A man accused of cheating at video games may lose his Guinness World Record

https://www.washingtonpost.com/news/early-lead/wp/2018/01/29/a-man-accused-of-cheating-at-video-games-may-lose-his-decades-old-guinness-world-record/

by danso • 8 years ago

96 92 8 years ago

washingtonpost.com

stefs• 8 years ago

i've never played the game, but if it's on mame, wouldn't a simple evolutionary search yield good results?

warp• 8 years ago

I really enjoyed watching Ben Heck trying to get a faster score with the help of Todd Rogers using simulated inputs? on the real hardware:

https://www.youtube.com/watch?v=qfXwSYzEjcE

awesomepantsm• 8 years ago

With only 5.1 seconds of input, you could brute force it if you really wanted.

EtDybNuvCu• 8 years ago

Could you?

The game runs at NTSC rates, which we'll round up to 60Hz. We'll also say that 6s of inputs need to be considered, and scores which are above 6s will be discarded. We'll finally say that the possible input states consist of the joystick in either left ("shifting") or neutral ("popping") position, and the gas button either up or down. This generates four possible input states.

The number of possible input states in total will be 60 * 6 = 360, and so the total input space is 4 * * 360 = 2 * * 720, which is a little big for brute-force exploration. Many of these states suck; after all, the total number of shifts is not ~300, but ~5. So a guided search is likelier to pay off.

Edit: Thanks HN for eating asterisks and having no math mode.

sp332• 8 years ago

That depends if the best result is similar to a bunch of slightly less-good results, or if the best is a wild leap that doesn't look anything like other good results.

NietTim• 8 years ago

Good!

woliveirajr• 8 years ago

Beyond the software analysis evidence, which speaks directly to Todd Rogers’s Dragster 5.51 score time

I didn't understand this sentence: it means that, looking at the code and calculating all possibilities of gear changing moments, it would be impossible to achieve 5.51 ? If so, why would any other evidence be necessary ? Or, if it wasn't such analysis (and I think that this one would be definitive), why wasn't this one done ?

jperry• 8 years ago

No, it's exactly that and it was done years ago. He's just been untouchable until recently.

watwut• 8 years ago

Why was he untouchable?

ranger207• 8 years ago

Yes, the analysis proves 5.51 is impossible, and it's been confirmed that 5.57 is the best possible time through the use of an emulator with preprogrammed inputs, meaning that it plays the game perfectly. Todd Roger's response is that the analysis failed to consider the "human element" and how the console would respond to that.

jerf3• 8 years ago

Ah, yes, the "human element", the same thing that leads people to believe they can beat a chess AI by "confusing it" by moving randomly. Still a surprisingly popular belief despite how easy it is to test the theory.

mackal• 8 years ago

Been Heck show also console verified the TAS (they ran a TASBot like machine to play)

The score has also been achieved by humans on real hardware.

tzs• 8 years ago

Yes, the analysis proves 5.51 is impossible, and it's been confirmed that 5.57 is the best possible time through the use of an emulator with preprogrammed inputs, meaning that it plays the game perfectly.

An emulator with preprogrammed inputs doesn't mean it plays perfectly. It just means that it plays reproducibly. Whether or not it plays perfectly depends on the preprogrammed inputs.

How did they choose the inputs?

Looking at the game manual, it comes down to deciding when to shift, and deciding how to use the gas pedal.

Shifting looks like it is easy to deal with in emulated testing. You have neutral, 1, 2, 3, and 4. You start in neutral. So, an optimal run will start with you shifting once at t=0. You cannot down shift, so your only choice when shifting is up, and once you reach 4 you can no longer shift.

From another comment, it seems the game works in ticks of 3/100th second. That means that for a 6 second race, there are around 1.3 million possible shifting timings, assuming that you cannot skip gears.

For using the gas pedal, it looks like if you are not red lined, there are no circumstances where it would be better to not give it the gas. When red lined but at a time when your shifting plan calls for staying in the current gear, I can't tell from the manual what you should do. It says if you go past that you risk blowing the engine. What it doesn't say is if you get more power when going past the red line, so that it may be worth it.

Assuming that red line is max power, then when at red line but not shifting optimum would be to modulated the gas so as to keep it right at the red line.

For each shift pattern, it should be relatively easy to figure out when the gas has to switch between steady and modulated. Putting it all together that would give 1.3 million candidate inputs, one of which should be optimum.

If going past the red line does give more speed (at increased risk of engine blow out), it is more difficult. If engine blow out is probabilistic, you have to worry about the possibility that someone could get insanely lucky.

At that point, you probably need to analyze how the random number generator works, and how it is used, and how user inputs affect it. One particularly worrisome possibility would be that the timing of user input affects it. For example, if whenever the user presses or releases the gas it uses a random number to pick from several sounds associated with that action, that could be bad news. It would mean that the earlier analysis that the gas would optimally in each gear start out as constant and then switch modulated at the red line is no longer applicable. It could be that there is a run where if someone releases briefly the gas during the "constant" phase, that happens to get the random number generator to a position where they will get very lucky on the engine blow out rolls later.

If that meant we had to try every possible gas pattern, instead of just "constant followed by modulated", we'd have a combinatorial explosion.

Probably the best to do in this case (the "past the red line gives more speed, and the random number generator has a lot of user provided entropy" case) would be to tweak the emulator so that the engine cannot blow out, change the gas pedal input to simply be full gas all the time, and see what the results are. That would at least give us a lower bound on what the luckiest person in the universe could conceivably achieve.

Superleroy• 8 years ago

Exactly, they calculated the best possible time and found out that 5.51 is not even possible.

Todd Rogers however has multiple impossible scores which are still considered "records" even though they can't possibly be obtained, some of them are stupidly obvious, like games in which the score increases in steps of five, yet he has a number not evenly dividable by five and such stuff.

I guess nobody really knows why he still holds these records.

tomku3• 8 years ago

You're exactly correct about the kind of analysis that was done. The code was analyzed, and the game was played frame-by-frame in an emulator to demonstrate perfect play. I believe some members of the speedrunning community also ran brute-force searches of the input space to make sure there wasn't some kind of obscure glitch that could trigger that wasn't obvious from the code.

The reason that Twin Galaxies didn't accept that analysis is because they're run by members of a clique of old-school gamers who stayed behind when the speedrunning community abandoned TG, and Todd Rogers is part of that clique. The analysis was largely done by Omnigamer, who is a well-respected speedrunner outside of TG but isn't part of the clique. TG's staff chose to believe their own in-group rather than a evidence-based challenge from an outsider until their hand was forced by a viral Youtube video[1] calling them out.

This kind of behavior isn't new for Twin Galaxies. They have a long history of mismanagement and corruption. Most of the times/scores posted on their site lack any publicly-viewable proof because until very recently, proof only had to be shown to TG staff "referees" so that people could keep their special techniques secret from the competition. Modern speedrunning eschews secrecy and requires publicly-viewable video proof for all runs.

[1] https://www.youtube.com/watch?v=8e-H4sEHB54

zimpenfish• 8 years ago

Alternate link for people who can't be bothered defeating WAPO paywalls: https://kotaku.com/infamous-atari-player-gets-world-record-r...

danso2• 8 years ago

Also worth pointing out the extensive writeup on Twin Galaxies (the score/records tracking site mentioned in the OP and Kotaku):

http://www.twingalaxies.com/feed_details.php/87/dragster-des...

Contains some interesting insights about the Activision's spotty history of archiving game records. The most surprising thing to me about all of this is how Dragster's creator, David Crane, feels so sure that the record is sound: "I have no doubts, then or now".

millzlane• 8 years ago

I find that if you right-click on the link, then open it in a private tab or incognito. It will allow you you to read the article.

doodpants• 8 years ago

I find that viewing the article with JavaScript turned off works too.

whiskyant• 8 years ago

Here's a good video summary about Todd Rogers: https://www.youtube.com/watch?v=8e-H4sEHB54

graphitezepp• 8 years ago

This video stands by itself for all you need to know about the situation. It is pretty clear cut that this Rogers guy 'scum bagged' his way to his prominence when gaming was in a quite less developed state.

xutopia• 8 years ago

For those who can't watch. Some games were scored in increments of 5 and he somehow got a score finishing with an 8. Many scores were ridiculous like 650000 for a game that goes in increments of 100 and the second highest video verified score was in the 90k range. Then the fact that all his records were referee verified instead of video verified. The referee is serving a sentence for child rape. Then there is the lie about the videos being taken by the police department and never returned which was proven false by a call to the police department. Then the fact that the record holder had direct access to the Twin Galaxies database and ent3ered his own records though he says he only did it once.

AdmiralAsshat2• 8 years ago

Then the fact that all his records were referee verified instead of video verified. The referee is serving a sentence for child rape.

While that's deplorable, it does not logically follow that a child molester is necessarily an unreliable referee.

danso2• 8 years ago

Ironically, in the case of the Dragster record, his exact time does match the nature of Dragster's clock being only precise to 3/100ths of a second:

Despite Omnigamer’s claim, as previously mentioned, Activision supposedly had created a machine-assisted run that determined 5.54 a perfect score, which is .03 better than Koziel’s model, thus making it impossible in his determinations. Crane explained to me that the “Dragster timer is .03 seconds per tick, so the difference that is being discussed is one tick of the game clock in a record set 35 years ago.”

http://www.twingalaxies.com/feed_details.php/87/dragster-des...

lifeformed• 8 years ago

Not even 650,000, it was 65,000,000!

Reference: http://webcache.googleusercontent.com/search?q=cache:0kQ1V1Z...

bbarn• 8 years ago

So, is the opinion that he just fabricated the photo, or that he had a knack for figuring these glitches out? Because as far as I know, glitching things is considered all well and good with speed runners, and lots of his other high scores imply some kind of glitching or another.

jerf3• 8 years ago

Glitching is not universally considered "well and good" by speedrunners. There are multiple categories. Claiming you have a glitchless run that uses glitches is cheating. You may also run in a category that permits certain glitches, in the no-holds-barred category that must be run on real hardware, or the tool-assisted category, or any other game-specific category. Same deal.

This is the current state of the world. 10 years or so ago and before, there was a much stronger presumption of glitchlessness. Though you'd still encounter the problem that rigorously defining "glitch" can be quite challenging.

tomku3• 8 years ago

Twin Galaxies is notorious for banning glitches completely and it's one of the many reasons the speedrunning community abandoned TG. If he glitched, his runs would be invalid by TG's own rules.

As far as glitching in speedruns nowadays, it depends. You play by the rules of the category you choose to run, which may allow no glitches, some specific glitches or all glitches. To give an example, let's take a look at a popular speedgame, The Legend of Zelda: Ocarina of Time[1]. The "Any%" record, meaning beating the game with all glitches allowed, is 17m09s. The "Glitchless" completion record is 3hr48m38s. Categories like "No IM/WW" and "MST" are glitch limit categories where specific glitches are banned and the times are intermediate between any% and glitchless. Less-popular games will have fewer categories, but if there's interesting glitches, there'll usually be at least "no/few glitches" and "more/all glitches" categories.

[1] https://www.speedrun.com/oot

xamuel2• 8 years ago

banning glitches

How can you even define glitches?

In the original Super Mario Brothers, you could dash and then crouch and then slide through narrow tunnels while crouching. It wasn't documented in any official manual or anything, but everyone did it. Is that a glitch?

unreal372• 8 years ago

There's no photo. There's no evidence at all, except that Activision accepted the score back in the day.

fortythirteen2• 8 years ago

"It's unethical!" cried the group so obsessed with protecting those in their inner circle that they regularly find loopholes to keep from recognizing legit records.

There are some good mods at TG, but I find that many, if not most of them, treat it like an elite social club instead of the supposed impartial record authority that they promote themselves as.

Jeema101• 8 years ago

One thing that has maybe not been considered is that the Atari 2600 is susceptible to what's known in the online Atari community as 'frying', whereby cycling the power off and then back on very quickly while the console is running can lead to weird changes in the program state.

Not saying this is what occurred, but it could be a possibility, depending on the code...

xamuel2• 8 years ago

Reminds me of a controversial technique discovered in the TAS community. Start saving your savefile, then cut the power mid-write. Usually creates a savefile with a bad checksum, but with certain timing on certain games, can result in playable corrupt savefiles which can be used for a faster win.

One of the reasons this idea is so controversial is the TAS community always assumes games should be played frame by frame, but this technique would open sub-frame "gameplay" (timing when you cut the power with much higher precision)

nightcracker2• 8 years ago

I used to do something similar as a kid, cloning Pokemon in Pokemon Gold by turning off the Game Boy mid-save.

lovehashbrowns• 8 years ago

There are some games where you can modify the program state to glitch to the end, like Super Mario World: https://youtu.be/vAHXK2wut_I The explanation for the glitch is really interesting.

But something like this has to be given its own category. There's a Super Mario World category that allows for that "Credits Warp" glitch. The other categories have different rules for what you can and can't do.

So it's possible that Todd did something funky, but that would still make his scores illegitimate.

dkonofalski• 8 years ago

While that could be a possibility, that would still invalidate his score. It needs to be a full run and the game resets its memory at the start of each attempt so there'd be no way to start a new attempt with a "fryed" console.

xelxebar• 8 years ago

This reminds me of the slimy feel I get about Billy Mitchell every time I see a video of him.

vanadium• 8 years ago

Reportedly, that's just a persona and he amped it up for the King of Kong film. But I definitely get where you're coming from.

acomjean• 8 years ago

The King of Kong film makes him look bad. He clearly fills a kind villain role.

In the movie extras he's seen delivering a Q-Bert machine to a excited senior citizen so she can practice for a tournament. Seemed oddly out of character from the way he was portrayed in the movie.

I don't know him personally, but he keeps showing up at video game events and clearly is enthusiastic about them.

throwaway675309• 8 years ago

The movie definitely makes him look a bit mustache-twirling Bond cartoonish but I've seen him at things like the Southern-Fried Gaming Expo and he does come across as pretty conceited.

CrystalLangUser• 8 years ago

TwinGalaxies had quite the extensive writeup on Todd's various cheating [1].

There was a staggering amount of evidence, including TAS (Tool Assisted Speedruns) proving that Todd's time is physically impossible. That blew up in a 274-page thread, but ultimately I believe it took Robert T Mruczek's writeup to finally ban Todd.[1] Not surprising as Mcruzek is part of the "old guard".

I think this reflected very poorly on TG and I have no intention of using them in the future.

[1] http://www.twingalaxies.com/showthread.php/175364-Dispute-Di...

dmix• 8 years ago

Why does it reflect poorly on TG?

Todd's "Dragster" performance is one that I have qualified as being one of the top five video game performances of all time within multiple forum posts and in several interviews.

I'd be curious to know what the other 4 performances were.

s4vi0r• 8 years ago

Because they accepted beyond ridiculous scores without any real proof, or even skepticism? In like every game this dude has a record in his score is impossibly higher than 2nd place. And these aren't even games or records which have been around for a couple weeks so that you can argue there's room for improvement.

JohnJamesRambo• 8 years ago

Wait what? I was with you until the last line and now I'm confused. Why does this reflect poorly on TG if they removed it when there was good evidence it was false?

nightcracker2• 8 years ago

They have fabricated records and turned a blind eye for decades to Todd's outrageous 'records'.

viridian• 8 years ago

The evidence had existed for games like centipede many years, and the dragster game had been taken apart and proven fake via the actual machine code for 6+ months. TG didn't act on it until one of the inner circle got involved, that is the shameful bit.

otakucode• 8 years ago

They refused to remove it after having the evidence of falsehood for a very long time... until they got the personal arguments of one of their 'old guard' members. In other words, they don't care about objective reality and care only about the emotional investment their friends have.

hanasu4• 8 years ago

I think it's at the very least equally likely that the PR beating they were taking as a result of the Apollo video led them to make the decision rather than Robert's guidance.

lulmerchant• 8 years ago

Since watching King of Kong, I haven't been surprised about any of the shady stuff I hear about TG. Also, I'd recommend that anybody who hasn't watched that movie to watch it, it's hilarious.

wnevets• 8 years ago

Rogers’s record on the classic Atari arcade game Centipede, for example, had been listed on the Twin Galaxies site as being 65,000,000. The second-highest score, 58,078, barely comes close

Wow and nobody thought that was a bit suspicious?

BoiledCabbage• 8 years ago

I don't follow this stuff, but Twin Galaxies deserves zero respect after this.

To be considered a reputable organization and not even question scores like this is laughable. And there were multiple this absurd.

fortythirteen2• 8 years ago

Yet try to submit legitimate high scores as an _unknown_ and you'll find them repeatedly dismissed on "technicalities".

Many of their arcade "high scores" for lesser known titles are easily broken by someone practicing for less than a week, but happen to be long held by TG regulars, with hundreds of "high scores".

hanasu4• 8 years ago

Most, if not all, of Todd's scores were entered by a single referee. At the time, each 'platform' (console, computer, arcade etc.) had its own referee that would verify scores. That referee clearly was not trustworthy. However, Todd's scores were as equally valid as any other score on the site - all verified in secret by a single other human.

There has been years of doubt about Todd's scores, among those by other gamers. TG was recently purchased by a new owner. That new owner launched a 'dispute' system whereby any score could be challenged and examined by the community. There was never a formal process until now that would have allowed for this, regardless of how many doubters there were.

ChuckMcM• 8 years ago

Interesting article. For as long as I've been playing games there have been people claiming they did things which seemed unlikely if not impossible. However, given today's cycle accurate simulators and computation power it seems like you could certainly answer the question of whether or not the software would ever get to a certain point.

534b44a• 8 years ago

The referee that decided most of the scores in prison for 30 years, he was a very close friend to the cheater [1]

[1] https://www.youtube.com/watch?v=8e-H4sEHB54

A man accused of cheating at video games may lose his Guinness World Record

Related Stories