Game objectives

  • Emulator used: FCEUX 2.6.6
  • Uses glitches to save time
  • Takes damage to save time
  • Super-Attack in final BOSS

Comments

Stagedragonxyk(old)dragonxyk(new)Improvement
stageXX3036829931+437
  • Personal best reclaimed — forgive the lazy strat call on skipping stage-by-stage breakdowns. Total gains: 437 frames casually flexed.
  • Regarding the more in-depth mechanics of this game, the 1-player run have already provided detailed explanations. Due to my limited English proficiency, I can't fully articulate them. However, just because I don't mention them doesn't mean I don't understand. Moreover, some superficial mechanics that appear flashy don't actually contribute to improving the run's speed. I believe that aside from speed, there's no other criterion to measure a SpeedRun—isn't that right?

Stage by stage comments

<STAGE1-1>

Waste half of 2p'health through a Ultimate.Allocating time to charge the ultimate added a 4-frame delay.

<STAGE1-2>

Losing half HP in 1-1 and performing two ledge suicides to kill off Player 1 (1P) resulted in significantly reduced lag

<STAGE 2>

No route changes — just cleaner movement tech to shave off lag.

<STAGE 3>

While climbing to the rooftop, I delayed killing off 1P and escorted 2P to the long staircase to accelerate horizontal screen scrolling to the edge.

<STAGE 4-1>

<By manipulating enemy positions, we enabled 2P to effortlessly take out four enemies on the rooftop; optimized inputs also slashed latency significantly.>

<STAGE 4-2>

No route changes — just cleaner movement tech to shave off lag

<STAGE 5-1>

This stage's optimization is significant: By redirecting enemy aggro, the final two mech crabs were forced to attack via jumps instead of projectile sprays, drastically cutting animation delays and refining the speedrun route.

<STAGE 5-2>

Another stage with massive upgrades: During the final boss fight, both players opted for chain whips. By stacking the whip hitboxes with blade slashes, we achieved double damage output at max attack speed. Additionally, I preserved 1P's super for the finishing blow to trigger the replay timer cutoff early.

Other comments

  • Thanks to Xipo and Klmz the two biggest contributors to this game's speedrun records.
  • Shoutout to DeepSeek — the ultimate language crutch that bridges my broken English into precisely coded intent.

DrD2k9: Claiming for judging.
Side-note: While it sounds like it wouldn't have been allowed when the previous run was published, a 1-Player run is now an option due to rule changes regarding what's allowed in our Standard class. Below is the applicable rule.
Multiplayer: For games where you can control several character at once, up to 3 branches are allowed:
* Minimum players (usually 1)
* Maximum players (usually 2)
* Some other player count if it's faster than both of the above.
fsvgm777: Processing. McBobX is handling the encodes for this one.


TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 15969
Location: 127.0.0.1
This topic is for the purpose of discussing #9579: dragonxyk's NES Shadow of the Ninja "2 players" in 08:18.03
Reviewer, Skilled player (1242)
Joined: 11/18/2011
Posts: 362
Location: Morocco
Ayyy great to see this run updated! Gotta give it a watch in a bit. EDIT: A temp encode for this run. Was fun to see "2-players" but only "1-player" is used most of time xD. Yes vote! Link to video
I still learn more about English. https://www.youtube.com/user/McBobX100
I wrote:
Working is the best way to achieve goals in speedruning. Hardworking is a pain.
Active player (457)
Joined: 7/7/2007
Posts: 29
Location: China
Alright, considering my limited English proficiency hindered clear communication of the technical aspects in this RUN, which led to significant controversy surrounding my previous RUN(in 8:25) - and given that the game‘s core technical principles have already been thoroughly demonstrated in 1-player mode(NES Shadow of the Ninja "1 player" by TASeditor in 09:30.41 https://tasvideos.org/3153M) - I‘ll now focus specifically on mechanics unique to the 2-player experience. The core challenge in two-player mode lies in lag management, rather than mechanically applying 1-player run rules. Following single-player movement routes in co-op mode would generate excessive lag. This forced me to make strategic trade-offs: while pausing after landing (by either not triggering boosting or failing to obtain the odd rule before boosting) causes temporary stalling, it effectively prevents lag accumulation.While there are moments when I deliberately refrain from accelerating at 25-speed thresholds or sustaining momentum post-landing – actions that might superficially appear unrefined – this strategic restraint stems from frame-perfect LAG mitigation calculations. Every movement in this run constitutes a continuous technical balancing act: optimizing velocity while preemptively neutralizing lag through frame-specific input optimization, requiring constant evaluation of whether momentary acceleration gains justify potential LAG-induced trajectory instability. Someone said this:"Player 1 can double hit enemies in odd RAM addresses, while player 2 can double hit enemies in even RAM addresses, which is all bosses. And you can't hit an enemy with both players at the same time." It's wrong.This argument stems from a misunderstanding of the game's core mechanics. As I've ‌visually demonstrated‌ in my current run:‌Player 1 (1P)‌ can trigger double damage against bosses through precise frame-perfect inputs‌‌;‌Both players‌ can simultaneously execute double-damage attacks via coordinated hitbox manipulation‌‌.These mechanics allowed ‌100% double-damage consistency‌ during the final boss fight. While language barriers limit my technical explanations, the FM2 documentation provides frame-by-frame verification of these interactions. The implementation logic (hitbox priority, damage multipliers, and player collision flags) is thoroughly annotated and surprisingly accessible ‌Third‌, in 2-player run, the Y-axis vertical scroll rate doubles compared to 1-player operations‌. This system-enforced acceleration mechanism becomes particularly evident during Stage 5-2's elevator sequence, where the ascension velocity increases exponentially‌.Strategic applications of this paradigm extend across multiple key stages:‌Stage 1-2‌;‌Stage 2-1;‌Stage 3-1;‌Stage 4-1;‌Stage 5-1‌. Finally, I would like to clarify that the techniques demonstrated in 1-player run were actually publicly disclosed and mastered within the Chinese TASers community as early as 2008. However, not all of the showcased techniques are beneficial for speedrunning in the two-player mode. This is precisely why I refrained from employing these tricks in the 2-player run - they simply don't contribute to time efficiency improvements. It's rather ludicrous that some individuals have unilaterally assumed I lack mastery of these techniques.
The reminiscence of childhod. I love this game.
Post subject: Movie published
TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 15969
Location: 127.0.0.1
This movie has been published. The posts before this message apply to the submission, and posts after this message apply to the published movie. ---- [6461] NES Shadow of the Ninja "2 players" by dragonxyk in 08:18.03

1742985655