Submission #8821: nymx's C64 Pegs "maximum score" in 00:25.54

(Link to video)
Commodore 64
maximum score
BizHawk 2.9.1
1528
59.826089499853765
0
PowerOn
1986-12.d64
Submitted by nymx on 1/3/2024 2:32:17 AM
Submission Comments

Pegs (Compute's Gazette)

"Pegs" is a game that has been around for decades. You've probably played it when you've eaten at Cracker Barrel, where you try your wits against that golf ball tee triangle game! So...can you do it? or are you "Just plain dumb!"? LOL
The article for this game can be found on page 52 of Compute's Gazette Issue 42 (December 1986)

Why TAS This Game?

The continuation of TASing games from my all-time favorite magazine, Compute's Gazette. This makes my 56th TAS from this series.
This was a very famous magazine issue for me. I clearly remember wanting to play Q-Bird only. After that...I played that game over and over and never got to this one. Well, now that I know what it is...I regret not giving this a try.

Game Difficulty and Ending

There is no selection of difficulty, as it is purely a strategy game. The ending is when you remove all but one peg.

Effort In TASing (BOTing)

This game may not be exciting to watch, as a TAS, but it certainly was one of the more exciting ones to produce. In fact, this is the first game that prompted a BOT that I didn't expect to write. Every input was BOTed and was done so in a most unique fashion. Here, I ran an solution file that was pre-generated by the web owner of https://www.joenord.com/triangle-peg-game-all-solutions/. The site owner found that 438,984 solutions were possible for winning the famous triangle game.
So why not just pick one solution and call it done? Well, this game was written in BASIC and along with such writing...you get inefficient responses. When I analyzed the code, I saw that it was enumerating a number of arrays, that pertained to the playing field, and the speed of each response was unpredictable. This was something that I didn't expect, after initially completing the game with a certain solution and noticing slowness between some moves and not others. So..I looked up some other solutions and found the aforementioned site and decided to create a unique BOT that I now call "Play BOT". So I used my Ryzen 3950X to run 16 instances to finish executing all solutions in roughly 15 hours, where it ran through all solutions and found the fastest ending at frame 1528, where the previous manual effort was 1817. So this was an exciting cut of 289 frames...where man vs machine!
One other reason why this new BOT is exciting to have...it can serve as a routing BOT for many other games that have systematic navigation, like Q-Bert or Pacman. So now I believe most of my future submission can be BOTed!
So far, I now have these BOTs that I use...depending on the situation:
  • F-BOT: for racing games
  • T2: A template driven BOT that I've only used on Tetris 2
  • Brutus: a strictly brute force algorithm for short segments
  • Splotch: a random input generator that I use only on Commodore 64 games
  • Play BOT: My new addition to the family, which is featured with this game.

Human Comparison

None to be found. :(

DrD2k9: Claiming for judging.
DrD2k9: Nice work. Accepting as Maximum Score branch/goal.
fsvgm777: Processing. Also replacing the movie file with one that sets the re-record count to 0, due to the submission having botted rerecords (per current policy).
Last Edited by fsvgm777 on 1/26/2024 9:04 PM
Page History Latest diff List referrers