occurred:
Warning [2] Undefined property: MyLanguage::$thread_modes - Line: 49 - File: showthread.php(1652) : eval()'d code PHP 8.4.24 (Linux)
File Line Function
/inc/class_error.php 157 errorHandler->error
/showthread.php(1652) : eval()'d code 49 errorHandler->error_callback
/showthread.php 1652 eval



Login Register

Poll: what do now
You do not have permission to vote in this poll.
pray to the ghost of wade boggs
25.00%
5 25.00%
drink poison
20.00%
4 20.00%
dissuade kman from making threads when hes stoned
25.00%
5 25.00%
derelict
30.00%
6 30.00%
Total 20 vote(s) 100%
* You voted for this item. [Show Results]

Population Erasure Really Thrills Hitler thread
#14
im currently trying to correlate features of a jigsaw puzzle to the time it takes to do the puzzle

things that factor into it:
- size of pieces
- number of pieces
- shape of the pieces
- inter-piece color contrast (a puzzle of blank white pieces would have none of this)
- intra-piece color contrast (a puzzle where every piece is a different solid color would have none of this)



[Image: WgvEy3X.png]

consider this puzzle of a tree. the circled/shaded red area would have some intra-piece color contrast because each piece would have a bit of trunk and a bit of sky for example. the non-circled areas of just solid color/texture would have no intra-piece contrast because they're all the same color. the puzzle as a whole is pretty high contrast color-wise, even if it's simple. sometimes a puzzle has very visually distinct pieces but with very low contrast. look at these:

[Image: 16133867433_1062863bed_b.jpg]
bombardment of algiers - 9000 pieces
[Image: puzzle-2.jpg]
CMYK - 1000 pieces

ive seen it posted to weird jigsaw puzzle mommy blogs that when you double the pieces, you quadruple the time taken but this is just demonstrably false. i traced the claim back to a paper that was (iirc) just measuring a computer's ability to descramble squares which is far from what a human doing a jigsaw puzzle is.

ive been doing puzzles of different sizes and timing how long it takes me. here's a plot of those averages:

[Image: Juxh66g.png]

a 2000 piece, which i havent timed yet, should be a little under 14 hours for me. if you go by the double-quadruple estimate, it should take me more than two and a half days straight

of course, there's a few problems. the more i do the same puzzle, my times get faster. should i use the fastest time possible? should i force myself to go at a leisurely speed? i also dont adhere to any real strategy other than putting pieces that are alike together

in this way, a puzzle is treated almost like a set of many puzzles and then linked together. in fact, many of the largest puzzles are just mosaics of smaller puzzles:

[Image: 4005556178261-0.jpg]
[Image: 71U45YjvEHL._SX425_.jpg]

to this point, i wonder when a X piece puzzle is just two X/2 piece puzzles and the time taken might just be linear. also, you know what's bullshit? they mail you these giant, 50K piece puzzles in SEPARATE BAGS so that you can do them as individual puzzles out of the box and then just stitch them together later. 

this post brought to you by [Image: autism-awareness-magnet-fhautism.png]
[Image: Sd9744R.png]
Reply


Messages In This Thread
RE: Population Erasure Really Thrills Hitler thread - by Ben - 05-06-2019, 12:27 AM

Forum Jump: