Why Can't This World Record Be Beaten? "The Impossible Speedrun" 

What do you guys think? Let me know here: strawpoll.com/5w7c11e Hope you all enjoyed :)

maybe they thought the 7 was a 1

I once got a 5.49 but i was working for the CIA at the time and the footage cant be released for matters of national security.

I got a time of 5:32 in 1997. My goldfish was there, and so was my dog. Both dead, and no video proof. Believe me pls UPDATE: Holy shit, yesterday (May 7th, 2020) I replicated the record. I only needed to pop the clutch straight to third gear, then switched to second when the timer hit 0:57 and afterwards spam the clutch for insane speeds. My chinese action camera broke while recording, so no footage. Sorry.

Mr. Activision is actually the worst nickname ever.

I got a time of 1.34 I started by immediately shifting into 10th gear and doing a backflip with a frame perfect button press

You get my upvote for 1 simple reason: You made "Dragster" interesting. I had this game as a kid and it was fun for approximately 5.51 seconds.

He got banned :D and all his records removed. lol

Todd Rogers seems to have submitted a number of impossible records, if wikipedia is to be believed, such as 999,990 in a game that tracks score in increments of 50 with a maxium of 999,999, or 1698 in a game that tracks score in increments of 5, with a killscreen around 1300 points, so there's really no reason to believe he didn't fake this record too.

Had no idea Ron Jeremy was such a gamer.

It's literally impossible to get the time he did, people have dug through the code for the game

I was able to get a 5.51 on my calculator, started with a 6 then used the - key with a 0.49.

I'm here after he was confirmed to be cheating...

The Biggest Cheater in Gaming History?

Many people don't know this but I myself, achieved a time of 3.XX, at the tender age of 6 years old. It was completely on accident but happened to be in front of 56 uncontacted (until me) amazonians who drew the event with mud on wood. I have forgotten the exact location but if found, I am the only human being alive who can translate their words. I'm sure they would agree with me on my record but the wood may have rotted to dis-figuration.

If Todd keeps starting on 2nd gear he's gonna fuck his clutch up.

Darbian has just joined the 5.57 club

He got the 5.51 out of a supply drop.

If he was able to recreate it, it must be legitimate. However, if he used his console and his cart, maybe not. Guess we'll never know?

I think we’re missing a very vital nickname here: Todd Todgers

I honestly believe he got 5:57 and took a photo of the screen at an angle such that the 7 was ambiguous/blurry/pixelated and wrote "5:51" under the picture and the record keepers believed it based on his reputation and that the "1" digit with the diagonal tail, looks a lot like 7

I have a rather messy history with Todd. The most infamous of which was his Barnstorming record. At the time I was working for TG, I would play and review the games Todd had records for, and Barnstorming was the first one where I realized his record was simply impossible. I and others that suspected something was amiss, immediately got flamed and attacked for ever daring to question Todd's accomplishments. It was only after I hacked the code of the game and removed all obstacles that I was able to conclusively prove Todd's record was impossible. His time was FASTER than simply flying straight to the finish line, and when people saw this, we finally got some traction on investigating the record. However, we were still attacked with venom and vitriol whenever the question was raised, and often told "You don't know how good he is", "He'll do it live and you'll be eating your hat", etc. Well we were ready to eat our hats, but such demonstrations never happened. Ultimately, the referee finally had to admit he never actually saw Todd do the record, and merely recorded an "error caused by a coffee stain in a magazine". Keep in mind Todd was made aware of the debate over his record, yet never stepped up to say that indeed the record was misreported. He let it stand as is for decades, and his fans would violently defend his honor over it. As a result, I lost all trust in Todd, and indeed the friendship was broken. It still brings back very bitter memories for me as to how I was treated when I knew for a fact the record was bogus.

Nothing wrong with top ten lists, but this original, independently researched content is way more interesting to me. Never had any clue about any of this. Good work EZScape, this is your best video yet.

Why can't this world record be beaten? Answer: Because Todd is a lying cheater.

Guys I got a time of 0:00 in dragster. I have witnesses too including “Hitler but in an alternate universe where he did nothing wrong”, and “Disney’s good star wars trilogy” They’ll both vouch for me I swear

"You got the high score!" Input your name Todd: "hmm... T..O..D..D.. no wait.. "5 . 5 1 ..hehe yeah thats my name"

Maybe there's something about the emulator that makes it different from how the actual console operates?

