Super Mario World 96 exit TAS - by Fabian

  • Aims for fastest time
  • Takes damage to save time
  • Abuses programming errors
  • Manipulates luck
  • Plays all levels
  • Emulator Used: SNES9x 1.43 +v10
    • Use WIP1 Timing - on
    • Allow Left+Right / Up+Down - off
    • Volume Envelope Height Reading - on
    • Fake Mute desync Workaround - off
    • Sync samples with sound CPU - off

Introduction

This is an improvement of VIPer7's 96 exit Super Mario World run. It's 7724 frames, or 2 minutes, 8 seconds and 44 frames, faster. However, VIPer7 did not end his movie at the final hit on Bowser, the time difference at that point is 4340 frames (1 minute, 12 seconds and 20 frames). Techniques and strategies leading to this improvement are detailed below.
My biggest challenge when making this run was to come up with something I felt was worth watching. With all the recent SMW and SDW runs, ending up with a fresh looking and original final product has been very difficult. Trying to make each level as entertaining and fun as possible, to prevent the viewer from getting bored, has been my number one priority at all times. I do hope you will enjoy watching this movie, even if you've seen several Super Mario World or Super Demo World movies in the past.

Techniques used

Nearly every technique described on the Super Mario World Tricks Page has been used. The most important difference from VIPer7's run is the abuse of Mario's oscillating speed. When holding "forward", Mario's speed oscillates between several different values. If "forward" is released on the right frame while in mid-air, the oscillation will stop at Mario's fastest speed. This speed can then be kept up by continuously jumping on the first frame Mario touches the ground, which is commonly known as the "hopping glitch". This works both with flying and running, as well as Mario's medium running speed. This "hopping glitch" can be used to your advantage in many situations and you'll see me using this technique in one way or another in many levels.
There's one other thing I'll describe, since it's not on the tricks page yet; Yoshi flight. Flying with Yoshi is usually the exact same thing as spin flying, which is the same speed as regular running. With the Yoshi flight glitch, however, Mario and Yoshi move as fast as regular flying. This can be initiated in several different ways, two of which are used in this movie. The first is to dive onto Yoshi while Yoshi is standing on a shell (see Vanilla Dome 3 and Vanilla Dome 4), the second is to dismount Yoshi when flying pressed to a ceiling (see Valley of Bowser 3 and Valley of Bowser 4b). The second method is often considerably faster, and it's worth noting this technique was not known at the start of this run. Had it been, it could have been used to save time in Cheese Bridge Area, and possibly have been faster overall in Vanilla Dome 4, although this has not been tested.
If you're unsure about anything, I recommend checking the tricks page, and if that doesn't answer your question, ask me.

Other things

I want to thank JXQ for his constant support and feedback, always listening to me whine and helping me with hexediting, among many other things. Thanks man.
I also want to thank jimsfriend for help with level strategies and lots of ideas.
Lastly I want to thank all of you who have offered encouragement, helped me test ideas and shown interest in my run. I really appreciate it everybody.
If you have any questions at all, feel free to ask me, I'll be happy to answer them. Thanks for watching.
Suggested screenshot: Frame 129112


