April 27, 2017, 05:00:10 am
0 Members and 1 Guest are viewing this topic.
The TM47 discovery was a semi-joint discovery by both entrpntr (Pokemon Speedrunner, although he does more routing stuff) and I. We knew that TMs and HMs could execute code from invalid addresses, so we went out and tested each and every one. Out of all the TMs and HMs, only 3 stuck out:TM47: This item jumps to this subfunction , which increments the wOverworldMapAnchor. While using this only visually affects the map, reloading the map or crossing a connection applies the map changes to the loaded metatiles, allowing one to slowly clip through walls and access unintended parts of the map.TM36: This item jumps here, which is in the middle of the Poke Ball function. While this could hypothetically allow you to catch whatever was the last loaded enemy mon, unfortunately the b register (used to store catch rate) is always 0 when entering this function (both through an invalid pocket and through select menu), so you'll end up with a 1/256 chance of catching the enemy mon, or 11/256 if the mon has a FRZ/SLP status.HM03: I just found this out today while trying to figure out the interesting TM efffects, and it has a very useful application. This jumps here, which is in the middle of the function called when withdrawing the stored daycare mon from the daycare lady. The function doesn't check whether a daycare mon is present (or clears the relevant data either, as it wouldn't be needed), so you can continually use HM03 to clone the daycare lady's stored mon. The only stipulation is that the new moves learned (by calculating level based on raw experience, which may not reflect the current level as the daycare step function does not recalculate level for every exp point gained) is based on the last loaded Pokemon, in 0xd004.
Page created in 0.079 seconds with 17 queries.