Submission Text Full Submission Page
I'm sure you've never heard of Super Mario Bros., but it's pretty fun. I'd check it out if I were you. In any case, for this April Fools' Day, because my Cat Planet attempt failed (Good thing too, as FuzZerd beat me to it), I decided to submit my Minimum Presses TAS of SMB that I completed on 01/19/12. This is the 7th and best attempt at this feat, this time completing the game in 05:46.33 and 145 presses. Although I've been told that this kind of run will likely never see publication, I figured this was a worthwhile attempt for today, and with FractalFusion's help, the video might be entertaining enough after all.
The encodes all say that the length is 05:46.33, but that's the number that FCEUX gave me, so it's seriously not my fault gosh.

Game Objectives

  • Emulator used: FCEUX 2.1.5
  • Attempts to complete the game with the least amount of button presses possible
  • Avoids wasting time
  • Aims to maximize score / kills without jeopardizing the above goals
  • Takes damage to save presses
  • Moderate luck manipulation
  • Becomes a dancing queen

The Run

Title: 1 press

1 press required to get past the title screen. Duh.

1-1: 14 presses

I grab a mushroom here without losing any presses. This will be helpful in future levels. In the underground section, I jump so I don't have to slow down to exit through the pipe. At the end, instead of jumping right to the flagpole, I fall down and jump later to avoid getting fireworks.

1-2: 13 presses

I wallkick to get through the ceiling in the fastest time possible. As I need to move left to do so, I continue holding left while holding right to moonwalk. Doing so allows me to simply release right at the end to walk to the warp pipe instead of having to press left again. At the platform section, I jump into a platform near the top of the screen to slow down so that the lower platform can rise up and bring me back to the ceiling.

4-1: 14 presses.

Nothing needs explaining here.

4-2: 17 presses.

I take the normal beanstalk route because it'd take a lot of presses to perform the glitch. I choose not to moonwalk to the warp pipes after exiting the beanstalk as the 1 press I'd save trades off the 1 press I save by running up the stairs (Running takes 2 jumps, (moon)walking takes 3).

8-1: 29 presses

I take damage from a Koopa in order to run through him and 3 Goombas. This saves 2 jumps and I don't need to be big Mario any longer. I dodge a pair of Paratroopas using a wall.

8-2: 17 presses

I use the top Paratroopa to ascend the beginning stairs in 2 jumps. I then jump on the spring instead of over it in order to avoid the 3 Paratroopas. I then run into the last Bullet Bill cannon in order to spawn a Bill from the second to last cannon. I use this Bill to clear a gap at the end of the level without jumping.

8-3: 13 presses.

This level manipulation of the Hammer Bros. They jump based on timers that run when they're on the screen. They don't reset when they leave the screen, and the states of the timers carry on to the next Hammer Bros. spawned. Using this, you don't have to jump to avoid any of them. This part can probably be optimized better for speed, but this seems good enough for now, and I wouldn't know a way of proving it's optimal as the timers are a little chaotic. I again avoid fireworks at the end.

8-4: 27 presses.

Holding A when exiting a pipe makes you jump, which allows me to save a press going into the third room. I continue holding down while entering the water section as it makes Mario stop running regardless of the fact that I'm still holding right, allowing me to easily dodge the firebar.

Lua Script

To make the encode more aesthetically pleasing, FractalFusion provided a Lua script that displays the press count along with the input styled like a song on Dance Dance Revolution. The result looks something like this:
Dance Dance Revolution: Mario Mix
Just kidding. Here's an actual screenshot:
Mario jumping on a platform near the beanstalk.
This script is used to create all the provided encodes.

Special Thanks

Of course thanks to FractalFusion for creating this amazing Lua script. I also would like to thank him, along with MUGG, for creating some of the previous MP TASes of this game; this run obsoletes their latest collaboration. Finally, thanks to LexSfX for helping out with the encodes and not making fun of me for being the most prolific completely incompetent encoder of all time.
Happy April Fools' Day!

FractalFusion: Changed the branch.

adelikat: Claimed for judging.



