Post favorite algorithm, go!
Post favorite algorithm, go!
Other urls found in this thread:
en.wikipedia.org
youtube.com
en.wikipedia.org
algo.inria.fr
en.wikipedia.org
people.eecs.berkeley.edu
twitter.com
heap sort
...
heapster sort
if (a > b && b < a)
Ok, now I am intrigued, 3 IPs, all calling heap sort, why?
Also, op here, Möller–Trumbore triangle intersection is my choice.
Wouldn't be fair to have all you chaps tell me yours without showing you mine.
>feminist frequency
yeah, you can fuck off to stack overflow, reddit or wherever is not here.
The algorithm that sorts randomly, and if the assortment is correct then it ends
A* search
sleep sort?
it's the best most inefficient way to do anything.
Looking at
en.wikipedia.org
What is cross? The external product?
I posted the gif, it's like some kind of anime move
HIDDEN TECHNIQUE:
HEAP SORT
Simplex
>What is cross? The external product?
as in cross product, yes.
it's still one of, if not the fastest ways to do non-proprietary triangle intersections.
Shor's algorithm.
Ayyy just gimme dat (You)
simple but very useful
>d^2=(a2-a1)^2+(b2-b1)^2+(c2-c1)^2
Haha yeah that one.
I've always found the Cholesky decomposition interesting
Yes. According to wikipedia, the cross product is the external product.
Giving someone name to that algorithm is like the "relation de Chasles" (Segment addition postulate). We shouldn't give people name to such trivial task.
I haven't decided which one is my favourite, but I remember that I really liked Particle Swarm Optimization.
sleep sort
Oh, and Fuzzy C-Means!
Bubble sort
xor swap.
useless and obsolete nowadays, but when i saw it the first time it blew me away.
Weak quickheap sort.
I think there was a variation that added another similar word to the beginning, but I can't remember it.
One Time Pad
Very hard to use irl tho
Yes, but I like the fact that it's mathematically proven to be secure and that it's simple.
True
Hashlife
Comparing the distances of two objects and only having to compare dot products, saving myself a square root.
Anything involving simd. AVX in particular
FFT multiplication is useful and cool as hell.
draw a circle of radius r (pixels)
count pixels inside circle
divide by r^2
---
approximation for pi
Currently implementing a very fast merge sort using AVX. How rich am I gonna get.
Cross product. Ray marching. Euler method integration.
0x5f3759df
recursive descending parsing
>google this
>it's a link i've already visited
slightly salty about this
Shunting yard, because it's the progenitor to operator-precedence.
Also are you using this thread to find out useful algos, OP? Or are you just curious?
Here's one, twos compliment.
...
The only way to sort your shit, figuratively and literally.
Currently making a git and write-up of a string/generic-array reversal algorithm that beats std::reverse with a speedup of x12 on large arrays and is only marginally slower on smaller arrays due to slight overhead
>someone put in the time to make that
...
Kruskal's algorithm
Floyd :^)
xorshift+, literally 4 xors, 3 shifts and one add for 128 bits of quality pseudorandomness.
That's more or less how pi was originally calculated, before everyone figured out the infinite series.
Hyper log log counting.
algo.inria.fr
en.wikipedia.org
Such a fascinating idea.
super useful in time series statistics. big fan
Tarski fixed point theorem. While the algorithm to find the fix point is very simple, the power of doing so is incredible.
Where as determining if two regular languages are equivalent is done in exponential time, while determining bisimularity can be done in polynomial time.
I recognize C++ when I see it
Streaming algorithms are neat.
// what the fuck?