A first look at the Hero amiibo

Super Smash Bros. Ultimate mod has recently surfaced that tricks the game into thinking a Link amiibo is actually a Hero amiibo! After managing to get the mod working, Supernova streamed Hero amiibo training on YouTube for the first time ever. Now, Hero’s amiibo is likely more than six months away, but we were able to study the character’s basic movements and come to a few conclusions!

Hero’s base AI seems competent enough. It uses all of its normals efficiently and even combos its up throw into a string of up aerials. Most interesting, however, is its usage of special moves. It rarely charges its neutral special (Frizz / Frizzle / Kafrizzle), but sometimes uses the uncharged version to keep enemies away. We haven’t seen Hero use its side special (Zap / Zapple / Kazap) very often; in fact, it used one uncharged Zap and that was it. Whether or not this is because the AI is hardcoded to steer clear of Zap is unknown. What is known, however, is that Hero is surprisingly proficient with its down special, Command Selection! It’ll open the menu occasionally and look through its spells. If none of the spells are appropriate for the given situation, it will cancel the menu without using anything. Hero likes to edgeguard with Sizzle, Bang, and Kaboom, and will often buff itself with Oomph or Psyche Up. It will even use Magic Burst as an edgeguard or kill move! It seems trainers won’t be able to influence which spell Hero selects; instead, it will be able to influence how often the FP opens the menu. Additionally, Hero’s AI is really smart when it comes to conserving MP. It rarely lets itself run out of MP and mostly relies on its normals. Overall, the community is really impressed with Hero. We’re looking forward to the inevitable release of the Hero amiibo, though sadly, it may be a long time before we’re able to get our hands on one.

If you would like to read more informational posts, please follow this link.


3 thoughts on “A first look at the Hero amiibo”

  1. Has the usual training method using Best of 5 been patched? It didn’t work nearly as fast for me since the last update

    1. Unfortunately yes it has. Now the “optimal” method of raising to Level 50 is doing one first to 5, turning learning off, and then doing whatever you want until it reaches Level 50.

Post a Comment