Jungon
He/Him
Player (50)
Joined: 4/2/2009
Posts: 376
Location: Porto Alegre - Brazil
I vote No because there is already an improvement, to 143 presses,.. but ... wait, I voted Yes, now what? >_>' I think this should be tried in other games and other consoles, and if it don't make it to publication, it should invent a new 'Minimum Presses' forum and entire set of new movies... =D
*Jungon*/*Johnnyz* smilie weirdo =P ^^ o.o @__@ +D
Experienced player (702)
Joined: 2/19/2006
Posts: 742
Location: Quincy, MA
Brandon wrote:
OK, so I was able to complete 1-1 in one less press thanks to andrewg, and 8-2 in one less press thanks to...you'll have to wait and see. Assuming the rest syncs and I don't find any further improvements, that brings us to 143 presses.
I think I know what you did in 8-2 ;) I bet I do at least. I almost managed one less press in 8-4 by the way. I'm not sure if what I'm thinking is possible... but I really feel like it should be. I'll look into this deeper once I finish school. I just find this to be an interesting concept.
Super Mario Bros. console speedrunner - Andrew Gardikis
Editor, Experienced player (570)
Joined: 11/8/2010
Posts: 4036
Jungon wrote:
I think this should be tried in other games and other consoles, and if it don't make it to publication, it should invent a new 'Minimum Presses' forum and entire set of new movies... =D
There are some "Minimum Presses" and "Minimum Holds" TASes on Brandon's page here, and there's also the main Minimal Presses thread here.
Brandon
He/Him
Editor, Player (191)
Joined: 11/21/2010
Posts: 914
Location: Tennessee
Jungon wrote:
I vote No because there is already an improvement, to 143 presses,..
That's not a reason to vote no...I might just have a judge replace my submission file. :(
andrewg wrote:
I almost managed one less press in 8-4 by the way. I'm not sure if what I'm thinking is possible... but I really feel like it should be.
Care to explain? Perhaps by PM?
All the best, Brandon Evans
Patryk1023
He/Him
Joined: 3/1/2011
Posts: 288
Location: Inside out house.
Jungon wrote:
I think this should be tried in other games and other consoles, and if it don't make it to publication, it should invent a new 'Minimum Presses' forum and entire set of new movies... =D
Oh yeah. I need to come on and try in... Mario 2? Haha. An yeah, I vote YES.
<Nach> scrimpy is fretty with her sunglasses on I'm here. never visible.
Experienced player (702)
Joined: 2/19/2006
Posts: 742
Location: Quincy, MA
8-4 won't work actually... Well, I still need to test more, but I've nearly exhausted my ideas on it. I'll tell you if I make any progress.
Super Mario Bros. console speedrunner - Andrew Gardikis
Experienced player (702)
Joined: 2/19/2006
Posts: 742
Location: Quincy, MA
In 8-4, You can hold "A" going out of the pipe into the water area instead of pressing it again. (Unless that was just an error of DDR input), I think it was just an oversight.
Super Mario Bros. console speedrunner - Andrew Gardikis
Brandon
He/Him
Editor, Player (191)
Joined: 11/21/2010
Posts: 914
Location: Tennessee
andrewg wrote:
In 8-4, You can hold "A" going out of the pipe into the water area instead of pressing it again. (Unless that was just an error of DDR input), I think it was just an oversight.
Nope, that's another press...142 presses. I wonder if it might be possible to use a Cheep Cheep to get you onto the pipe after the Pirahna Plant has already retreated. It doesn't look very likely, though.
All the best, Brandon Evans
Experienced player (702)
Joined: 2/19/2006
Posts: 742
Location: Quincy, MA
I tried to eliminate the walljump in 8-4 by using the turtles, but I came about 1 block too short on the vertical, which was disappointing. hmmm...
Super Mario Bros. console speedrunner - Andrew Gardikis
Brandon
He/Him
Editor, Player (191)
Joined: 11/21/2010
Posts: 914
Location: Tennessee
andrewg wrote:
I tried to eliminate the walljump in 8-4 by using the turtles, but I came about 1 block too short on the vertical, which was disappointing. hmmm...
I can look into that. If it's really that close, it might be a matter of brute force...you'd be surprised how little room for error there is for these things, as you'll see in the new 8-2. Hey andrewg, you seem fairly interested in this stuff; after I'm done with the new Minimum Presses run of Super Mario Bros., would you like to try making a Minimum Holds run? That'd be very much appreciated.
All the best, Brandon Evans
Personman
Other
Joined: 4/20/2008
Posts: 465