Omni gamer just got 5.57 XD

I got 3.xx but unfortunately I smashed the vhs (with footage on it) and forgot my Dropbox password. I am the real King of the Video Games!

He got all the DLC upgrades

I guess Activision was not such greedy fuckfaces back in the old days.

He was stripped of his records this morning!

All these “beaten by Darbian” comments are hilarious… Darbian’s Time, a 5.57, is tied for WR, because it’s a legitimately possible time. Todd’s “record” was a 5.51, which has no proof behind even being possible. Darbian did tie the WR, but he didn’t beat Todd’s supposed “time.”

wow they still claim its a record, with no evidence he actually did it, times have changed man

In 1706 I beat this record! At the time I was at my grandmas house showing my dragster skills on her television. Her dog, me, Her cat and her herself witnessed the attempt. It was a time of -69.69 by using a wrong warp in which I forgot how to replicate by the time video cameras were introduced. Because of this I have no footage and I haven't beat 5.57 with the normal method. Of course, my grandma, her dog and her cat are dead and I'm only alive due to being frozen in ice for 312 years. I know this because I was told on a piece of paper. I'll try to shift into second gear and get an untied wr!

He’s a liar. He faked one of his dragster times. I️ don’t believe for one second he got a 5.51

Why don't they just have this guy do his trick now?

4:18 when hes not running for his life, john wick enjoys playing video games

5.51 is impossible. Its not even in the code. Many people have known he cheated, he cant even do it again.

Hey guys! I got a 5.51 in Terraria by summoning the Moon Lord before - Oh wait wrong game.

0:28 TheLegend27...

Maybe it was an earlier version/revision of the game which would enable you to start in 2nd gear?

Ah this video has aged like milk. If only we knew about Todd back then with what we know now.

I got a time of 5.46 back in 2012. The way I did it was holding the A button as I entered the race before building up speed before the race began for 12 hours, letting me go to an alternate universe and, by a series of precise inputs, obtained the 5.46 in half an a press.

It could be the physical hardware of the console or even the tv. I remember watching something about how some older consoles would actually run slower or faster depending on their power source. Maybe the way he moves his controller affects the inside of the console slightly. idk just some random ideas.

I love these videos, they're so interesting!

Wow. I didn't now he held the record for over 33 thousand years. Must be a really hard record to beat.

I love watching this video and then your later videos on Todd.

Emulation is well, emulation. I think if people could find a way to put perfect inputs on an actual Atari 2600, then we might get more information, if that hasn't been done before.

I think this was a publicity stunt, actually. Think about it, there's no proof that it happened outside of people at Activision, the time is literally impossible with how the game is set up, and this happened right around the time of the video game crash of 1983. Activision gets to promote their game, Todd gets fame as the "King of Video Games", and Atari gets to at the very least try to market to adults, now that adults can see another adult playing video games. This also ties in with the convenient facts that some of his runs turned out to be fake, and if you could actually start in second gear, you could get a time much lower than 5.51. So, either publicity stunt or console defect. Either is possible

I got a 4.29 but the timeline I was in at the time was disintegrated by Cthulu. Ask them, they can back me up.

Who is watching this after Todd Rogers' Guinness World Record was officially revoked?

5.57 HAS BEEN ACHIEVED BY OMNIGAMER!

my question is: why anybody cares about twin galaxies speedrun records anymore? specially with the current community of speedruning that streams everything to prove its autenticity rather than believes of an old dude's rant.

Darbian got 5.57 like a month ago

I love how chill Ben Heck was while utterly destroying Todd's legacy forever.

