(Shortly after the conclusion of the Billy Mitchell score dispute in 2018, I went back to get up to speed on what I’d missed on the Todd Rogers dispute. And what I found was that others had uncovered an exploit that allows much faster times on Barnstorming, the very game I had written about previously. While I’m not much of a high score chaser, given my prior work, it felt appropriate that I be the one to breach the 30-second barrier on console. I also felt it was best to announce this discovery in the proper context, explaining how it does not absolve Todd Rogers of his past lies. As I expected, Todd’s supporters have indeed referred to this very development as evidence that video game consoles are magic mystery boxes capable of justifying any bogus high score claim anyone wants to make. Lastly, this is the piece that resulted in me becoming a Kotaku-certified speedrunner.)
Last year, I wrote about Todd Rogers’ fraudulent time on Barnstorming on Atari 2600. Today, thanks to a newly rediscovered exploit from fluddowich and subsequent TAS work by MrWint, Barnstorming is my first ever WORLD RECORD! 😀
PREVIOUSLY, ON BARNSTORMING
For those that missed the last round of this, here’s a recap. In 1982, Activision credited Todd Rogers with completing Barnstorming (game 1, novice difficulty) in 32.74 seconds. That would be, to this day, a world class time, which unlike his bogus Dragster time, is actually in the realm of possibility. Alyosha’s TAS of Barnstorming published last November clocked in at 32.69, in-game time, and only in the last three months have two players, Clay Karczewski and SteamedYams, managed to beat Todd’s original credited time with 32.72 (though a few had tied 32.74). Of course, this is Todd “The Fraud” Rogers we’re talking about, verified by the same people who verified his Dragster score, so who knows what really happened.
Oh, but that’s just the beginning! His 32.74 time got reprinted by Activision for a few years. But then, in 1986, it was misprinted as 32.04, allegedly the result of a coffee stain. That erroneous time was then reprinted from that point on. When Todd gathered his old scores and documentation to submit to Twin Galaxies in 2000, he just submitted the 32.04, which was technically documented. Because hey, why wouldn’t you get away with what you can get away with? (I meant that as a joke, but I’m sure someone reading this thought “Yes, exactly. If they verified it, who are you to say you didn’t achieve it?”)
In 2002, people started snooping around this score. Turns out, this time was hilariously impossible. In Barnstorming, moving up or down to avoid obstacles slows you down. Well, some people edited the ROM to remove all obstacles, and found that you still couldn’t fly that fast in a straight line.
TG brass soon showed up, in the form of head referee Robert Mruczek and chief editor Ron Corcoran. Ron spoke to Todd, and instead of coming to an understanding that the score was incorrect, they discussed holding an event where Todd would try to re-perform the impossible time live. Meanwhile, then-TG referee Wolff Morrow spoke up, saying that he had previously challenged this score behind he scenes. He did his own research, and in his words, “I came to the unfortunate conclusion that the time of 32.04 simply is NOT possible on a glitch-free run.” Years later, Wolff Morrow under a different username recalled how poorly he and his brother Mike (C64 referee) were treated for daring to question the legendary Todd Rogers. Yup, this was totally not the first time this score was questioned and then swept under the rug.
Soon afterward, Ron Corcoran (who, it is now known, is an entirely different kind of scumbag, currently serving a 20-year prison sentence for something you may not want to look up) got out in front of it, declaring that the erroneous time of 32.04 would be rescinded and replaced, not with 32.74, but with a 32.77 time they claimed to have on VHS. Todd, who was in the background for most of this, finally spoke up, saying “Sometimes in the jog of things second and third partys handeling such delicate information like high scores and employees who work for major companies do not have the 1) passion as we gamers have ,and 2) the will to obsserve what material that they are forwarding”. Those dang second- and third-parties, always messing things up for honest gamers like Todd “Self-Proclaimed King of Video Games” Rogers.
Ron assured the public Todd would do a public performance of Barnstorming at the upcoming CGE 2002, and that this performance would be captured on video and made available to the public. Well, the event came, and Todd never did find time to play Barnstorming during the event (even though he did find time to play both Dragster and Robotron). But they did get together after the event and, so the story goes, Todd achieved a 32.50 captured on VHS. Funny enough, this matches an earlier claim by Ron that Todd had performed that exact time of 32.50 at the previous year’s CGE, allegedly in front of three TG refs no less. (Recall that Alyosha’s recent TAS was 32.69 seconds.) Of course, this alleged video was never made available to the public as promised.