Post subject: 20/20 (10+10)
Banned User
Joined: 12/23/2004
Posts: 1850
I can't finish the movie. For the second time in a row I was unfortunate enough to hit '6' (Joypad Switch -- why is something that can desync a movie so easily put right next to the useful layer buttons? arhghghgh)... Oh well, everything up to the end (Bowser Fight) was nice. :) I'd have to say Chocolate Island 2 (or whatever the goofy one that uses arbitrary exits) would be my favorite, since it's the only level where we can take a break from the rush and just enjoy some superhuman playing around. It's definitely a nice break from the stale "fly through levels as fast as possible" (though it wasn't nearly as bad here)... I'm actually suprised this hasn't gotten a star yet, seeing as it's ranked #1 overall (0.2 points above Excitebike). (also, we should probalby consider having some other little graphic to denote the top x movies; mabye a little trophy for the top 3 and a blue ribbon for 4-5 or something, just so that thre's an easy way to tell what the top one is without appending ?ratesrt=Y or whatever it is to the end of the URL, but that's a matter for another thread somewhere else)
Perma-banned
Senior Moderator
Joined: 8/4/2005
Posts: 5777
Location: Away
JXQ wrote:
Under a minute for sure. I'd be surprised if you could get more than six seconds.
I don't want to spoil the fun, but this is the moment where it becomes funny: • as some of you may know, Megaman TAS can be improved by no more 3 frames (contrary to the popular belief about its perfection) in any of the known ways, its technical quality rating is 9.3; • even 3 seconds of possible improvements in this run are 60 times more than 3 frames, and the technical rating is 9.4. Erm… wait, I'm a bit confused here. Giving so many tens to a run that can be improved in a known way… isn't it that scary cross-contamination the rating guidelines talk about (applies to the Megaman run as well, but AFAIR, the improvement was found by Vatchern and mentioned on IRC a few times, so most people don't even know about it)? :D Don't get me wrong, not that I don't want it to have high rating, but it still is a bit… strange. For the record, I gave nine (and not ten) to both of the aforementioned runs for the sole fact that they can at least be improved, even by one frame, blah blah blah, disregard that.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Banned User
Joined: 12/23/2004
Posts: 1850
moozooh wrote:
Erm… wait, I'm a bit confused here. Giving so many tens to a run that can be improved in a known way…
6 seconds in a 1:21:30 tends to mean considerably less than, say, a few frames in a much shorter movie. Besides, has it even been confirmed that any of this time is actually savable?
Perma-banned
Former player
Joined: 6/4/2006
Posts: 267
Location: CO
moozooh, for OoT, people are talking about a sub-3 hour run. This means the current version could be improved by about 35 minutes. That's 700 times longer than the 3 frames for Megaman, and the techical rating for his movie was a 6.4, only 0.1 less than the entertainment value. That's a much bigger injustice I think, even if you consider a 6.4 "low." Plus there's the 89% approval rating for this movie. Just pointing it out. Xkeeper, the time that is known to be saveable is attributed to the following: First, a few frames were lost here and there from missed corner boosts in underwater levels, which was discovered midway through the run. Second, a few frames here and there from imperfectly using the hopping glitch, was was not fully understood until midway though the run. Third, a few frames sacrificed for entertainment, which I think was well worth it. Finally, about 2 seconds could be gained using a glitch on Tubular that was discovered well after Tubular was done. Any proposed new routes to save time are not for sure known to be possible. I would say this would be about 4 seconds, I only said "under a minute" earlier to leave open lots of room in case new routes were discovered. EDIT: An easier way to compare this run to OoT (only because I mentioned it earlier) would be by percent. This run could be about 0.08% faster, while OoT could be about 20% faster. So this run is about 250 times less improvable than OoT.
Former player
Joined: 6/15/2005
Posts: 1711
Xkeeper wrote:
moozooh wrote:
Erm… wait, I'm a bit confused here. Giving so many tens to a run that can be improved in a known way…
6 seconds in a 1:21:30 tends to mean considerably less than, say, a few frames in a much shorter movie. Besides, has it even been confirmed that any of this time is actually savable?
I'd say 6 seconds is a pretty good estimate on time I know can be saved. JXQ was nice enough to help with hexediting for this run, but unfortunately it didn't work very well in most places. If it had, this run would have been faster. If you read the SMW thread, you'll see thegreginator has lots of ideas for strategy changes in levels (like more flying under ground and lava and stuff), but unfortunately I think none of them would work.
Zoey Ridin' High <Fabian_> I prett much never drunk
Senior Moderator
Joined: 8/4/2005
Posts: 5777
Location: Away
Xkeeper wrote:
6 seconds in a 1:21:30 tends to mean considerably less than, say, a few frames in a much shorter movie.
If we're talking about proportions, 96-exits run should have been several hours long to justify the difference. Anyways, "10" equals to "perfect", which means "can't be improved in a known way", and both mentioned movies are in no way perfect. Excellent, yes, but for the aforementioned reason — not perfect.