Can the stairs to the warp zone in 4-2 be entered a la the second gif here in fewer than 6 presses? I don't have experience doing the trick, but the page seems to indicate that it should be possible, if tricky. If I'm counting right, it seems like you could do it in 4 presses: 1. Add L to decelerate and steer away once ejection starts 2. Jump into the stair instead of onto the top platform 3. Release L and add R to walk through the wall 4. Fall onto the pipe and press D, keep holding R. I'm sure I'm missing something, or maybe it's just impossible. Sorry if you've already tried this a million times/already know it's not doable for an obvious reason, Brandon :)
A warb degombs the brangy. Your gitch zanks and leils the warb.
Brandon
He/Him
Editor, Player (191)
Joined: 11/21/2010
Posts: 914
Location: Tennessee
Personman wrote:
I'm sure I'm missing something, or maybe it's just impossible. Sorry if you've already tried this a million times/already know it's not doable for an obvious reason, Brandon :)
I haven't messed around with any of these types of glitches, but MUGG tells me that this involves a lot of jumps and would not be worth it. Please correct us if we're wrong.
All the best, Brandon Evans
Experienced player (584)
Joined: 2/23/2008
Posts: 266
Location: CA, USA
Been a while since I TASed smb, but I think for glitching into say the top of a pipe you only need 1 a press and > being held constantly, while going through solid walls needs more jumps or it pushes you out (been a while again, not sure).
Joined: 2/7/2012
Posts: 15
I very much like the idea of a "minimum presses" category, as there is much potential for creative ways of reducing presses (as the present run exemplifies).
Brandon
He/Him
Editor, Player (191)
Joined: 11/21/2010
Posts: 914
Location: Tennessee
Bobmario511 wrote:
Been a while since I TASed smb, but I think for glitching into say the top of a pipe you only need 1 a press and > being held constantly, while going through solid walls needs more jumps or it pushes you out (been a while again, not sure).
Would you or anyone else be so kind as to test this? I am not the best with glitching, and I'm currently working on the other improvements. I have a huge incentive not to find a glitch improvement in 4-2, which would in turn require me to do a bunch of additional work, anyway. :P
All the best, Brandon Evans
Expert player (2573)
Joined: 12/23/2007
Posts: 831
If it would get published, then TASVideos would open a gate to welcome all kinds of "minimum presses" TASes, and soon there would be SMW MP, Megaman MP, and so on. We would probably need a counter for key presses for every submission, and "minimum holds" TASes would want to join in TASVideos, too. If it would get published, then I think we would have to change the defination for the word "TAS" again, since it's neither a speedrun nor a playaround. Glitches of this game has hardly been shown in this run, only a walljump or two. And like Tub said, it's too similar to the warped run of mine, just avoiding needless jumps and more bumping into walls, really actionless for such a cool action game looking from the outside. People do enjoy incredible speed and amazing entertainment in TASes, but do they really care how many buttons are pressed? The sum of key presses are not that meaningful in my opinion, although I always keep the key inputs in its neatest way, and some TASers like music playing with inputs, but those are for making movies perfect and surprising the audiences. It's not like solving a Rubik's Cube with the least moves, key presses are something that buried deep inside the game and shouldn't be digged out. Besides, a little more presses won't hurt your keyboard or your gamepad. Again, this is not a bad movie or an April Fool's submission, but forgive me, I just don't see it worthwhile to be published.
Recent projects: SMB warpless TAS (2018), SMB warpless walkathon (2019), SMB something never done before (2019), Extra Mario Bros. (best ending) (2020).
Emulator Coder
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
HappyLee wrote:
If it would get published, then TASVideos would open a gate to welcome all kinds of "minimum presses" TASes, and soon there would be SMW MP, Megaman MP, and so on. We would probably need a counter for key presses for every submission
That's the idea. As for counting key presses per submission, that's not really a big deal.
HappyLee wrote:
If it would get published, then I think we would have to change the defination for the word "TAS" again, since it's neither a speedrun nor a playaround.
