hacker.org

Hi again!

I found out an pattern - and I really think it's ok.
But If I calculate the population at generation 10 000 000 000 and type it into the answer box, it calls me "incorrect".

In my opinion this Challenge is searching for an incorrect answer.

I've tested my pattern to generation 1 000 000 and it's always ok.
I think there's a program which calculates the false population - and every solver only types in the false answer...

I think the makers of this challenge should check this challenge without the "false program".

Greetings
abc

Do you mean that the creator and all the 39 solvers are equally wrong and only you are right?

I don't know if there are particular cycles or super-pattern in this example, but I do know that self-replicating and self-referencing systems in general and class 4 cellular automata (which Life belongs to) can get very complex. (After all, Life is known to be a universal CA). So I would be a bit careful with reducing it to some apparent cycles (which doesn't mean that they don't exist). But I bet that not all 39 solvers used the same approach nor the same program. At least I used a suitable program to simulate it to the desired generation.

@tails: you will have pm soon

Hi again!

Now I solved the challenge.
My pattern and assumptions were correct - but my program code was incorrect.

@all: plz be careful with some life-programs - there's one that got an counting error...

Grettings
abc

gfoot: What kind of life program calculates 10.000.000.000 generations in a few seconds?

I started with life32, which is quite fast for the first million. But as the number of cells increases
all the time it's getting slower and slower. One probably needs to wait weeks or months for
the solution.
I doubt that I would be able to write a faster one.
Is there a faster program to look for? Or am I thinking the wrong way?
I also thought about finding out some periodicity in the generations... and did not succeed...
but maybe there is an easier way?

There is a fast program to solve it via brute-force, and surely you could also arrange some kind of formula instead of running the program.

I found a program that could calculate the generations extremely fast. The trickier part was getting it to stop exactly at the generation i wanted it too

This one was blowing my mind off, because I did not really wanted to find some math tricks here.
I was not aware that there are other ways beside just calculating one generation after the other.
Finally, I guess we found the same program then and managed to make it stop at the right
nanosecond.

Thanks for the hint!

Well it's open-source so you could always hack the code to make it stop at the right time. I did it by hand though, and yes, it was a bit tedious.

I prefer the intelligent approach of calculating the periodicity of the components it degenerates into and then working out the population at an arbitrary generation immediately, but I'm ashamed to say it's not what sprang to mind.

Has anyone already made this picture to a correct golly rle file?
I´m a bit disappointed to do it by hand, but I can´t manage to scale it to
be read right... =(

Greetings,
Kjellski

kjellski wrote:Has anyone already made this picture to a correct golly rle file?
I´m a bit disappointed to do it by hand, but I can´t manage to scale it to
be read right... =(

Greetings,
Kjellski

I did it with a little script to convert the image into a text file. Worked perfect

How many blobs are alive at Generation 0? I think I must have a tiny error somewhere but I can't find it!
--EDIT - Nvm, I found it

Choice a)