Post subject: Mario party
Editor
Joined: 3/10/2010
Posts: 899
Location: Sweden
I know what you all think: What? Mario Party? Extremely bad game choose! It has too little action! Too much text! Too little actual gameplay! But these are based on the assumption that you would do a fastest completion run of the game. Instead I propose a run where no minigames are repeated and the computer is greatly outclassed to such an extent that it's fun again. With luck manipulation and super reflexes, this is easy to pull off. They are not just beaten in the minigames, but keep donating things to the player character. It would be easy to set up the most likely case only for it to suddenly end up with a major twist, making the cpu characters the best but monkeys ever. Oh, and there is developer times to beat too.
Editor, Expert player (2313)
Joined: 5/15/2007
Posts: 3855
Location: Germany
Mario Party 5 Minigames Will Flower (5''41) (mugg) Leaf Leap (180 Yards) (mugg) I thought about making a story mode TAS. However, it requires a lot of luck manipulation and with the few current savestate crashes I'm getting it's not really feasible. Also Dolphin crashes after beating Bowser...
Player (142)
Joined: 7/16/2009
Posts: 686
Isn't there a mini-game only mode in one of the games? That could make for a mildly entertaining TAS.
Editor, Experienced player (608)
Joined: 11/8/2010
Posts: 4012
Editor, Expert player (2313)
Joined: 5/15/2007
Posts: 3855
Location: Germany
In Mario Party 5, I'm trying to improve Will Flower and got 5'33 (old = 5'41). But I don't understand how I saved the frames... When looking at the input sequence, some input frames are duplicate because the emulator does that. What mystery is that? EDIT: Read the description in this video for more information: http://youtube.com/watch?v=BA7VV2qIAuw
RachelB
She/Her
Player (127)
Joined: 12/3/2011
Posts: 1579
The controller is polled at 60 fps, even when the game doesn't run at 60 fps. My guess would be lag. Even if nothing happens for a frame, the controller is still polled (and dolphin will skip the frame when you frame advance). If you look at a .dtm of a game that runs at 30 fps, you'll see that every single input is doubled. or occasionally tripled.
Skilled player (1637)
Joined: 11/15/2004
Posts: 2202
Location: Killjoy
MUGG wrote:
Leaf Leap (180 Yards)
Uh, wow, is this really that good? The reason I ask is they certainly animated interesting things that high - things the player would be unlikely to see if it required TAS tools. I could be wrong.
Sage advice from a friend of Jim: So put your tinfoil hat back in the closet, open your eyes to the truth, and realize that the government is in fact causing austismal cancer with it's 9/11 fluoride vaccinations of your water supply.
Joined: 7/2/2007
Posts: 3960
It looks like the game is simply iterating "press a direction and jump after a set interval" -- the interval is known, as is the time limit on the game, so you can easily calculate exactly how many times the loop can be processed in a perfect game. Then you animate enough to make it so even a perfect game doesn't run out of content, and you're done.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Editor, Experienced player (608)
Joined: 11/8/2010
Posts: 4012
MUGG wrote:
In Mario Party 5, I'm trying to improve Will Flower and got 5'33 (old = 5'41). But I don't understand how I saved the frames... When looking at the input sequence, some input frames are duplicate because the emulator does that. What mystery is that?
When there is a lag frame, Dolphin duplicates the input from the last frame to the lag frame, like rog said. You probably saved those frames from starting the minigame on a different frame, which affects lag throughout the minigame. I experienced this with Mario Party 7. If you find that specific frame that manipulates away all of the lag, your minigame TAS can go from this (0.66 second of lag) to this (no lag, just autofiring A).
Editor, Expert player (2313)
Joined: 5/15/2007
Posts: 3855
Location: Germany
I made a TAS that removes those lag frames by pausing. http://www.youtube.com/watch?v=RkWSpbZ4J_M I think 5"16 is the best time you can get.
Joined: 8/27/2009
Posts: 159
Location: California
Are the ones listed in this thread all the ones that have been done for GCN/Wii? If not, can I find a more complete list somewhere? Additionally, if I were to work on the minigames, is there anything I need to take into account when setting them up? Does the RNG seed of when the game starts play enough of a role that we need to do anything specific for it?
Editor, Expert player (2313)
Joined: 5/15/2007
Posts: 3855
Location: Germany
Derakon wrote:
It looks like the game is simply iterating "press a direction and jump after a set interval" -- the interval is known, as is the time limit on the game, so you can easily calculate exactly how many times the loop can be processed in a perfect game. Then you animate enough to make it so even a perfect game doesn't run out of content, and you're done.
I TASed it again, this time removing all the odd lag frames that I encountered the first time around. 180 yards really is the best score you can get; the beanstalk stops growing after that point and I'm ~3 frames or so short of doing another jump after reaching those 180 yards.
Joined: 4/18/2010
Posts: 103
Mario Party 4 - Slime time in 2-3 seconds http://www.youtube.com/watch?v=1ZZgDOq0kv8
Editor, Experienced player (608)
Joined: 11/8/2010
Posts: 4012