The Omega Pac

It has been said that, once sufficient computation power is achieved, every member of humanity might be "ressurected". This is based on the idea that there are a finite number of possible human brains possible bound by human biology, and that each of those brains has (presumably) a finite number of states. With sufficient computing power, it should be possbile to compute all brains in all states - effectively creating a database of every human state of being possible. There are problems with this idea, the least of which is the meaninglessness of redundancy, as a great many of these "minds" will be utterly mad by the standards of the day, and that in regions of relative sanity, repitition will run rampant. However, in theory it is a way to create/recreate every human mind that ever existed.

As a sidenote, banking on the assumption that this is possible, a sufficiently powerful civilization will do this many times - from a purely statistical standpoint the Copernican Priciple (that you, the person reading this, are the most average example of your type of thing in the most average situation) implies that the chances of this event (where "this event" means ANY event) beign experienced by a real and actual biological human are smaller than it's chances of being the product of a simulation. This is also referred to as the "Teh m4tr1x has j00" fallacy (by me, but probably by anyone who hears it, too).

All of the proceeding paragraphs lead to this - if it is technically possible to run all possible instances of a given intelligence, why not go ahead and run all instances of a very simple algorythmic intelligence? Ms. Pac Man seems like a good candidate for "Tiplerizing" - unlike the original the ghosts actually respond to the current position of the heroine, and there are comparatively fewer states to run through. Running every official iteration of the Ms. Pac Man software through every possible gamestate and history shouldn't take more than a few years with current distributed computing techniques, possibly much less.

Sure, it's a silly waste of computation cycles, but it's a bleeding edge waste of computational cycles.