Super Mario Bros. Speedrun Record Beaten by One Second

Super Mario Bros. Speedrun Record Beaten by One Second

A speedrunner by the name of Blubbler has beaten Super Mario Bros. in four minutes and 57.69 seconds.

I am not very good at Super Mario Bros. I'm just fine at the rest of the series but every time I play the first game I usually wind up staring at the Game Over screen within the first ten minutes. That being the case, I can't help but be amazed at just how freaking good some gamers are at this game.

Case in point, a speedrunner going by the handle Blubbler has recently set a new a record for finishing the classic game in an incredible four minutes and 57.69 seconds. Blubber's run beats the old record (four minutes and 58.09 seconds) by the space of about a second. To cut that sliver off of his time Blubbler employed an impressive combination of practiced precision and time-saving techniques. For instance, he avoids triggering any of the end level fireworks sequences and, in level 8-2, he manages to skip Mario's post-flagpole victory strut by hitting the level's end at exactly time as a Bullet Bill.

All in all, it's a fairly amazing feat of gameplay and watching it from start to end I can't even imagine the amount of practice and preparation that Blubbler had to have put into the game to achieve this record. Even more than that though, I'd marvel to see the speedrun that does better than this. As the old adage goes, there's always someone better. Even so, it's hard to picture how anyone could do much better than this.

Source: Eurogamer

Permalink

This time is particularly remarkable in how close it comes to tool-assisted runs of the game. The current TAS record is 04:57.33, generally believed to be a literally perfect run that can't be improved on. To achieve something so close to that in real time, without the benefit of frame-advance, ram viewers, or other popular TAS techniques, is nothing short of amazing.

Sadly, it was still played on an emulator, making it ineligible for Speed Demos Archive and most other non-TAS record keeping sites, but it's still incredible for a non-TAS run.

P.S. Thanks

Tool-Assisted Speedrun, I presume? If so, then I don't consider it a speedrun as that's just cheating if you can not only pause and play frame-by-frame but also roll back frames if needed. And that is - in my opinion - cheating. It can be used as guide, but it certainly doesn't count for me.

Typo, opening line Stew. "4 minutes ad 57.69 seconds"

Also neat.

Bindal:
Tool-Assisted Speedrun, I presume? If so, then I don't consider it a speedrun as that's just cheating if you can not only pause and play frame-by-frame but also roll back frames if needed. And that is - in my opinion - cheating. It can be used as guide, but it certainly doesn't count for me.

A TAS is used for exactly what Covarr walked about - with the help of tools, we have an idea of what the PERFECT game of Mario looks like. It's not meant to be impressive in the same way that THIS speedrun is, it's supposed to be a technical analysis of the game, generally to help figure out how the REAL speedruns should go.

Bindal:
Tool-Assisted Speedrun, I presume? If so, then I don't consider it a speedrun as that's just cheating if you can not only pause and play frame-by-frame but also roll back frames if needed. And that is - in my opinion - cheating. It can be used as guide, but it certainly doesn't count for me.

Why can't we go one post about speedrunning without this arguement? No one seems to appreciate that tool assisted runs are not categorized with normal runs, and even those are criticized for exploiting bugs for the sake of speed. All to preserve some notion of 'purity' or the 'right' way to play games. Those of us who run games do so for fun or charity, not for fame. Those who do often burn out after a year or two.

Bindal:
Tool-Assisted Speedrun, I presume? If so, then I don't consider it a speedrun as that's just cheating if you can not only pause and play frame-by-frame but also roll back frames if needed. And that is - in my opinion - cheating. It can be used as guide, but it certainly doesn't count for me.

Nope, this isn't a TAS. They did play on an emulator, but it was played in real time, without any of the benefits offered by emulators. If it were a tool-assisted run, it wouldn't be record setting; TAS's have gotten this game nearly half a second faster for years.

