An update on my CPU Learning theory in Smash Ultimate

by Doc

I’ve long speculated that amiibo training affects the CPU of that character and their playstyle, in fact as early back as one year ago, and I had the theory prior to this post. I’ve done some A-B testing with them, in the post-8.0 update, and I’ve found a few unusual behaviors around the Ness CPU, and theorized around them.

I should make note that I’ve long had the theory that patches reset the CPU AI to some degree, so this isn’t something that I just came up with to further insulate my theory against criticism.

  1. CPU AI has definitely been updated with 8.0, and any console changes (if they exist) have probably been reset. Prior to the patch, my level 9 Ness CPU wouldn’t jump. He would only spam PK Fire and PK Thunder, and use Up smash quite frequently. If I had had the forethought I would’ve liked to get a replay of him in action. In 8.0, every Ness CPU engages in typical amiibo behavior. They’ll spam PK Fire on the ground and in the air, PK Thunder juggle up to three times at the ledge, and use Up tilt to Up air juggling. It appears Nintendo has noticed how we teach our amiibo to play.
  2. The possibility exists that CPUs might be dependent not on the profile of the user, but the console itself. It’s possible that in order to save space, Nintendo would only create one database to draw the “CPU training” information from instead of having one for each profile.
  3. I don’t think that amiibo behavior affects CPU play as much as I did previously. While it might have affected CPU behavior much more in the past, it seems to me that post-8.0, the CPUs are much more rigid. Granted, I haven’t been able to do much training since the patch, so the rigors of time may have not taken their toll yet. I do still believe that, at least pre-8.0, amiibo behavior changed the CPU.

I plan to lab out these changes to the best of my ability – I don’t have a separate Switch, but I’ve heard from many, many people outside the amiibo scene that they’ve had this theory for a long time too. I don’t think that so many people would have a theory like this if there weren’t some truth about it, but it’s so unspecific that it’s hard to prove.

I think at the end of the day we won’t be able to really prove or disprove this theory until the updates have stopped coming in and we’ve had several months to let our CPUs change without the possibility of resets.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s