But this is the best part! That impossible time of 32.04? When talking about it to people covering it as a current event, it was all the fault of those dang third-party scorekeepers. But years later Todd decides, no, he really did get that time after all, probably on some magic prototype cartridge or something.
NEW DEVELOPMENTS
So now we get to current events. Hold on to your butts, folks.
It is in fact possible to get 32.50, 32.04, or even faster, on Atari Barnstorming. Last night, using an original Atari console and Atari joystick, I achieved a time of 29.97 seconds, now verified as a new world record at speedrun.com. (It’s in its own category, as it should be, but it is the fastest completion of any of the categories.) However, I did not go through ten barns like you’re supposed to.
While researching the Dragster dispute, I came across this mention of a recently rediscovered trick on Barnstorming. (I say “rediscovered” strictly because there’s no way nobody had found this in the last 36 years, even if references are lacking on the Internet.) I couldn’t track down the post Barthax referred to, but it was followed up with a link to this video by fluddowich showing the trick in action. Further research led me to this recent TAS by MrWint (a TASer most well known for breaking Pokemon Yellow wide open). With tool-assistance, MrWint used the trick to finish Barnstorming with an astonishingly low in-game timer of 15.22 seconds! (EDIT: I guess it’s worth making clear, these were done on emulator. The trick was not yet console-verified.)
Here’s how the trick works: The game operates on a barn counter, starting at 10. Each time you pass through a barn, the counter decrements, until you get all the way down to 0, and the game is over. It turns out, if you go just barely into the barn, that’s enough for it to count toward your barn counter. If you then move up and hit the roof of the barn just right, it sends you hurtling backwards. Normally that’s bad, but in this case it’s great, because the game still counted that barn, and you’re right there to go through it again (or try the trick again). TAS of course can do this with precision, but in real time, you can get some consistency by getting low and then flying in an upward trajectory into the corner where the roof begins, trying to hit just the right spot.
Because of the way the stage is set up, getting the trick once lets you skip three windmills that precede the tenth barn, whereas getting it multiple times on one run will have diminishing returns. The whole stage is always the same every time you play, but the number of windmills between each barn is different. For the other nine barns, skipping it would usually mean skipping one windmill. Granted, a time save is a time save. It’s just a question of how consistently it can be done.
“So wait, lemme get this straight. This means it actually would’ve been possible to get the times Todd Rogers claimed on Barnstorming, which we thought were impossible?”
That is correct. With the use of glitches, these Barnstorming scores, once thought impossible, actually are achievable. Facts are facts, and must be accepted as such. (I mean heck, if it weren’t for Todd Rogers lying about his scores on this game, this would just be a really cool discovery in its own right.)
And don’t forget, all Activision required back in the day was a polaroid! If someone knew this trick, yes indeed, it would’ve been quite possible for someone to send in a practically legitimate photo of a 32.04 time even back in 1982.
“Hold up, though. Does this mean Todd Rogers wasn’t lying after all?”
lol Heavens no! The hole Todd Rogers has dug with his own lies is far too deep for one new discovery to fill.
Let’s start with the fact that we know Todd Rogers didn’t get credited for 32.04 in 1982, because the records of the time credited him with 32.74. The time of 32.04 didn’t appear until years later as a typo. No matter what Rogers wants to say about “Nuh-uh, I really did get that time in 1982 after all”, no amount of cosmic rays allow him to play a video game simultaneously in 1982 and 1986. That’s what changing your stories does to ya!
Then you gotta talk about Twin Galaxies. Since the early days, TG has banned glitch abuse, and this would certainly fall into that category. (This is in contrast to the speedrun community, which allows glitches and exploits, using different categories to separate glitch run leaderboards from glitch-free ones, thus keeping competition fun and fair.) In the 2002 Barnstorming discussion, Ron himself said “If there is in fact a ‘trick’ that allows Todd to go faster than anyone else, I will be the first to disallow it as the Atari Editor” (whatever scumbag Ron’s word is worth). The far more scrupulous Wolff Morrow stands behind Mruczek in saying “a trick that allows you to go faster than is possible in the game, or gives you some sort of unfair advantage MUST be revealed to all gamers in fairness of the competition.” Of course, TG being what it was, who knows what would’ve happened if Todd had submitted a glitched 32.04 for private verification by his buddies (which he didn’t do anyway).