Tool-Assisted "Superplay". It's "super", because it requires long range planning in finding how to trade multiple buttons presses for less. It also shows off things in the game you just don't normally see. Mario walking by Hammer Bros like they're not even there? They just jump out of his way? Or bouncing off lots of things with no effort expended. You don't look at the game as "oh I beat it in no time at all", it's more of "I beat this in my sleep, I barely even pushed anything". Or "Hey, I tied a rubber-band around the controller and walked away, and look, the game was beaten, this *game* is pathetic". Unlike fastest time, this really is about "effortless play".
HappyLee wrote:
People do enjoy incredible speed and amazing entertainment in TASes, but do they really care how many buttons are pressed?
I do. It's another metric to ridicule classic games.
HappyLee wrote:
It's not like solving a Rubik's Cube with the least moves
However, a Rubik's Cube competition can consist of doing it as quick as possible, in the most entertaining way as possible, or in the least amount of spins as possible. With TASing, we currently are missing this last metric.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Brandon
He/Him
Editor, Player (191)
Joined: 11/21/2010
Posts: 914
Location: Tennessee
andrewg wrote:
I tried to eliminate the walljump in 8-4 by using the turtles, but I came about 1 block too short on the vertical, which was disappointing. hmmm...
Can you provide me your input file in which you got this close? I've looked at the section and I have no idea how you would have even made it this far. I currently have a file that completes the game in 142 presses, which is roughly 2 seconds longer than the 145 press version. I'll be without internet until Sunday, and during this time, I'll play around with 8-4, but after that's done, I'll be going back to SMB3. Unless someone finds an improvement to 4-2 using a glitch by Sunday, I'm going to redo the subtitles, encode, and submit this. Actually, judges, should I just have one of you replace the submission file? Only 4 of the levels change at all, but the improvement in 8-2 is really impressive and might change votes. I'll heed your suggestion. I invite everyone to try to find an improvement in 4-2 or experiment with this idea of andrewg's for 8-4, though I must say, I really really don't want to have to redo 4-2...all of my enemy manipulation will have been wasted... Edit: Nach suggested I resubmit it, preferably as soon as possible. I'll try to get it done by tomorrow, but there's no guarantee.
All the best, Brandon Evans
Joined: 6/12/2008
Posts: 84
Don't take this offensively but I would really be looking forward to seeing this run from Happylee. I still fail to see why you made an extra jump at 0:15 when you can clearly slow down and walk through the pipe. I am also aggravated at how you jump high on the flag poles when you could probably save the viewers time by jumping as low as possible (on those flag poles). And I do think it can be further optimized on top of all that so...at first I thought I would abstain from voting but after making such a worthwhile comment, I am going to vote no, since meh would bother me after all that I said
Skilled player (1742)
Joined: 9/17/2009
Posts: 4984
Location: ̶C̶a̶n̶a̶d̶a̶ "Kanatah"
BrainStormer wrote:
Don't take this offensively but I would really be looking forward to seeing this run from Happylee.
Agreed.
HappyLee wrote:
If it would get published, then TASVideos would open a gate to welcome all kinds of "minimum presses" TASes, and soon there would be SMW MP, Megaman MP, and so on.
Yes, but if those TASes weren't entertaining enough, or looked too similar to a current published run, then it may not be accepted. I think...
Joined: 10/20/2006
Posts: 1248
Maybe there should be more discussion about possible improvements before it gets resubmitted, I'm thinking the first published minimum presses run should really be a very high quality one to get people interested in the concept. I'd also be interested in seeing a more detailed submission text, explaining why these are the minimal presses required, including possible alternative routes that have been considered. I know it might be asking for much, but a detailed submission text would greatly enhance the entertainment value I'd personally receive from watching minimal presses runs. The DDR input display is very fitting for them and should maybe even become the standard imo.
