Submission #4423: PoochyEXE's N64 Magical Tetris Challenge "maximum score" in 01:57.38

Nintendo 64
maximum score
(Submitted: maximum score)
(Submitted: Magical Tetris Challenge (U) [!].v64 USA)
BizHawk 1.7.0
7043
60
559
Unknown
Submitted by PoochyEXE on 9/5/2014 5:38 AM
Submission Comments
Thanks to Spikestuff for the encode!

Introduction

This obsoletes my previous run, beating it by 5,666 frames (1:34.43) and by 1:14 in-game time. I have another (now cancelled) submission which is faster by 123 frames and 5 seconds in-game time but uses Easy instead of Expert and a significantly different route. This is intended for Moons and possibly Notable Improvements.

Game objectives

  • Emulator used: BizHawk 1.7.0
  • Aims for fastest time to 999,999,999
  • Uses hardest difficulty
  • Genre: Puzzle

Comments

As before, luck manipulation was still the hardest part. Additionally, the Expert AI does some luck manipulation of its own, stalling until the piece in the shared next box is one it wants then dropping its current piece immediately on the next frame. It's possible to counter this with some stalling of your own, since the AI still needs to wait the 31 frames of ARE before it actually gets the piece.
As with the Easy difficulty submission, I exploit the fact that line clears appear to advance the seed, as well as the hard drop + rotate on the same frame.
I've moved my comments on the game mechanics to a separate game resources article.

Stage by stage comments

Stage 1

This time around I pull off a 17 combo for a x3000 multiplier. Since the player is saddled with a longer meter on Expert, I can safely do a 16+ combo without prematurely triggering the meter. However, this took a lot more luck manipulation and I had to take one pentomino to make it happen, which made the number of blocks on my playfield odd. This made it impossible to get an All Clear until I did something to make it even again. Thankfully, the meter is always at an even height on Expert, making it easy to make it leave behind an even number of blocks. I wasn't able to luck-manipulate my way into 2 consecutive I pieces before the meter triggers an I piece rush, so I settle for an L and a J, which combine with 3 additional I pieces from the rush to create a 2x10. A bit more luck manipulation gets me two more All Clears, which is just barely enough to push the bonus over a billion.
The stalling leading up to the third All Clear is not for luck manipulation, but rather to ensure the AI dies. Whenever either side clears a single line, they get a number of "charges" equal to the current intensity level (displayed in the clock in the top center). These go towards countering magic pieces first, then any left over go into the meter. I had to intentionally stall to avoid giving the AI "!" pieces, which can't be countered, so that the AI's single-line clears will counter my attacks instead of filling up the meter. I do this just enough so that the AI dies one charge away from filling its meter.
Then I build a crude "TAS" out of the blocks while waiting for the AI to die.
Score for stage 1:
  • 13,146,450 from line clears & combos
  • Bonus subtotal: 337,000
    • 35,000 points for time
    • 2,000 points for 2 countered magical pieces
    • 300,000 points for 3 All Clears
  • Bonus multipler: x3000 for 17 combo
  • Bonus: 337,000 x 3000 = 1,011,000,000
  • 1 point for clearing the stage (added during the stage transition, but not displayed since I had already hit the cap)
Total score for stage 1: 1,026,146,451

Stage 2

Not much to say here. I intentionally top out as soon as possible, since the in-game time only stops when you lose. Then I input "TAS" for initials and end the movie.

Other comments

As with my other submission, there may be room for more optimization, either via better luck manipulation or by finding a way to make the AI die a couple seconds earlier on stage 1 (but not too early, since you have to rack up a bunch of All Clears first).
I still wanted to build a much tidier-looking "TAS" at the end, but I wasn't quite able to luck manipulate my way into just the right series of pieces.
Screenshot suggestion: Frame 5800 showing the "TAS" I built out of blocks.

FractalFusion: Accepting as an improvement to the published run. In addition, this run, which was done on the hardest difficulty, was chosen in favor of the Easy mode run.
Spikestuff: Hey, I still have the files... might as well get a quick publish out of the way. Publishing.
Last Edited by adelikat on 10/12/2023 12:37 AM
Page History Latest diff List referrers