Just like DonutCannon said. Timing is the key here: You need a system which changes its state over time, so when you need the random number you just copy the state of your system at that specific moment. That way you get pseudo-random-numbers.
A little example:
i think a better way would be to have a number of clocks set up, and you sample them. that way, on average, half the output bits will be high. in the example you showed, most of them were off most of the time.
this is the one i built into my cpu:
a similar option is to have get a binary counter, run it as fast as possible, walk around for a few seconds, then sample it's output. the random factor being the time you walked for.
I think I may have come across one of the most true-to-random designs. I will try to upload a video soon.
It is fueled by dispensers which shoot either arrows a a block directly ahead or stone blocks into a stream of water with ice coating the bottom (to give the stone greater speed.) This speed enables the block to activate the plate for a few ticks (about equal to the press of a button). I then duplicated the design, adding one to each of the previous one's output. I managed to achieve eight total random outputs before it got too large.
I think I may have come across one of the most true-to-random designs. I will try to upload a video soon.
It is fueled by dispensers which shoot either arrows a a block directly ahead or stone blocks into a stream of water with ice coating the bottom (to give the stone greater speed.) This speed enables the block to activate the plate for a few ticks (about equal to the press of a button). I then duplicated the design, adding one to each of the previous one's output. I managed to achieve eight total random outputs before it got too large.
The dispenser method has been popping up many times recently, and is quite a compact way of making very fittable randomization. The other most popular method is to have a series of clock with odd delays so that they have very oddly dispersed multiples, which leads to an output which appears to be random. I recently made a sort of slot machine which involved a constantly running clock and a pressure pad, both of which where hooked up to an AND gate. Basically, to make a signal go through, one had to step on the pressure plate just as the clock triggered the and gate. Since the clock was hidden from view, this is sort of random. I prefer to call it skill-based, though, to be accurate.
i think a better way would be to have a number of clocks set up, and you sample them. that way, on average, half the output bits will be high. in the example you showed, most of them were off most of the time.
this is the one i built into my cpu:
a similar option is to have get a binary counter, run it as fast as possible, walk around for a few seconds, then sample it's output. the random factor being the time you walked for.
minecraft has lots of space for stuff. use MCedit, and put it underground.
It is fueled by dispensers which shoot either arrows a a block directly ahead or stone blocks into a stream of water with ice coating the bottom (to give the stone greater speed.) This speed enables the block to activate the plate for a few ticks (about equal to the press of a button). I then duplicated the design, adding one to each of the previous one's output. I managed to achieve eight total random outputs before it got too large.
This sounds like what I'm looking for!