Submission Text Full Submission Page
This game is a pretty simple game in which you control a helicopter by clicking to make it gain altitude and that's about it. It's an infinite game and the goal is to go as far as possible. Unfortunately, the game is poorly coded and frequently generates sections that are impossible to pass as they're tighter than the helicopter hitbox.
The official name of the game seems to be "Helicopter Game", as used in the page on the website named in beginning of the game (somehow still having this 20+ year-old game in its navigation bar).

Emulation

Note that I modified the ending to reduce number of frames in the end since the encode.

Comments

I did a run with the System Time 1 seed, but it turned out to generate an impossible section pretty fast (around 1000), which ended the game on a lower score than what you can find in RTA runs on youtube. I tried another seed, which allowed to go as far as 4636, which seems good enough for a submission.

Potential screenshots:

Room for improvement

This is a pretty low-effort TAS.
Improving this TAS would mean looking for better seeds. This would mean either:
  • TASing it manually for each seed to see how long it goes, which would be a tedious process
  • finding a way to automatically determine how far the first impossible section is. I'm not exactly sure how to calculate this. This is a geometry problem.
  • Somehow implement terrain generation to see the whole terrain without having to TAS the level, so you can see if a seed seems promising (or just patch the game to be invincible and explore the seed)
  • training a neural network on this very simple game so you go directly go to the hard sections and manually test if they're possible to pass (which is something I might do in the future... Don't mark my words).


TASVideoAgent
They/Them
Moderator
Location: 127.0.0.1
Joined: 8/3/2004
Posts: 17533
Location: 127.0.0.1
This topic is for the purpose of discussing #10201: zapkt's Flash Helicopter Game "High score" in 02:33.900
GJTASer2018
He/Him
Location: Stafford, NY
Joined: 1/24/2018
Posts: 407
Location: Stafford, NY
I'm skeptical of this submission being acceptable as it is right now, but instead it might make a "low bar" for a bot-based effort to have to beat...
c-square wrote:
Yes, standard runs are needed and very appreciated here too
Dylon Stejakoski wrote:
Me and the boys starting over our games of choice for the infinityieth time in a row because of just-found optimizations
^ Why I don't have any submissions despite being on the forums for years now...
Patashu
He/Him
Joined: 10/2/2005
Posts: 4146
Aw yis my childhood. If this beats all RTA records I think it's publishable, but the potential to push it further is obvious.
Puzzle gamedev https://patashu.itch.io Famitracker musician https://soundcloud.com/patashu Programmer, DDR grinder, enjoys the occasional puzzle game/shmup.
Editor, Player (183)
Joined: 7/10/2022
Posts: 41
Hmm there's a 11k RTA score. Perhaps this should be bot'ed then https://www.youtube.com/watch?v=ZiebMObNEi4 I'm not doing that in the short term though (although it would be interesting to work on libTAS to allow training neural networks without game patching)
TASVideosGrue
They/Them
Location: The dark corners of the TASVideos server
Joined: 10/1/2008
Posts: 2958
Location: The dark corners of the TASVideos server
om, nom, nom... sweet!