Live data from Hacker News

Show HN: BreakLock – A hybrid of Mastermind and the Android pattern lock

maxwellito.github.io

11–20 of 40 posts

Re: Show HN: BreakLock – A hybrid of Mastermind and the Android pattern lock

#11

Maxwell, you have a spelling mistake on the Success page. "Lock found in 20 attemps. You must hate this game by now."

Dammit! Good catch! I'll fix it tonight, once at home. Thank you!

PS: Pull requests are welcome ;-) https://github.com/maxwellito/breaklock

Re: Show HN: BreakLock – A hybrid of Mastermind and the Android pattern lock

#12
post #3

Solved the first puzzle on "easy" in 53 attempts. The first 20 attempts were me figuring out that the 4 dots beneath each try are not in the same order as I drew the pattern, so I can't directly deduce if a single dot is correct or not. Then it took me another ~8 attempts to find the four correct dots and then a very long time to get the order right. Fun game, interesting idea and nicely done! I'll probably play a fe…

52 attempts first here, then 25 then 20... exactly the same thought process.

Re: Show HN: BreakLock – A hybrid of Mastermind and the Android pattern lock

#15
I saw this game (Mastermind or originally called Moo or Bulls and Cows) in an article in Software Practice and Experience[1] in the 70's. A couple of years later I saw a very clear explanation of how to write an effective software opponent, written by Knuth[2]. It was a lot of fun to implement back in those days (first in Fortran and then in Pascal). I recommend the easy paper by Knuth.

It's surprising how effective Knuth's algorithm is. If I recall 4 or 5 guesses to solve 4 digit mastermind.

[1] Computer Recreations, Software – Practice & Experience 1.2, 201-204 (Apr-Jun 1971) [2] D.E. Knuth, The Computer as Master Mind, J. Recreational Mathematics Vol. 9 (1976/7)

Re: Show HN: BreakLock – A hybrid of Mastermind and the Android pattern lock

#17
post #15

I saw this game (Mastermind or originally called Moo or Bulls and Cows) in an article in Software Practice and Experience[1] in the 70's. A couple of years later I saw a very clear explanation of how to write an effective software opponent, written by Knuth[2]. It was a lot of fun to implement back in those days (first in Fortran and then in Pascal). I recommend the easy paper by Knuth. It's surprising how effective…

This is harder than Mastermind because you're not allowed to repeat colours/numbers in your entries.

Interestingly, I find I'm using some similar logic to what I use playing Minesweeper. "Three of this quartet are in the pattern, one of this quartet is in the pattern, they overlap in two places, therefore the two non-overlapped points in the three-scoring quartet must be in the pattern."

First try took 19 attempts, second try took 13 and I think I can do better.

Re: Show HN: BreakLock – A hybrid of Mastermind and the Android pattern lock

#18
post #3

Solved the first puzzle on "easy" in 53 attempts. The first 20 attempts were me figuring out that the 4 dots beneath each try are not in the same order as I drew the pattern, so I can't directly deduce if a single dot is correct or not. Then it took me another ~8 attempts to find the four correct dots and then a very long time to get the order right. Fun game, interesting idea and nicely done! I'll probably play a fe…

52 attempts first here, then 25 then 20... exactly the same thought process.

gave up around 15 tries on my first attempt because I refreshed to review the instructions again but my 2nd try on easy I did it in 7

Re: Show HN: BreakLock – A hybrid of Mastermind and the Android pattern lock

#19
Great time waster, worth making lots of guesses just to see the performance reports. My first go took 65 guesses which included working out the use of the 4 dots, the game appreciated my determination. my second go was 3 guesses but I admit I got very lucky there.
Post reply on HN