quote:Original post by ph33r
If you had a 100 doors with 1% lockpicking ability, I thought you would open one door, but I guess that is not the case.

As always with statistics, you have to be very careful with what you say.

If you have 100 doors with a 1% lockpicking ability, you would be most likely able to open one door. In other words, if you had 500 players with a 1% lockpicking ability try to open 100 doors each, most players will have succeeded only once.

For a big picture view of this, if you had 500 players try to open 1000 doors with a 1% lockpicking ability, and you recorded a histogram of how many people had N successes, you would see a Gaussian distribution centered about N=10.

quote:
Think about it; lets say that your method with the door ID''s ends up generating RAND_MAX.

Mithrandir: this is why it was suggested to also hash with the player ID--if one guy can''t get it, maybe another guy can. Also, if you want to be ''nice'', you could hash the day number to let them retry once a day, or whatever other factor you wanted to add.