The 32.50 time he and Ron claimed was achieved at CGE is a bit more interesting. If Todd knew of and was using the trick, he could certainly have gotten that time. It actually wouldn’t have even been that great a time, given that I’m not a great Barnstorming player and I used the trick to get 29.97 without too much effort. But again, TG policy on glitch abuse comes into play. Even if the referees weren’t paying attention and didn’t notice, there’s absolutely no way for the player not to notice, due to the pattern of windmills. From your starting position, the windmills in between barns go 1-1-1-2-3-1-2-0-1-3. You don’t achieve records in this game without knowing that order, because at full speed you have to steer for the next landmark before you see it, and since steering slows you down, you never steer unless you have to. You pass one windmill before going into the ninth barn, and you go through that ninth barn knowing you have three windmills ahead of you. If the game suddenly ended, you the player would immediately know something was up. And that presumes the trick was done on accident, and not intentionally with the hopes your live ref didn’t notice or didn’t care. Personally I suspect the 32.50 time was still just made up (both with regards to CGE 2002 and the previous event Ron spoke of), but glitch abuse is now a credible theory as well.
That one claim is the only claim of Todd’s affected by this revelation, and it still doesn’t excuse Todd bragging about a time which his competitors are unable by rule to fairly compete with. Given TG rules, this simply suggests an alternate way he could have cheated.
Now, as those of you who followed the Dragster dispute know, there has been no shortage of straws grasped at by Todd and his defenders. (Yes, I am about to beat a dead horse, but some people still swear they can hear the horse breathing.) Todd’s supporters could point to this revelation, and claim there’s some secret undiscovered glitch-that-doesn’t-count-as-a-glitch in almost all of the games Todd made up scores for, secret tricks for which we have zero evidence but which we also couldn’t disprove without great effort. This includes Wabbit (which scores in increments of 5 and which kills the player at 1300 points, but somehow Todd got 1698), or his 65,000,000 on Atari Centipede (second place was 58,078), or his 960,001 score on Kaboom which Robert Mruczek described as “beyond insane”. Maybe they can even cover their eyes and plug their ears as Mruczek recalls Todd and Ron one-upping each other day after day on Atari Ms. Pac-Man. Or the old chestnut of “Sure, I guess I entered my own scores, but it was only once, and only the time I got caught.” (That was a paraphrase, of course.)
But Todd’s biggest lie of all stands testament. 5.51 on Dragster, aka “The Impossible Record” (photo credit to Zoyx), has been conclusively proven impossible by nothing less than a complete disassembly of the game code by Omnigamer (backed up by multiple peer reviews). Dragster game 1 is exceedingly simple, and is now comprehensively understood. That ship has sailed, and it ain’t never comin’ back. Again, Todd’s own lies do him in. It wasn’t enough for him to say he did it once. He had to say he did it three times, on three different cartridges and consoles, with alleged eyewitnesses (none of whom were ever found), and alleged evidence (which doesn’t exist). Todd Rogers gave us no choice but to know that he’s a liar.
So you know what? Forget that cheating fuck! This discovery on Barnstorming should be taken for what it is: A cool new skip on an Atari classic. Not only did we break new ground on a very old game, not only did we beat The Human Element^TM , but we beat the original coffee stain, 26 years later. Score one for the good guys!!
I do want to make clear one last time that this record time is for a glitched run of Barnstorming, and should not be competitively compared to non-glitched runs. Full credit to Clay Karczewski and SteamedYams for their tied performances of 32.72 seconds on glitchless, which I consider far more impressive than what I did. Credit again to fluddowich for the trick and to MrWint for the TAS work. Also, big thanks to my older brother for loaning me his Atari and showing me what all the little knobs on it do. (Hey, I’m old school, but I had a TI-99/4A instead.)