Submission #4791: nitsuja & Alyosha's NES Kabuki: Quantum Fighter in 09:14.19

Nintendo Entertainment System
baseline
FCEUX 2.2.2
33306
60.0988138974405
38654
Unknown
Kabuki - Quantum Fighter (USA)
Submitted by Alyosha on 8/8/2015 1:57:01 PM
Submission Comments
Kabuki Quantum Fighter is unique platformer with interesting swinging, weapon, and hair mechanics. It is a particularly difficult game to optimize since there are many small tricks and health and weapon energy management need to be planned out in advance.

Game objectives

  • Emulator used: FCEUX 2.2.2
  • Beats the game as fast as possible

Comments

Improving Nitsuja's run was the toughest I have faced so far. For a long time I was stuck on the first level unable to improve anything. Only recently did I get by it and the also challenging second boss. After that the run came together pretty quickly. Most of the improvement though comes down to the last boss fight and different energy management.
Originally I thought that Nitsuja's energy management route was the best. This is because I couldn't find convenient enemies to recover the extra spent energy. When sack_bot asked in this submission thread and I actually wrote out the math, I suddenly realized that I didn't have to save all the energy to save frames. I knew 1 enemy I could get a big refill from and still save time. In testing, I discovered another enemy that I thought didn't drop anything was actually a random drop (the blue teeth enemy at the first part of level 5.) This was enough to make the complete energy depletion route feasible. This whole idea was originally discussed a while back by Nitsuja and others, well, theory confirmed!

Stage by stage comments

Level 1

I improved 1 frame right at the end. Nitsuja hits down at the last jump which costs him a frame. I had known about this for a while, but since it through off level 2 enemy patterns I couldn't implement it. Randomly adding a jump at the start of level 2 fixed it. This fix would be needed for other levels as well so it was fortuitous that I waited to figure it out.
Also in level 1 I burn the extra energy that was previously carried throughout the other levels.

Level 2

In the second part of level 2 Nitsuja misses (or purposely doesn't use) a 6 frame improvement by moving early. I implemented this but then it ruined the boss pattern, which was the fastest possible as far as I could tell. I eventually got enough of the fight to sync to finish it in an equally fast way but needed to take 1 hit of damage. I make up for this in a level 3.

Level 3

Nothing new in the level. I save one hit on the boss.

Level 4

Nothing new in the level, but I save a hit on this boss also, as well as a few frames. I'm not sure why Nitsuja originally got hit here.

Level 5

The hit I saved on level 4 boss gave me a free damage boost in the first part of level 5. I recover one big refill right at the start. I recover the other 2 in the second half of the level. You might notice that I spent 8 energy originally but got back 9. This actually saves time on the last boss, giving me one more triple shot to use.

Level 6 (Final Boss)

The rest of the savings come from the much different boss fight. It turns out that beating the first phase with a more optimal pattern and then beating the second phase in 2 cycles is faster by a couple seconds then Nitsuja's approach, which was to set up the second phase pattern to be beaten in one cycle, but being much slower on the first phase.

Other comments

Hurray its finally done! This was pretty challenging and I'm glad its over. Special thanks to sack_bot for making me think a little deeper into this TAS. In the end I am happy with the result, although there is a slim possibility that a much quicker first phase pattern of the last boss can be found such that the second phase can also be beaten in one cycle. This would save another 2 seconds I think, but all my attempts to find such a pattern have failed.

ars4326: (replaced movie file w/ Alyosha's 117 frame improvement) Judging!
ars4326: Hello again, Alyosha. Another nice improvement to a run which already had some serious optimization! The improved final boss fight, and the superior energy burnout strategy, were immediate standouts compared to nitsuja's publication. Cross off another one on the 'First 500' list. Good work, once again!
Accepting as an improvement to the published run!
feos: Pub.
Last Edited by adelikat on 10/14/2023 2:29 PM
Page History Latest diff List referrers