I’m putting a pause on my hiatus for the release of the Pichu, Pokemon Trainer, and Isabelle amiibo, but more importantly the news which are the subject of this post. I don’t plan on coming back to Smash for a few more weeks.
About a month ago I wrote up a piece on the 20XX of the amiibo meta. The finality of the scene would basically consist of first selecting a most-optimal amiibo (in the piece, it was Bowser) and then editing the amiibo piece-by-piece to make it use only the most optimal moves. After all, amiibo are just spreadsheets containing move probabilities, so the final theoretical ending of the amiibo metagame would always end up getting them to use the moves we wanted them to use. That’s the irony of amiibo: we could just play the game ourselves and use the moves we want them to use, but instead we send these half-functional AI into battle instead.
So what would happen if we skipped over deciding the most optimal amiibo and just figured out how to edit their individual pieces of data to prune out the undesired moves and enforce the most desired moves?
Well…
User @Fudgepop01 on Twitter has done just that, and found out which bytes control which moves. He posted a demonstration video here:
This seems to be the equivalent of “designer baby” amiibo- is it better to raise an amiibo with traditional in-game methods? If edited amiibo are banned, how can you prove they were edited? What is the end result of the metagame if edited amiibo become common?
I won’t get too far into this topic just yet, as the information is still coming out, and the method is not widely available at all. We’re still not sure what the impact will be, and at this point most of this is left up to imagination. One thing’s for sure- this may change everything.
Amiibo Doctor 