Billy Mitchell at the Pac-Man 35th Birthday Celebration, May 22, 2015, in Schaumburg, Ill. Photo: Barry Brecheisen/Invision/AP
A prominent Donkey Kong forum has removed some scores supposedly earned by legendary arcade game player Billy Mitchell after evidence that there was possible foul play. Mitchell's scores, which had set world records in Donkey Kong and Donkey Kong Jr., had been recognised by the forum for eight years.
Sorting out what happened the day he set his record leaves numerous questions.
On July 31, 2010 Mitchell recorded world record scores for Donkey Kong and Donkey Kong Jr. after a session at Boomers Arcade in Dania, Florida. He supposedly logged a score of 1,062,800 points in Donkey Kong and 1,270,900 in Donkey Kong Jr.
A dispute over Mitchell's scores recently came to a boil on the forums for Twin Galaxies,, an organisation that tracks world records and where Mitchell long ago served as a referee.
On the forum poster Jeremy "Xelnia" Young accused Mitchell of using the Multiple Arcade Machine Emulator (MAME) to record his scores instead of an arcade cabinet. Young announced that he was removing Mitchell's scores from the Donkey Kong Forum, a long standing community site dedicated to the arcade games.
"Short of live, time-stamped, complete footage (including full views of cab hardware) of the games in question, I will not be reinstating any of Billy's scores in question," Young said. "If this community (and others like it) are built on the idea of friendship through competition, camaraderie through our shared pains in pushing ourselves, our friends, and these games to their limits, then we must strive for honesty and legitimacy."
Twin Galaxies does accept scores achieved in a special version of MAME called WolfMame. Mitchell's scores, accepted as authentically occuring on an arcade cabinet, would not have gone under the proper verification process for emulated gameplay. If it was determined that the gameplay was emulated, it would not align with claims that the score was performed at an arcade.
Young's argument centres upon the way original arcade cabinets and MAME draw images to the screen when loading the game. While the original hardware loads levels from one side of the screen to the other, MAME loads levels in large chunks.
Comparing the loading screens on Mitchell's scores and comparing them to footage of arcade machines, Young concluded that Mitchell's 2010 scores were actually performed through emulation.
An image from Chris Gleed's 970k score. On an original arcade cab, images are drawn from one side of the screen to another.
Video for Mitchell's scores were directly captured from the game's screen and do not show him playing on a cabinet. Twin Galaxies submissions require players to show video proof that they were playing the game at the time.
At the time of Mitchell's score, a referee also needed to be present to validate the scores. The referee for Mitchell's 2010 scores was infamous Atari score-setter Todd Rogers who has been in the middle of a score dispute of his own.
Last week, following a case built against his alleged score in the Atari 2600 game Dragster, Twin Galaxies concluded the score was "impossible" and removed all of Rogers scores from their records, banning him from further participation. When asked by Kotaku to discuss witnessing Mitchell's record, Rogers did not reply.
According to former Twin Galaxies employee David Nelson, Twin Galaxies representatives met with Mitchell at the Big Bang event in Ottumwa, Iowa to discuss further verification of the score before adding it to their leaderboards. Mitchell planed to announce his records to the public during the event on August 6.
According to former Twin Galaxies employee Patrick Scott Patterson, Rogers presented Twin Galaxies with VCR recordings of the scores but said, to his knowledge, the tapes could not be viewed or verified at the time.
"I do remember there was talk of a video tape," David Nelson told Kotaku over the phone. "I think it was in Todd's possession. Did I see it? No. I don't know if Todd was hanging on to it for safekeeping or not letting people see it but even if we did get it there would have been no time to review this tape in the manner that it required. "
It is unclear what happened to the tape if the tapes truly existed and what may had happened to them, but what is clear is that the lead up to Mitchell's announcement was a contentious period.
In a phone call with Kotaku, Nelson described a talk with Mitchell wherein he outlined concerns with the verification process. According to him, the conversation was initially cordial before taking a turn.
"The conversation started getting a little ugly," Nelson said. "He seemed like his plans were being threatened." Nelson claims Mitchell threatened to make things bad for Twin Galaxies.
Both Patterson and Nelson recall a night of lengthy arbitration where Twin Galaxies representatives whether or not to take Mitchell's scores.
"This was a pretty big deal and I certainly wasn't going to make the call alone." Nelson said. "We talked about it for hours and we hashed it out."
"We were keeping [the scores] quiet with everyone else," Patterson told Kotaku via phone. " We held a secret ballot if we should take the scores. I don't know what the results were."
Complicating matters was the fact that Mitchell's Donkey Kong Jr record had already been broken prior to his announcement by Mark Kiehl. Kiehl recorded a score of 1,307,500 point score. Kiehl's record was not verified until after Mitchell's record was announced and eventually recognised by Twin Galaxies.
"He satisfied the requirement for the score," Patterson said. "Right after that Steve Wiebe legitimately beat Billy's score on Donkey Kong so Billy lost those scores anyway."
Footage showing the supposed arcade board switch between Mitchell's high scores. Wiebe surpassed Mitchell's record in September of 2010 with a score of 1,064,500. Still, doubts surrounded Mitchell's score due to the fact that he had allegedly beaten both games on the same cabinet with arcade board swapping in between sessions.
The swap, which was ostensibly done just to change which game Mitchell was playing, hatched wild theories that Mitchell could have swapped in a doctored board. A video supposedly showing the Donkey Kong Jr. arcade board being returned exists but no footage of the actual scores outside of Mitchell's direct feed capture have surfaced.
Another elite player, Wes Copeland, has accused Mitchell's Donkey Kong score as being bogus, citing what he categorized as a suspiciously fortuitous point gain from a part of the game governed by random number generation, or RNG.
"Most of Billy's points in his direct feed games comes from blue barrel smashes. These are completely RNG-based, yet somehow his averages are way above the mean," the former Donkey Kong world record holder wrote on Twitter. "This is evidence of splicing. Billy replayed the boards over and over until he got the right smash RNG to lock in his pace."
"For all we know, he did a white lie and did it on MAME or something on the computer," Patterson said. "I've had lots of questions over the years."
Meanwhile, David Nelson has a slightly more charitable opinion of what happened at the time.
"Did anyone have a reason to distrust his honesty or his ability? No," Nelson said.
Mitchell's direct feed recording does not show him playing the game nor does it verify the kind of joystick he was using at the time. Without additional video proof, Mitchell's score may now also be in jeopardy on the Twin Galaxies leaderboards.
"Twin Galaxies is in the process of fully-reviewing the compelling evidence provided by Jeremy Young to support his current score dispute case against Billy Mitchell's Donkey Kong score," Twin Galaxies commissioner Dave Hawksett told Kotaku via email.
"We will do this thoroughly and impartially. In the meantime we will continue to observe this discussion by experts in the community and will also examine any further evidence that may be provided during this review period. We are taking this matter quite seriously."
Kotaku reached out to Billy Mitchell, who did not respond after multiple attempts for comment.