rta_investigation

Speedrun investigation stream

This stream is to make a record of some of the various things I've worked on for speedgaming for games I've played or contributed to in some way. Most of the discussion has to deal with some sort of programming approach, data mining or otherwise intensive research to speedrunning for the games I've interacted with. "Simple" things, like finding faster strats by using a timer, aren't part of this - the examples I want to talk through are more about uncovering unknowns, understanding mechanics better, or identifying strats that are otherwise hard/inefficient/near impossible to do without a method to take some mysterious process and to make them RTA-friendly.

I've worked on quite a few things like these projects over the years, and I've felt like I never got a good chance to explain how the process worked to anyone. So I decided to just have a day where I walk through a number of examples, both for anyone else to hear and for myself in the future, sort of as a documentation of work. Me personally, I'd be very interested in hearing other examples from other people & communities of similar stories, because I know they exist- but often are fairly arcane and difficult to understand.

Most of the examples are for RPGs, in particular for Dragon Quest games, but there's a few other examples. The narrative around this discussion is to show off a bunch of varying examples, including data collection and analysis, investigating RNG manips, and assembly level breakpointing to figure out mechanics.