I haven't worked on it at all for 4 years. The last thing I ever did for the game was making a 2nd WIP that didn't even make it past the "this is GOLDWOOD FOREST SAVE THE TRIBALS cutscene. All I've done in glitch hunting since my old WIP was finding the fire-weapon-switch-spam trick, and jumping up into the darkened ceiling of doorways to load the next room faster. Even though I think it'll be a very good finished TAS product, here's my reasons for not working on it:
1. The overall level of optimization for 3D N64 TASing is set high with Super Mario 64. For a 4+ hour long run, I know that there could be improvements in lots of places, simply because for three dimensional movement, there's always some faster way to do a room. I'd be forced to write it off as "contains speed/entertainment tradeoffs" and that wouldn't be not ideal. There are so many ways that a complicated 3D shooter game like this can be optimized to the extreme (my old WIP is piss poor, every single room can be done with a much higher level of optimization, I could probably save minutes if I were to redo it), and I don't want to dedicate years to a project like that. Add that to the extra element of a second character that can help you kill stuff faster, and suddenly the optimal way to kill all enemies the fastest becomes even more complicated.
2. It's a big task (the final movie will be over 4 hours), and I kept looking for a TASing partner to help me with it, but never got one.
3. Now that I'm out of college and on the (painful) hunt for full-time employment while at the time juggling part-time work and an internship, I barely have enough time for my other hobbies, let alone TASing (it's slow, slow goign for my Mischief Makers Gold Gem TAS). That old WIP took me, I think, 4 months to make that 1+ hour progress, and I would go way more slower with the level of optimization I know now as well as way less free time. To compare, it took me over a year to finish Mischief Makers, and that's when I had a lot of free time on my hands.
4. The biggest reason is something I mentioned in the recent N64 Castlevania submission: the optimal fastest way to run in this game isn't a straight line. When you run with the joystick and use the left/right C buttons to strafe, you run faster, and it's possible to run in a straight line, but that straight line is actually a very long curve. You don't go in a 100% straight line like in, say, Goldeneye, rather you take a curved route. This sort of movement makes optimization over long distances insanely impossible, as when I first started TASing this game I'd just move the analog into one direction and tested to see if I'd reach my destination, not suspecting that I wasn't going as the crow flies, which in a finished TAS is extremely sloppy. On a minute, frame-by-frame scale, I'd have to keep adjusting the analog stick.
It's too bad, because I really think this would have been a very entertaining movie despite it's long length. If anyone wants to ever take this on, please feel free, I may even help out where I can.
