23 Jun 2011 dangermaus   » (Journeyer)

of mice and men

we just solved the Cavern challenge on the hacker academy. our bash warrior fought and died several times deep down into the dungeon, but a final modification into his mind brought him in about 16 hours down to the diabolic beast at the bottom. our warrior slaught the beast in two moves :-) we did not think someone could create such a deep dungeon!
dungeon's challenges are our entry to the more complicated challenges that require games to be solved by automation.

digging deeper

our "challenges solved" counter scores 102, but now the next challenges require us to invest much more time. we solved the first 90 challenges in two weeks, and then it took one week to solve another ten. this week we solved only one or two... also, we need to dig deeper into our toolbox. in particular, we start performing statistical analysis on ciphertexts (like done on one of my previous jobs)... i also printed out the Python code of the HackVM to learn its inners. and we need to decompose cellular automata into its components to know the number of cells at the 10th billion generation...

a life worth living

the Conway's game of life challenges are interesting, too. and make me feel younger, as my first Life implementation in BASIC for C64 was around 1989, after reading some old articles in Scientific American. i still remember the flash experience i had; my life simulation was not behaving as expected: only when i introduced a copy of the lifeboard for the previous generation, things started working out :-). with my father, we did another implementation in Turbo Pascal for Windows 3.1 back in 1992. the third implementation i did was just some days ago in Freepascal to solve the 'life skills' challenge. but in all my implementations were on bounded universes.

with please, i then noticed the Golly project, a simulator for Cellular Automata including Life. Golly has an unbounded universe, still it is very fast when the lifeboard size is small and can adapt lifeboard size if cells grow out of the box. and it is amazing feeling, as there are plenty of patterns in Golly and even a complete Turing machine working on the Life universe. it is amazing because i remember reading in Scientific American about it and how it theoretically should be possible to construct it in Life, as some NAND patterns were already working when simulated on a big supercomputer. 20 years later, i can run a complete Turing machine on the Life universe on my old laptop :-D.

in Scientific American, i read also about Life in 3 dimensions, and i wonder if there were progress on this as well. but probably not, as we 3D beings we think best in 2D. or at least me...

thinking different

all this time spent on the hacker academy is producing some chemical changes in my mind: i start looking at the detail, giving importance to an end of line or looking at ranges were a function is presumed to work. errors get more important than all the rest! and evidence just becomes obvious background noise... too much evidence can become even a misleading trail.
numbers previously thought big get small like a quarter hour or like four billion keys XORed on a ciphertext. numbers never considered before (because greater than the universe's atom count) are solution to challenges...
writing use-once code and combining tools and operating systems... burning CPU time, network bandwidth and hard disks... in short, hacker.org is wonderful!

Latest blog entries     Older blog entries

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!