TASVideos

Tool-assisted game movies
When human skills are just not enough

Submission #6439: ktwo's NES Castelian in 13:58.29

Console: Nintendo Entertainment System
Game name: Castelian
Game version: JPN
ROM filename: Kyoro-chan Land (Japan).nes
Branch:
Emulator: Bizhawk 1.13.1
Movie length: 13:58.29
FrameCount: 50380
Re-record count: 12616
Author's real name: dc
Author's nickname: ktwo
Submitter: ktwo
Submitted at: 2019-06-25 19:58:47
Text last edited at: 2019-07-03 06:28:02
Text last edited by: ktwo
Download: Download (10812 bytes)
Status: published
Click to view the actual publication
Submission instructions
Discuss this submission (also rating / voting)
List all submissions by this submitter
List pages on this site that refer to this submission
View submission text history
Back to the submission list
Author's comments and explanations:

Game objectives

  • Emulator used: Bizhawk 1.13.1
  • Primary objective: speed
  • Playing on the fastest combination of version (JP) and difficulty (Hero)

Background

This game was the first I made a TAS of before having completed a console speedrun of it first. This meant I continued playing the game after submission 6194 and, not completely unexpectedly, I came across a few improvement ideas. This new submission is just so I can have peace of mind again and be able to say that all known time savers have been implemented.

Game mechanics

Not much to add in this section. See 6194 for a description of some aspects of the game mechanics. I've also created a wiki page with some additional information: https://kb.speeddemosarchive.com/Castelian_(NES)/Game_Mechanics

Improvements

There are only two minor gameplay improvements in this submission. When trying to sync the rest of the run, there were also quite big swings in the "random" lag frames. Below is a summary of what has changed since the last submission, both in terms of the gameplay improvements, but also the "random" lag. "Random" lag and screen transition times between levels can vary significantly, but for reasons unknown to me. So it's essentially a source of randomness that I haven't been able to control other than through some far from non-exhaustive trial-and-error.
  • Instead of jumping at frame 6035 (1:42 in the encode), the first jump was delayed to the next step. This triggered an earlier spawning of an enemy above and 18 less frames of waiting to jump over it.
  • With a well-timed shot, the bouncing ball was delayed enough to make it possible to enter the opening at the start of the third tower. This saved 6 frames of countdown at the end of the level (see the previous submission for the game mechanics behind this). 1 frame was however lost at the level entry screen to manipulate this trick to be possible.
  • 2 frames lost in tower 3 to "random" lag.
  • 31 frames saved by shorter screen transition to bonus 3.
  • 20 frames saved by shorter screen transition to tower 4.
  • 6 frames lost in tower 4 to "random" lag.
  • 41 frames saved by shorter screen transition to bonus 4.
  • 5 frames gained in tower 5 from less "random" lag.
  • 1 frame lost in tower 6 to "random" lag.
  • 17 frames lost from longer screen transition to tower 7.
  • 3 frames saved in tower 7 from less "random" lag.
  • 38 frames lost from longer screen transition to bonus 7.
  • 16 frames lost in tower 8 to "random" lag.

In total, an improvement of 18+5-2+31+20-6+41+5-1-17+3-38-16 = 43 frames.


ThunderAxe31: Judging.

ThunderAxe31: Accepting as improvement over the current movie.

feos: Pub.


Similar submissions (by title and categories where applicable):