At any rate, TASing a game is hardly cheating, considering tool-assisted runs are only ever compared to other tool-assisted runs. It's just a different thing entirely; they compete for the best route planning and the most thorough understanding of the game's engine, rather than physical reflexes. It's only cheating if someone tries to pass off a TAS as a standard speedrun.

P.S. Thanks

Super Mario Bros. only has five minutes of single-player gameplay? What a rip-off. ;d

MinionJoe:
Super Mario Bros. only has five minutes of single-player gameplay? What a rip-off. ;d

Still more than MGS: Ground Zeroes :3

OT: Damn, that's really cool. In some places it looked like it was PIXELS to death!

One question about speed-runs though, why do they count segments that you have no control over in the timer?

PsychicTaco115:

MinionJoe:
Super Mario Bros. only has five minutes of single-player gameplay? What a rip-off. ;d

Still more than MGS: Ground Zeroes :3

OT: Damn, that's really cool. In some places it looked like it was PIXELS to death!

One question about speed-runs though, why do they count segments that you have no control over in the timer?

Part of the techniques people need to develop in order to beat the game the fastest often times revolve around methods that shorten the length or bypass these unplayable sections.

Bindal:
Tool-Assisted Speedrun, I presume? If so, then I don't consider it a speedrun as that's just cheating if you can not only pause and play frame-by-frame but also roll back frames if needed. And that is - in my opinion - cheating. It can be used as guide, but it certainly doesn't count for me.

Nope, not tool assisted. This is a human played run of the game.

Also, tool assisted runs are a different thing than human run and not really comparable to human controlled speed running. It isn't cheating because it isn't the same. That is like saying it is cheating when Nascar racers use cars because they are not racing on foot like in the 100M dash. It is a completely different sport, they are not comparable, and they are never seriously compared. Similarly, tool assisted run are a completely different thing from normal speed runs and they are never seriously compared.

Incredible run (including a number of pixel-perfect jumps with those piranha flowers), but with that tiny a difference (less than HALF a second actually, OP), every frame counts. Considering slight computer stutters or imperceptibly small differences in load-times and similar, I'd have to guess this run falls within "the margin of error" of the previous run, if you will. Still very impressive, but really, really hard to judge when a few frames shaved off one way or another (including simply due to tiny differences in how the emulator and the hardware are running rather than any player actions) might be what makes the difference.

Trude:
Why can't we go one post about speedrunning without this arguement?

Because people love policing things they know absolutely nothing about.

WhiteTigerShiro:

Trude:
Why can't we go one post about speedrunning without this arguement?

Because people love policing things they know absolutely nothing about.

At least it isn't an any% glitch run. 90% of the comments would be about people complaining that it "didn't count".

I saw an absolutely fantastic speedrun of Ocarina of Time, I believe the current record for any %, that finished up in about half an hour I think. There were so many mad people in those comments, I tell you.

I'm having trouble finishing the first world in 5 minutes -.-
This is INSANE!!!!
Props to any speedrunner, this takes some skill to master.

Ftaghn To You Too:

WhiteTigerShiro:

Trude:
Why can't we go one post about speedrunning without this arguement?

Because people love policing things they know absolutely nothing about.

At least it isn't an any% glitch run. 90% of the comments would be about people complaining that it "didn't count".

I saw an absolutely fantastic speedrun of Ocarina of Time, I believe the current record for any %, that finished up in about half an hour I think. There were so many mad people in those comments, I tell you.

Yup, I remember that one. People just don't realize that there's several different ways to speed run a game. There's TAS's, any% vs 100%, glitches allowed vs no glitches. Like someone in this topic said, it's like complaining that the NASCAR races don't count because they use cars instead of running on foot.

It's a pity he used an emulator. The Speed Demos Archive does not accept runs done with emulators, only runs done with original hardware. When you get this good, you should set up a NES and a DVD recorder and run on that.

 

Reply to Thread

Log in or Register to Comment
Have an account? Login below:
With Facebook:Login With Facebook
or
Username:  
Password:  
  
Not registered? To sign up for an account with The Escapist:
Register With Facebook
Register With Facebook
or
Register for a free account here