thegreginator wrote:
moozooh, for OoT, people are talking about a sub-3 hour run. This means the current version could be improved by about 35 minutes. That's 700 times longer than the 3 frames for Megaman, and the techical rating for his movie was a 6.4, only 0.1 less than the entertainment value. That's a much bigger injustice I think, even if you consider a 6.4 "low." Plus there's the 89% approval rating for this movie. Just pointing it out.
I have nothing to object here, quite possibly that you're right. Too bad I'm not familiar with OoT in the slightest (I've never had N64, and playing its games on an emulator — more precisely, using a keyboard — is too much of a pain to bother), and I haven't even watch the movie yet, though I certainly will (maybe even tonight). As you may see, I'm not a Zelda fan, too. :) Some other runs get [seemingly] underrated (or overrated) for no particular reasons, let alone forcible arguments, too, and most of the time ratings can get messed up only because of the personal disposition towards the game itself, the circumstances of the run's creation, or the author's personality. Probably, can't be helped. The idea of rating objectiveness is far too utopian to work as intended, anyway. ;)
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
JXQ
Experienced player (761)
Joined: 5/6/2005
Posts: 3132
Technical perfection is tough to judge if one doesn't know the game quite well. And there could always be techniques that are not yet discovered, which would mean the values are based on incomplete knowledge (although that's a bit of a stretch). To circumvent this, I have always judged that rating on how technically perfect the run looks. There is only one improvement I know of for this run that would change how things look. The rest are either visually minor, or completely undetectable without frame counters and/or memory watching.
<Swordless> Go hug a tree, you vegetarian (I bet you really are one)
Former player
Joined: 8/1/2004
Posts: 2687
Location: Seattle, WA
moozooh wrote:
Anyways, "10" equals to "perfect", which means "can't be improved in a known way", and both mentioned movies are in no way perfect. Excellent, yes, but for the aforementioned reason — not perfect.
Have you given out any 10s? 9s? The rating system is relative: what is perfect to you is imperfect to others, and vice-versa. However, if you're holding out for your own perfection, you might as well make the ratings scale 1-9, since you're never going to see a perfect submission.
hi nitrodon streamline: cyn-chine
Former player
Joined: 6/15/2005
Posts: 1711
Zurreco wrote:
moozooh wrote:
Anyways, "10" equals to "perfect", which means "can't be improved in a known way", and both mentioned movies are in no way perfect. Excellent, yes, but for the aforementioned reason — not perfect.
Have you given out any 10s? 9s? The rating system is relative: what is perfect to you is imperfect to others, and vice-versa. However, if you're holding out for your own perfection, you might as well make the ratings scale 1-9, since you're never going to see a perfect submission.
I agree with Zurreco in that I don't think the rating system was designed to never giving out 10s. I imagine it's more of a bell curve kind of thing, with the very best at 10, and the very worst at 0. Since no run is perfect, that would mean the 10s aren't either, of course. I didn't want this to turn into some rating debate though. I haven't said to anyone "hey guys my run rocks, make sure to give it a 10 on technical perfection!", and you won't here me saying it to anyone either. I only hope people rate it in a way they think is fair.
Zoey Ridin' High <Fabian_> I prett much never drunk
Former player
Joined: 8/1/2004
Posts: 2687
Location: Seattle, WA
Fabian wrote:
I haven't said to anyone "hey guys my run rocks, make sure to give it a 10 on technical perfection!", and you won't here me saying it to anyone either.
By referring to this as a Fabian, you have implied that this submission would be the best thing since pre-sliced bread.
hi nitrodon streamline: cyn-chine
nesrocks
He/Him
Player (247)
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
Fabian wrote:
I didn't want this to turn into some rating debate though. I haven't said to anyone "hey guys my run rocks, make sure to give it a 10 on technical perfection!", and you won't here me saying it to anyone either. I only hope people rate it in a way they think is fair.
You say that because you know your run deserves 10 out of 10 so you're safe!!! :) Well, i've already said it on IRC, but, great run, thanks for making it, enjoyed every bit of it.
Former player
Joined: 6/15/2005
Posts: 1711
FODA wrote:
Fabian wrote:
I didn't want this to turn into some rating debate though. I haven't said to anyone "hey guys my run rocks, make sure to give it a 10 on technical perfection!", and you won't here me saying it to anyone either. I only hope people rate it in a way they think is fair.
You say that because you know your run deserves 10 out of 10 so you're safe!!! :) Well, i've already said it on IRC, but, great run, thanks for making it, enjoyed every bit of it.
It's up to each person to decide if it deserves a 10 or not :) And thanks again FODA, I'm glad you enjoyed it.
Zoey Ridin' High <Fabian_> I prett much never drunk
Former player
Joined: 6/15/2005
Posts: 1711
I will say this though, regarding the ratings: Right now after 24 votes my run has an average of 9.5 (second highest rated run has an average of 9.2). I really can't tell you guys how happy I am that you all like it that much, I mean I know I sound corny but I truly think it's awesome and that it makes all the hours I spent on this totally worth it. Thank you very much everyone.
