karl e832407ea0 Mostly revert ef01f37
That commit message was a lie, it did not improve performance but actually reduce it significantly...

I had accidentally added a lot of complexity due to the std::list data structure. this approach might seem more redundant, but it's much faster with rectangles (like 100x faster) and doesn't seem to make a difference in the other cases.
2020-12-07 00:17:20 +01:00
2020-11-18 15:33:47 +01:00
2020-11-30 20:11:16 +01:00
2020-11-17 12:39:20 +01:00
2020-12-06 22:09:35 +01:00
2020-12-06 22:59:01 +01:00
2020-12-07 00:17:20 +01:00
2020-12-05 22:29:28 +01:00

quickhull

dependencies

for the visual graphic display SFML is needed (https://www.sfml-dev.org/download/sfml/2.5.1/) if you want to compile the program on your own, the deployed release uses static linkage.

program arguments

-v                   mode with enabled visualization
-akl                 using Akl-Toussaint heuristic beforehand
                     https://en.wikipedia.org/wiki/Convex_hull_algorithms#Akl%E2%8
-f <filename>        read random numbers from file
-rngcount <count>    generates count random float values
-rngmode <mode>      if generating float values this specifies the mode:
                     1. random numbers in screen range
                     2. numbers are ordered as rectangle
                     3. numbers are ordered as circle around screen center
-runs <count>        runs performance mode multiple times
-stepsize <size>     time between autosteps; if 0 or not provided -> manual steps
Description
No description provided
Readme 1.2 MiB
Release Latest
2020-12-08 13:25:53 +01:00
Languages
C++ 95.2%
C 2.8%
Makefile 2%