Timing ends when the parser says it does, and I am once again going easy on the encoders and judges with a recording made on an actual emulator. I seem to have a knack for picking RNG-heavy games with certain yet completely invisible potential improvements. I'm also using the comments template this time, because I am tired and lazy.
Completes one full cycle at maximum difficulty (DIP switch setting of 8 rather than default 4)
Loses lives intentionally to save time
Ignores rescuing the ditched pilots, because nobody cares about those LOSERS anyways, cya
Ends input early, thanks cheese!
Comments
This one was an easy one, barring me having to redo basically the entire thing because I missed out on a completely obvious timesave, that being destroying the stage boss by crashing into it and skipping the time-warp cutscene (we have the lives to spare). I chose max difficulty, because that's literally just faster and more fun, both to watch and to record, even if it means this gets put into alternative. The RNG in this game is both forgiving and inflexible- it will always spawn enemy planes in front of you with the exception of the last stage, yet the RNG seed takes your input to basically the very last moment until the enemies spawn. This means that prediction is largely limited to waiting until the enemy spawns offscreen so we can be already aimed when they become visible, and otherwise just shooting whatever we see as efficiently as possible. This is another instance where the rerecord count is complete BS, because hitting these enemies with the surprisingly restrictive firing angles requires a lot of fine adjustment, and thus a lot of scrubbing on the TASStudio piano roll; I wasn't keeping count, but just multiply it by at least 10 and that's more accurate.
As for potential improvements... I mean, yeah, probably? This is an RNG game, so it's a crapshoot in this regard as usual. Movement optimizations are probably less than you would expect, because enemy spawn rates are rather steady, and you can only turn your ship and make target so quickly; if you're killing everything that shows up on-screen in a timely fashion, then you're already doing WAY better than an RTA run, and just in general you're just doing fine for yourself. The homing missiles add slightly more required forethought, but not all that much. In terms of trivial timesave though, nope, and that means this fulfills the optimization requirement.
Since the faster you shoot the enemies, the sooner the boss spawns, killing them optimally is the main point of the game. Each time you shoot you fire 3 bullets, so it may look like shooting is not done very precisely, but this can't be helped (however it can be used to our advantage since there's more room for rotation adjustments).
Manipulating enemy positions is actually very easy. You shoot way in advance and then after shooting and before the enemy spawn, you adjust your direction to make the enemy spawn right where your bullet already is. Since enemies tend to get farther away from you as you rotate, you need to rotate in a different direction when the bullet is already around the edge of the screen.
108 frame improvement of the first level.
Link to video
I haven't checked if the frame the object spawns in memory has any relation to the frame it gets onscreen and to the angle (is it faster to spawn/kill them in corners or in the middle?). Neither did I check if delay between their spawns can be manipulated. But since your progress is displayed on the screen, it makes sense to compare different strategies on certain progress chunks. If the new chunk can't be accessed earlier anymore, that means the pace is good.
Once again I don't think this movie is optimal enough for publication, and attention needs to be paid to actually trying to kill enemies ASAP.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