Expert player (2573)
Joined: 12/23/2007
Posts: 831
Nach wrote:
As for counting key presses per submission, that's not really a big deal.
As a matter of fact, it is kind of a big deal to me, because it would look stupid on the submission page.
Nach wrote:
It's "super", because it requires long range planning in finding how to trade multiple buttons presses for less. It also shows off things in the game you just don't normally see. Mario walking by Hammer Bros like they're not even there? They just jump out of his way? Or bouncing off lots of things with no effort expended. You don't look at the game as "oh I beat it in no time at all", it's more of "I beat this in my sleep, I barely even pushed anything". Or "Hey, I tied a rubber-band around the controller and walked away, and look, the game was beaten, this *game* is pathetic". Unlike fastest time, this really is about "effortless play".
Well, it doesn't seem super to me. I see lots of people running through Hammer Bros that way. It's a lot easier and safer. I often use it myself, not in TASes. Saving some presses ain't "cool", and getting the lowest score would make the game more "pathetic". And this run didn't "barely even pushed anything", actually 145 is still a great sum. Without the lua script, I don't think people would care or remember the sum of key pressing in a movie.
Nach wrote:
However, a Rubik's Cube competition can consist of doing it as quick as possible, in the most entertaining way as possible, or in the least amount of spins as possible. With TASing, we currently are missing this last metric.
I know, I'd very like to see a a Rubik's Cube soved with the least moves, but I don't like to see Super Mario Bros soved with the least key presses. The idea would kind of ruin my favorite action games. Maybe the idea works better with puzzle games.
Recent projects: SMB warpless TAS (2018), SMB warpless walkathon (2019), SMB something never done before (2019), Extra Mario Bros. (best ending) (2020).
Banned User
Joined: 12/14/2009
Posts: 386
Location: Santiago, Chile
At first it took me understand the main idea of ​​this TAS. It seems strange to try to avoid pushing buttons as often as possible. Without understanding the main objective I would vote NO, but in my opinion, does not seem very relevant for me this, also I think there were some unnecessary things. Sorry, but my vote is meh.
Editor, Expert player (2078)
Joined: 6/15/2005
Posts: 3282
BrainStormer wrote:
I still fail to see why you made an extra jump at 0:15 when you can clearly slow down and walk through the pipe.
Slowing down costs a press.
BrainStormer wrote:
I am also aggravated at how you jump high on the flag poles when you could probably save the viewers time by jumping as low as possible (on those flag poles).
It doesn't matter whether it is high or low; it takes the same amount of time.
Experienced player (702)
Joined: 2/19/2006
Posts: 742
Location: Quincy, MA
@brainstormer Actually, jumping at the top of the flagpole is faster due to the fact that once Mario drops off the flagpole, he accelerates quicker to the castle (thus, activating the timer sooner). The jump at 0:15 was made because he must hold B and right, but if he does not jump, he will just end up on the top of the pipe instead of dropping down and going through it. Yes, I'd say it could be improved time-wise, but probably not to a degree you'd imagine. Limiting the button presses puts a great restriction on how fast the run is completed. EDIT: In all honesty, I don't know if TASvideos should accept it. However, I'd be interesting in making these TASes, just because of some of the creativity involved. I mean, it's a great concept demo. In some ways, I don't think the walkathon TASes even deserved publication. It really is along the same lines. I mean, the restrictions in both cases are very arbitrary. However, they are both interesting concepts (to me at least), so I would definitely enjoy helping out and looking for solutions to these kinds of runs (with Mario 1 being a game I know well). Is it right for TASvideos? I'm not sure. Is it fun to watch? for me, yes, therefore regardless whether it's published on tasvideos, I'd be interested in these kind of runs. EDIT2: I mean, the fewest amount of jumps could prove to be a lot more interesting if it were done well. At least, I'd imagine. I think there's a lot of potential for creativity in TASing.
Super Mario Bros. console speedrunner - Andrew Gardikis