Submission Text Full Submission Page
This may be the fastest that one can possibly complete this game (or any SNES game for that matter). It is kind of an odd choice for a speed run, and rather confusing to watch because it goes by so fast, but conceptually it is pretty interesting (and quite amusing to see the game engine get totally trampled by TAS). Enjoy!

Game objectives

  • Emulator used: Snes9x 1.43+ v15.3 rerecording
  • Aims for fastest time.

Comments

In this case, Clue has little to do with detective work and much to do with menu navigation. There is literally nothing to this game except navigating menus and stepping through cut-scenes. This might possibly be the shortest completion time of any SNES game. 1624 frames (roughly 27 seconds).
There is no luck manipulation involved in this video. All button presses happen at their earliest possible moment, and there are no extra frames inserted (intentionally or otherwise).
We discovered a crucial fact about the menu system of this game. The game will accept cursor-motion input one frame earlier than it will accept input from the action button. So when navigating through a list of menu items or a palette of person/weapon/room choices, you get one free move in any direction. This allows us to do the following without spending any extra frames:
  • Choose "Detective" difficulty level.
  • Make players 2 and 3 into computer characters.
  • Choose characters other than Mustard and White as the computer characters.
  • Accuse a person/weapon/room other than Mustard/Knife/Hall. This is crucial. A previous attempt used luck manipulation to force a scenario wherein the answer is Mustard/Knife/Hall, because these are the items at the cursor's default position (thus saving the extra frames needed to move the cursor). However with the ability to make 1 free cursor move, there is no additional cost to choosing any of the items adjacent to Mustard/Knife/Hall.
The scenario (person/weapon/room) is chosen when you select "Start Game" after choosing characters. The outcome depends on the exact frame at which you press the action button at the "Start Game" screen. If you provide input at the earliest possible moment in all menus (up to and including this one), the outcome scenario is "Plum/Knife/Conservatory". Knife is the default selection (no time cost). Plum is within 1 step of the default selection (no time cost -- the first step is free). But Conservatory is 2 steps away from the default selection, so it costs 1 extra frame to reach it. Any attempt to use luck manipulation to get a better scenario would require the insertion of 1 or more frames. Therefore a faster solution can not be found using luck manipulation.
Unless someone discovers something about the game mechanics that we've missed, this is the absolute fastest time in which the game can be completed. Even in TAS, how often can you say that?

mmbossman: Seems you've been bested

TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14873
Location: 127.0.0.1
This topic is for the purpose of discussing #2234: feasel, Oguz's SNES Clue in 00:27.07
Joined: 11/4/2007
Posts: 1772
Location: Australia, Victoria
If this is REALLY the fastest possible, would you mind me youtubing this?
Banned User
Joined: 8/2/2008
Posts: 420
Location: italy
I liked this. Short and apparently optimised to the max. Yes vote, and have a free encode, flygon. MKV'd for a change :P
Gone.
Joined: 5/2/2006
Posts: 1020
Location: Boulder, CO
Is the randomness the same every reset? The implication is that this is optimal because it would be impossible to get an ideal set of room/person/weapon without loosing frames, but would this be the case regardless of the seed? Or is there no seed for this game?
Has never colored a dinosaur.
Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
As speculation: Does pressing extra buttons at any point alter the randomness, or does where you move to choose the difficulty?
Joined: 4/18/2009
Posts: 3
I've experimented a little bit to see how the pseudo-randomness works. The guilty scenario appears to depend only on frame number. Choosing other characters or giving extraneous input does not seem to alter the outcome. And adding extra frames anywhere in the opening part of the game would produce a slower (or at best equivalent) completion time, even if it managed to get a more fortunate scenario (e.g. Mustard/Knife/Hall, or one of the adjacent possibilities). I don't know how the randomness would be affected by a soft reset, but this video is of course recorded from a hard reset, and the distribution of scenarios is the same every time the system is booted (else it wouldn't work if you tried to replay the video, right?)
Active player (441)
Joined: 3/21/2006
Posts: 940
Location: Toronto, Canada
I'll throw in a yes vote. I was entertained by it and it looked perfectly optimized, so...yeah :D
My current project: Something mysterious (oooooh!) My username is all lower-case letters. Please get it right :(
Joined: 11/4/2007
Posts: 1772
Location: Australia, Victoria
It took longer to log into youtube than to actually download the encode, ignoring Rapidshares wait time of course. ^.^ Fastest game of Cluedo ever played. Yes, I said Cluedo. Thanks nineko for the encode.
Joined: 4/19/2009
Posts: 2
Ha! That's hilarious. Clue in 27 seconds...
Joined: 6/26/2007
Posts: 147
There's one thing I wonder, would a different scenario result in shorter ending animations?
TASVideosGrue
They/Them
Joined: 10/1/2008
Posts: 2738
Location: The dark corners of the TASVideos server
om, nom, nom... juicy!