User Files from Cephla

interesting game I thought I'd have a look at. ended up finding a pretty sure-fire way to repeatedly launch the ball into the goal even with the previous iterations working away doing similar movements behind you. it didn't become particularly difficult until the last iteration and I had to experiment a lot more to get any balls in. I decided I'd go up to 100 goals because it's a round number and also because I was hoping it would cap out around there (also managed to accidentally get a bonus 101st goal in there after ending input). unfortunately there doesn't seem to be a score cap and once it gets to 32769 it just wraps around and the game never ends which makes working on a maximum score tas a lot less feasible. I suppose I could experiment with botting a run but I wouldn't really know how to start on that since there's so much variation in how the ball might be sent flying that I can't see how I could script something to cope with that. there may also be alternative solutions where I can position the robots in such a way they automatically launch the ball into the goal without needing to move much but considering the goal moves every iteration I can't see how a formation could be set up where this works every time and it would take until all the robots are spawned in to fully set this up. had fun doing this though and quite pleased with how this tas in its rough state looks

tried starting on the first possible board and got a run 26 frames slower than starting on frame 196. it is probably possible to clean this up by a few frames but it cannot beat a 196 start

saved a massive amount of time by holding A

I only found out that this game is a demo with a full GBC game planned for release later this year before submitting but I don't want to forget or lose this run for when it comes out. For such a simplistic and short game, I did have a lot of fun with it and surprised myself by actually finding glitches which I'm usually terrible at.

level 2 improvement thanks to lgr

It's been a while but I've managed to get the script working with the new custom board generator functions. This one is for the first level in the C64 flooder and I've already got a solution down to 7 moves which I think is the lowest possible but I'm testing currently to see if 6 is possible, though I'm doubtful

My attempt at making a DS TAS fails once again, though this is definitely a TAS I want to finish even if it takes 10+ years.

The requirements I've drawn up since there's no SRC leaderboard for 100%:

All emblems

There's more but pretty much anything that would contribute to 100% is covered by an emblem except crowns, I think but if they're not covered by an emblem, that's added to the 100% requirement too. The requirements for all emblems is:

All character emblems (completing all character's missions) Olympic Record Emblem World Record Emblem Track Emblem (Gold medals in all track events) Field Emblem (^ field events) Aquatic Emblem (^ aquatic events) Gymnastics Emblem (^ gymnastic events) Dream Emblem (^ dream events) Full Play Emblem (Play all events) Fencing Emblem (10-0 score in fencing) Archery Emblem (120 points in archery) Shooting Emblem (40 points in shooting) Bouncing Emblem (10 points in trampolining) Somersault Emblem (10 points in vault) Table Tennis Emblem (5-0 score in table tennis) Full Game Emblem (Unlock all events) Thanks for playing Emblem (Load your save 50 times) Collaboration Emblem (Play as each character once) Mario Family Emblem (Get a gold medal with every member of the Mario family) Sonic Family Emblem (Get a gold medal with every member of the Sonic family) Gallery Owner Emblem (Complete the gallery) Friendly Emblem (play versus mode except this one has to be skipped due to network play being required) Rocket Emblem (perform a start dash) Penalty Emblem (Commit a foul)

So far, I've unlocked the entire gallery, got the Thanks for Playing Emblem and have started on the first circuit but have stopped due to being unable to get a perfect 100m. I've reached out to MKDasher who has uploaded a perfect 100m for every character to YT but am yet to get a response (it's been a couple weeks already so doubt I'll get anything back).

The reason I'm doing circuits now is to unlock all events first. This allows me to do all missions next and I can also snag some perfect score emblems along the way. Getting all gold medals for every character will be the last objective.

Now, if you'll excuse me, I'll be over here sitting on WIP all words nes wordle tas, debating on doing a 100 flooder rounds in a row and perhaps starting on postman pat and peppa pig ds tases

huge

missed it

ok, this one's going to need a bit of an explanation. I found another flooder game https://josipretrobits.itch.io/flooder this time for the C64. The gameplay is identical to the GBC game which this script is written for except it's a level based game with all patterns being preset and the solutions gradually getting harder with the board getting bigger with each level. There's also no limit on the amount of moves that can be made, just as long as you can beat the level, you win. But that's not good enough for a TAS and I want to find the optimal solution for each level. My first thought was to adapt the brute force script for the GBC version for this version, but unfortunately the colour changes are not instantaneous and it takes a million years for the board to change colour one pixel at a time. It also doesn't help that you can only have 10 colours loaded at a time and you have to wait for the game to change colour ten times before you can load the next ten so this isn't a viable solution. Or is it? I've already got a script for the GBC game so why not recreate the C64 game in that game? What an absolutely fool proof solution. This file recreates the first level of the easy mode in the C64 mode by using colours 1-4 and a 6th grey colour to make the empty space.

I'm yet to modify the brute force script to make all pixels turn grey when the board is solved but I've got ideas for that. The C64 game isn't particularly fun to TAS and it takes ages so I haven't made much progress and I haven't actually used this custom pattern script properly yet. The best method I've tried so far to find the fastest solutions is good ol' google sheets (at one point I was using google slides lmao) but I haven't worked on this for weeks and probably won't again but thought it'd be worth uploading anywa

ror

actually wrote this one myself (which explains why it's so bad) and it works exactly as intended. to run, you need to start the game and do the first move yourself, then modify the section inside the first while loop before the second so it can only select neighbouring colours. After that, it should run smoothly and may give you a solution in 10 or so minutes depending on which pattern you choose.

I'm currently working on the second possible pattern and after slimming it down, it's 348 frames long, only 16 longer than my none botted tas I did ages ago. Will see what happens when I optimise this solution further

uses latest version

I've lost the little motivation I had for this game but I may come back to it in the future

improvement found in ruby 4

the logical step to take now is working out how to bypass wall collisions

I also found out that if his y level is at a sepcific value, he can walk in the wall

Fun puzzle game with an interesting central mechanic. Towards the end of where I'm up to, I realised you could just walk through the guy who talks to you and blocks your path which will save a bunch of time when I implement that. There are also 2 versions of the game, english and russian. Once I'm done with the TAS, I'll see if the russian version saves any time in cutscenes.

quite a bit of timesave I can see in lvl 1, level 2 maybe. Currently at a dead end with The Getaway due to dodgy collisions or something. This movie shows that the Jammy Dodger can't hit the boat for some reason (emulator bug?)

so far, I believe everything before mountains 2 is faster than the current publication even though it doesn't look like it because of framerate differences but I'm really stuck on this level. if I put in a lot more work, I might be able to but I'll probably shelve it for a bit unless someone else decides to improve upon this

TNtdiTT is a homebrew Gameboy game developed during the first 2020 lockdown and is heavily inspired by it.

The player must go around town completing 3 tasks to complete the game but there's not much to it; just some frame perfect A presses here and everywhere.

By RTA timing, this comes out (roughly) as 5.967 seconds faster than wr (a 3:32.2)

Objectives:

YT encode

This is one of three Peppa Pig games released for the NDS, with Peppa Pig: The Game being released in 2009.

The game consists of 11 poorly built minigames:

You can earn up to 3 stars for each minigame depending on either how many times you did something, how quickly you did something or how many times your kite got damaged.

This TAS aims to collect all 33 stars as quickly as possible.

I don't have much info, the game doesn't even have a Wikipedia page. It's basically just a bunch of minigames you can't speed up. Just imagine a very low budget Mario Party but without the multiplayer, and in 2D, and with Postman Pat's face slapped on everything instead of Mario. That's this game.

Here's a youtube encode anyway.