oh hey, that's my post at 2:28. shoutouts to dropbox for screwing over my avatar by removing support for public folders. i guess i'll vamp about my experiences with TASing this. the consensus at TASvideos back when i was working on it was that the emulation of BizHawk's Atari 2600 core was faulty in some way, something involving frame timing, but several people tested older revisions with a different frame timing without any luck there. a user named Alyosha then vastly improved the Atari 2600 core over several revisions, and each time there were changes to the core i'd go right back to Dragster and see whether or not i could push down the final time. if i remember correctly, three people posted files of 5.57 times using slightly different strategies, and i eventually spent nearly an entire day working out the most optimal shifting method through pure trial and error only to find out that i could still only get 5.57 times. i don't honestly think it's possible to get below 5.57 even on console without some fluke that i presume Todd was able to take advantage of. my memory's a bit faulty but i believe there's a thread on AtariAge somewhere where Todd posts a picture of a 5.5X time, i wanna say it was 5.57, and to my knowledge that's the only physical proof of a time below 5.61 outside of our TASes (which obviously don't count as human records). i couldn't tell you which thread, there are probably 5-10 threads there alone asking how to get good times and i read through each one several times trying to figure out what Todd did. that post of his shown in the video is burned into my memory with how many times i analyzed it. a couple theories i have: * Activision's simulation could have been programmed into the game directly and thus wouldn't have to deal with anything like input delay or human error, thus it may have been able to find a shifting pattern that only it could do to reach 5.54. my reasoning here is during my testing, i discovered a hacked ROM of the game on AtariAge called Dragrace, and someone had implemented an auto-shifter into that ROMhack, and after a few tests i was able to get a time below 5.50. i discounted it because of all the physical changes made to the ROM possibly affecting the timing (i could never find an auto-shifter for Dragster itself), but it could lend some credence to the theory * TASes start from power on and all our attempts would have been read by the game as the very first attempt made on the game after power on. i presume during verifications that it took Todd dozens, perhaps hundreds of tries to recreate the record, which leads me to believe that it could be something having to do with multiple attempts, perhaps the frame timing matching up with a perfect shifting pattern (since the atari 2600 runs in 60 fps but only polls input every other frame, with the occasional frame dropped) * on that same line of thought, maybe there's some faulty logic in the game where things get messed up after hundreds of attempts, allowing for a lower time than should be possible either way, i spent way too long working on this game only to get nowhere, so it's awesome to see Omni (who's helped me out immensely with several TAS projects in the past, thanks again man) putting in the kind of technical effort i never could in order to figure this out. as of right now i'm convinced it's not possible under normal circumstances, but knowing Omni i feel like it's not gonna take too long to get to the truth behind this.

i once had a 5.51 while sitting backwards with a pillow over my head blindfolded and eating pizza and drinking cola AND having a tattoo made on my arm while doing the eddy capo from tekken 3 movement while lifting my keyboard in one perfect frame and all this in 2nd gear!

"King of video games" 50 something year old who has a couple illegitimate records.

BANNED FROM TWIN GALAXIES LUL

Maybe it's some analog trick with the cabinet that just can't be calculated or accounted for in the code Idk man

If you cant recreate it after taking the human element out of it and performing a frame perfect run of the thing, then I mean... the record seems pretty sketchy to me.

It really tells somehing when you tell the cheater "We have proven that it is completely impossible to achieve this score" and he answers with "Well but I did anyway because I'm that good"

Apollo legends video was the video that finally pushed twin galaxies to do something.

i believe there will come a time where somebody will pull a record that no matter how hard anybody tries it cant be beat and that shouldnt negate that run from being a World Record

This score is currently being challenged now at Twingalaxies

This record was probably faked.

I would look at exact versions of Atari and the game that we’re out during the time he completed the record. It’s likely that the current version available for emulators no longer have the glitch that lets you start in second gear. Games in those days went through many revisions but seldom advertised this on the cartridge. So one person might have a rev 2 and another a rev 4 but you wouldn’t know it easily without a lot of effort. Edit: Turns out this was dug into to the bottom and it was a false record and two of twin galaxy’s employees were faking these records for years. Board is cleaned up now though.

Cool video, horrible ending song.

Excellent video. I'm pleased and impressed that you were able to dig up this information, even though there are still some holes from this taking place in the days before internet documentation. My gut feeling does say that Todd's time is probably due to some cartridge error, though without more information about the exhibitions of the run, there's definitely some doubts that can't be quelled. With that in mind, though, I have an alternative proposal: Is it possible that there is more than one publicly released build of Dragster, that are not well differentiated, ID'd, or documented? I know, probably not; but if it turned out to be the case, it may better satisfy the problems surrounding this record (especially if more details and verification of the exhibitions happens to come up). An alternate build could have the exploit Todd describes, or other, minute differences. Even if it was only produced in small numbers or distributed to a limited area, circumstance may make it more likely that all three runs were performed on that particular (theoretical) build, than of three separate carts having the same, specific error (if that were ever confirmed to be the setup). And it could explain why Omnigamer simply couldn't get a 5:51 out of the numbers or code: they were simply using the wrong build. How many people are really publishing dumps from their personal vintage game carts, let alone comparing them against the only known version? That build could be out there. The evidence, the transactions and documents that might suggest it isn't are probably long gone; and proving that something doesn't exist is a fundamentally flawed endeavor. It could take just one person dumping their cart and confirming it doesn't match the known version. I'm not at all saying this is the likely answer. Misreading a 6 or using the same, bugged cart for all three runs, or even that Todd didn't actually match the 5:51 during those exhibitions (but his claim deemed reputable anyway) are all more likely possibilities. I'm just offering it as a remote possibility that hasn't been mentioned. And certainly, if a previously-unknown build was identified, that's what I'd put my money on.

