Well, generations don't have to be rendered in realtime, so you could just grab some high cpu instances on AWS and compute really large sizes. For example, this

Well, generations don't have to be rendered in realtime, so you could just grab some high cpu instances on AWS and compute really large sizes. For example, this guy made a turing machine, and it's pretty huge: http://rendell-attic.org/gol/tm.htm http://rendell-attic.org/gol/tm.htm And that was in 2000. I'm sure we could do extremely large boards now.