Zoey Ridin' High <Fabian_> I prett much never drunk
Former player
Joined: 6/4/2006
Posts: 267
Location: CO
JXQ wrote:
Technical perfection is tough to judge if one doesn't know the game quite well. And there could always be techniques that are not yet discovered, which would mean the values are based on incomplete knowledge (although that's a bit of a stretch). To circumvent this, I have always judged that rating on how technically perfect the run looks.
Exactly.
Zurreco wrote:
The rating system is relative: what is perfect to you is imperfect to others, and vice-versa. However, if you're holding out for your own perfection, you might as well make the ratings scale 1-9, since you're never going to see a perfect submission.
Exactly. I gave this game 10's in both categories.
Joined: 8/10/2006
Posts: 25
Sorry if this is a little late and I know that this is already published, but I want to congratulate you on going through this run. Even though I couldn't have voted, I would've given this an obvious Yes vote. I know you had some trouble getting it done, but now that it is done, it looks freakin great. It would be hard for me to find a part of the run that I liked the best, it was all awesome. I hope to watch any other game you run in the future and I hope you make look as kickass as this one. Good job man!
Active player (407)
Joined: 3/22/2006
Posts: 708
Yes! I don't care if someone eventually beats this time or how badly they do so; this movie should never be obsoleted. Anyone else invariably would not do it with the same level of style. This is easily one of my favourites. Awesome job.
P.JBoy
Any
Editor
Joined: 3/25/2006
Posts: 850
Location: stuck in Pandora's box HELLPP!!!
  • Takes damage to save time
huh also, voted 10 and 9, i don't think i've ever voted 2 tens ^_^ i cant beleive that DPGH, how low can ya fly. maybe it's just me, but i'd be more entertained if you used fire mario more, like in the nes mario's keep up the great work
Former player
Joined: 1/17/2006
Posts: 775
Location: Deign
Fire Mario moves slower than Cape Mario.
Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign aqfaq Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign
Former player
Joined: 6/15/2005
Posts: 1711
P.JBoy wrote:
  • Takes damage to save time
huh also, voted 10 and 9, i don't think i've ever voted 2 tens ^_^ i cant beleive that DPGH, how low can ya fly. maybe it's just me, but i'd be more entertained if you used fire mario more, like in the nes mario's keep up the great work
You're right, that's badly worded. Gets hit to save time would be more accurate. And like jimsfriend says, the reason fire mario isn't used is because flying with the cape is faster than running.
Zoey Ridin' High <Fabian_> I prett much never drunk
Joined: 10/24/2005
Posts: 1080
Location: San Jose
Great run Fabian! I really enjoyed it. I don't have much to comment on, as most of the praise has already been said, and after viewing this run... I feel kind of humble. :)
<agill> banana banana banana terracotta pie! <Shinryuu> ho-la terracotta barba-ra anal-o~
KennyMan666
He/Him
Joined: 8/24/2005
Posts: 375
Location: Göteboj
Awesome TAS of decent game.
Det man inte har i begåvning får man ta ut i energi. "I think I need to get to Snoop Dogg's level of high to be able to research this post." -Samsara Read my fanfic, One Piece: Pure Corruption
Joined: 11/15/2004
Posts: 804
Location: Canada
I'm glad that screenshot was chosen for the publication page. Mario juggling a bunch of mushrooms and feathers at the same time was awesome. :)
TASing or playing back a DOS game? Make sure your files match the archive at RGB Classic Games.
Mitjitsu
He/Him
Banned User
Joined: 4/24/2006
Posts: 2997
Finally managed to watch it, the reason I couldn't watch it before was because I didn't have the most upto date emulator. Generally it was very good and I espeacially liked all the mushroom juggling around about the vanilla dome. However, I'm of the opinion that this movie isn't worthy of star status mainly because its so long and does get very repetive throughout and I could also see potential improvements and shortcuts espeacially ones that might avoid the timed sequeces i.e. moving yellow walls. I think sub 1:20 might be possible although that might be pushing the boat out too far. The bowser battle is definitley the best I 've seen out of all the movies. So to sum up "Did you enjoy this movie?" In terms of speed? Yes it was very quick and would certainly catch the eye of the expertly skilled gamer In terms of entertainment? Generally yes although some of the stuff got too repetive a bit more varation was needed. The vote? Well, there is no reason whatsoever to vote no, nor is meh a good enough option. I really do think a best out of ten system would be better, as I would of given it an 8 and in terms of percent I would have given it 92% (yes, confusing I know I'll explain why shortly). That might sound confusing but an eight is to signify that its not a high 9. Since I can't rate I'll simply have to vote yes.
Senior Moderator
Joined: 8/4/2005
Posts: 5777
Location: Away
Uh, actually you can rate.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Former player
Joined: 1/17/2006
Posts: 775
Location: Deign
Can you give specific examples where you see things that could be improved?
Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign aqfaq Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign