30

I was reading an article on Polygon which was talking about a longstanding record for an Atari Game called Dragster. In the article it was talking about how people challenged the record by using "tools" to dispute how the record would be achieved without cheats.

Would anyone have an idea what tools they could be referring to? They talk about analyzing the code. The link in the article takes you to a spreadsheet which lets you change values. Also is this a typical practice for challenging records in video games?

I've added the article below for reference:

https://www.polygon.com/2018/1/29/16944736/atari-dragster-game-world-record-banned

  • 1
    Just to be clear from the article - his original record that caused this controversy was based only on his word? Like he made the record run at home and just told people about it? And now it's surprising that it can't be verified? I feel like I'm missing something. – JPhi1618 Feb 2 '18 at 19:14
  • @JPhi1618 You're not missing anything. You have it exactly correct. The guy who claimed the record is becoming very well known for his massive history of claiming completely false records for many arcade games. He was friends with the record keepers at Twin Galaxies, which is how this happened for so long. – ell Feb 2 '18 at 22:13
  • @JPhi1618 Todd Rogers' Facebook post (linked in the Polygon article) claims that his record time of 5:51 was "achieved in front of Activision staff" at a live event. – Robert Feb 2 '18 at 22:47
  • @JPhi1618 - He allegedly sent Activision a polaroid of the screen showing 5.51, upon which he was invited to an event where he repeated the feat in front of a live audience, and then a second time at a second event in front of another live audience. Again, that's Todd Roger's recollection as recounted to Ben Heck when they try to simulate the perfect run. Relevant part is 0:36-1:44. – Robotnik Feb 4 '18 at 23:53
28

The "tools" they mean is something that can be used to analyze the game. There are probably a few ways to do this, but one video here discusses using a RAM analyzer for the Atari 2600. What it does is when the CPU writes to the RAM, the RAM analyzer will detect this, and copy the information into a different memory array that they can look at further. They then use a microcontroller that examines the RAM, and based off the information in the memory, it can make decisions on how to play the game.

A second video shows just this. By tweaking their settings for the "simulation" if you will, they can run the game using specific timing for when to change gear in the dragster, control acceleration, etc. The reason this world record is so controversial is that even when using the most ideal computer simulated run of the game, it was still not able to tie or beat the world record run by Todd Rogers.

  • 1
    Thanks for the information. Is this standard practice for video game records? – Yamcha boi Feb 1 '18 at 16:01
  • 2
    @user9285752 I don't think it is for every game. It's just that this record is highly controversial, and so many have tried to debunk it. Perhaps the Guinness world record website may contain information about this. – Timmy Jim Feb 1 '18 at 16:12
  • 7
    @user9285752 As mentioned in the other answer, there is actually a fairly common practice for this, called a "Tool assisted speedrun". The technique is generally used for optimal runs with better inputs than humans can consistently produce. In this case though, it seems the tools also helped them dig very deep into the code; which is not necessary for a TAS (though it could help find new techniques and glitches). – JMac Feb 1 '18 at 18:01
  • 1
    Also they probably just calculated the lowest possible time by simulating perfect input using a computer. – BRHSM Feb 2 '18 at 16:20
  • 3
    @Yamchaboi No. The reason why this was done here is because Dragster is simple in comparison to more modern games. Also one reason why the record is so controversial is because there was no video of it. I suspect either the referee was bribed or they mistook 5.57 for 5.51 – MechMK1 Feb 2 '18 at 21:12
32

They're describing tool-assisted speedrunning, aka TASing. A TAS uses an emulator, allowing a player to make frame-perfect button inputs, simulating absolutely perfect play, and allowing any errors to be rewound and redone quickly.

In the case of Dragster, we already know that perfect, down to the individual frame level (shifting and accelerating at the exact best possible moment), results in a time of 5.57.

The 5.51 time submitted by the player in question has no known way to reproduce, has never been able to be reproduced by anyone even with a TAS. Omnigamer has provided a spreadsheet indicating exhaustively what different combinations of input results in which times.

  • 3
  • 2
    makes me wonder if the 5.51 time was due to a glitch in either the player's cart or Atari or something. – Wayne Werner Feb 2 '18 at 0:49
  • 10
    the general consensus is that he faked it – Saya Feb 2 '18 at 4:26
  • 4
    Especially because, as I understand it, he "just so happened" to be unable to reproduce it when anyone was watching. – Bagahnoodles Feb 2 '18 at 15:35
  • 4
    @Bagahnoodles It's a lot worse than that. I saw a youtube video on this recently. Apparently he seems to have a pretty longstanding record of very suspicious high scores that were "verified by referee", as in confirmed by another person instead of by tangible evidence. A lot of them are just nonsense, like scores in the millions where the next top scores are all clumped around 50000. He basically seemed to submit scores without really playing the games; and sometimes would be way off with point scales. Also, the guy who verified his scores was his friend; and he verified some of his own – JMac Feb 2 '18 at 17:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.