First I want to make clear that this run and the \"best ending\" run weren\'t done all by myself. It\'s mainly a co-operation between me and Hotarubi, a japanese regular speedrunner who showed to have some excellent sense for TAS as well. During the creation of this and the other Gimmick! run we took some breaks in our work, the longest being from the beginning of December up until 3 weeks ago when I took it up again. I tried to contact Hotarubi but he apparently hasn\'t been available to respond, and because he\'s not a member of this site I can\'t add him as co-author - something which I\'d like to be done as soon, or if, he joins some time later.
  • Emulator used: FCEU 0.98.28 (submitted an fm2 that syncs with FCEUX 2.1.4a)
  • Aims for fastest time
  • Takes damage to save time
  • Uses death to save time (no life is lost however)
  • Heavy glitch abuse
This run beats my cancelled run by 2356 frames (~39 seconds) and the current published run by 3404 frames (~57 seconds). Never in my life I would\'ve thought such an improvement was possible when we started it since I only thought we would implement one new shortcut, instead it ended up in a cavalcade of new strategies. Depending of which new things you count as new strategies and which you just consider to be minor changes I\'d say there\'s about 25-30 new strategies (compared to the cancelled run). If you want to really see the changes themselves I suggest to first watch the cancelled run or current published run before watching this one.

Time savings: (CR = the cancelled run, CPR = the current published run)

LevelFrames saved over CRFrames ahead of CRFrames saved over CPRFrames ahead of CPR
Level 185858686
Level 210519010441130
Level 31805199518242954
Level 417821732673221
Level 5922265923313
Level 6912356913404
These improvement descriptions explain what\'s new since the cancelled run. It\'s explained only briefly but for surprise you\'ll might want to watch the run before reading it.
Level 1: Better precision and new strategies in more than half of the rooms, by more effective star toss and using an enemy as a platform.
Level 2: Better precision and a new glitch abuse by entering a pipe in a way the developers probably didn\'t consider you to:)
Level 3: Better precision, got one additional bomb in the first room only losing 1 frame, new strategy in the room before the underground, big timesaver by the waterfall and a huge timesaver at the boss.
Level 4: Better precision, lots of new strategies involving the star and better boss fight.
Level 5: Better precision, minor adjustment in the rooms before the first pipe, two new strategies in the room with lots of conveyor belts and better boss fight.
Level 6: Better precision, new strat for the warriors at the beginning of the castle, a new strategy in room 5 of the castle, a new interesting strategy for the first mini boss ghost and a new strategy for the second mini boss ghost. The big ghost couldn\'t be as fast in the previous run since its movement is controlled by a timer that starts at the beginning of the level, but not too much time was lost.
I start by giving you the any% run but the \"best ending\" run will show up very soon! Actually, it\'s here now!
Enjoy!

cpadolf: A very impressive improvement and a great movie. Accepting for publication.
Aktan: Processing... should be quick!

Post subject: Movie published
TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14904
Location: 127.0.0.1
This movie has been published. The posts before this message apply to the submission, and posts after this message apply to the published movie. ---- [1543] NES Gimmick! by Aglar & Hotarubi in 05:11.49
Joined: 7/2/2007
Posts: 3960
While a savings of 57 seconds might have provoked the authors to whoop a bit in excitement, I suspect the desired word in the movie description was "whopping". :)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Publisher
Joined: 4/23/2009
Posts: 1283
Derakon wrote:
While a savings of 57 seconds might have provoked the authors to whoop a bit in excitement, I suspect the desired word in the movie description was "whopping". :)
Oops, moozooh changed it for me. =)
Dwedit
He/Him
Joined: 3/24/2006
Posts: 692
Location: Chicago
I was just watching the very first TAS of Gimmick, which requires Famtasia to play, which was 8:37.50 in length. You can clearly see the difference in play style. For instance, it doesn't even bounce stars off walls to move faster, or use the death trick to skip collecting points at the end of a level. The current "Best Ending" run even beats it.
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
I've noticed the SMB-like wall-jump right after the first boss fight. Was it described anywhere?
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Expert player (3584)
Joined: 11/9/2007
Posts: 375
Location: Varberg, Sweden
feos wrote:
I've noticed the SMB-like wall-jump right after the first boss fight. Was it described anywhere?
I skipped decribing it since it didn't serve any speed purpose. If I remember right, the only thing you have to do is being as close as possible to landing on the platform, then you'll instead land on the block below - kind of like in Super Mario Bros 2, but it's harder to perform here since as when you enter the wall you start to get zipped out of it.
feos wrote:
Only Aglar can improve this now.
Site Admin, Skilled player (1236)
Joined: 4/17/2010
Posts: 11274
Location: RU
Aglar wrote:
when you enter the wall you start to get zipped out of it.
Irrelatively of direction you look to?
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Expert player (3584)
Joined: 11/9/2007
Posts: 375
Location: Varberg, Sweden
feos wrote:
Aglar wrote:
when you enter the wall you start to get zipped out of it.
Irrelatively of direction you look to?
(After some testings) Yes. The zipping speed is 1 pixel/frame.
feos wrote:
Only Aglar can improve this now.