Awaiting updated video

Holy smoke! It’s a scavenger hunt!

my new plant manager at unilever is named Todd Roger lol thats just funny

Didnt he give himself "the king of videogames" title?

I got a 4.97 in 1986 but then my alarm clock went off and I had to go to school.

Ron Jeremy would never lie

Brb gonna go spend 300 hours and beat his record

Well if it is not possible then he did not really do it. Furthermore there seems to be no proof at all. Even more, he made specific claims about how he did it that are debunked. I think we can conclude it is a hoax.

Idk why but when you said that you hope we have a beautiful life it caught me so off guard that you didnt say "day" that it made my day thanks

Never attribute to malice what could be attributed to stupidity. It’s possible that Todd legitimately believes he got a 5.54 because of either a hardware error in the 2600, a cartridge tilt (I don’t know enough about cartridge tilts so that might be easily disproven), dirty screen, manufacturing error in the cartridge console or TV, or just misremembering. He was nervous people wouldn’t believe him, so he made up the 2nd gear thing, which is still him lying, but the 5.54 might not be. Either way, this is very unlikely, and I still believe he’s lying, but it’s worth looking into.

Darbian actually tied the record

Give back the Chaos Emeralds

Time for that update video!

IDK dude I like his mustache I think he's legit

Toddzilla more like cheatzilla

Todd gave himself the nickname of "The king of video games" not the media.

This has been in my recommended for like 6 months

There was a game called Grand Prix for the Atari 2600. There were only 4 levels on it and the 3rd level was fast but very short. I think I had the fastest possible time on it because every time I played it (after memorizing the level) I got the exact same time.

all of his speedruns have a pattern it's much higher than second place it's approved by "referee" it most likely is impossible due to a mechanic in the game EDIT: oh and they're all fake

dude looks like ron jeremy .

That's all very interesting. I wonder if there are any updates to this story.

You're in Polygon, dude! Congrats!

I feel like he lied but a few witnesses went with it for either better publicity or maybe they were even bribed

We beat his record after all. Every single one of them at once🙌🏻

"Why Can't This World Record Be Beaten?" it should be "Why Can't This World Record Even Be Achieved?"

Anybody know the outro? Link in description is corrupted I think

Fun fact: You could have watched the real Dragster world record 71 times during this video.

I'd like to point one thing out, as a programmer: emulators are not a substitution for the real thing. Until a hardware TAS is attempted, there's no reason to doubt (or accept) the run. Why? If you write an emulator, you quickly figure out the a "realness" problem. First, the computer isn't just a big matrix of digital gates. You'd think every gate combination is specifically coded to do one thing, but isn't true. To save on transistors, certain pins drive certain basic operations, like pin 10 is "this loads a value into a register." Additional pins will signal from where and to where. The 6502 CPU (which runs the Atari 2600) has "unsupported" operation codes. For instance, the $AF opcode causes a value to be loaded into both the A and X registers simultaneously. It crosses the gates of LDA (load A) and LDX (load X), and just so happens everything runs ok. Thus, $AF is nicknamed LAX; although, it's not a "real" opcode. But, that's not the end of it. Second, a computer is not so much a digital device as it is a pulse-oriented voltage regulator. As voltages are applied to the pins of the CPU, various gates open and close. It takes time for the voltages to settle. Machines often have to have "debounce" their inputs (sometimes hardware, sometimes software) to wait until the voltages are steady before accepting their values. Voltages fluctuating at just the right time means the machine can do very odd things, such as miss executing and instruction or corrupting memory. Super Mario 3 can be beaten two seconds by changing the controller values so fast that the game loop waits "forever" on the voltages settling. Due to the way the SMB3's game happened to be coded as a series of "fall throughs" for game state, the game falls through all the game modes to the last one, the game ending. I would find it entirely reasonable that an exactly timed controller input change causes an unexpected opcode execution.

"The King of Videogames" is self-titled btw.

I was able to watch this entire video in under 5 minutes. I started watching in second gear.