This post is part of a series, examining various myths and stories around Billy Mitchell’s claimed performance of Pac-Man in 1999 and his subsequent trip to the Tokyo Game Show. The first post in this series can be found here:
https://perfectpacman.com/2021/09/02/dot-one/
The supplemental material for “Dot Two” can be found here:
https://perfectpacman.com/dot-two-supplemental/
THE SPLIT SCREEN
Before we begin today’s installment, I wish to be clear that we won’t be focusing so directly on Billy Mitchell quite yet. You’ll see him a little today, but not as much as you might have expected. That’s because today’s installment is about gaming achievements in the ’80s.
In “Dot One”, we discussed a perfect score up to and including Pac-Man’s kill screen. While this final board is a fully known and understood entity today, in Pac-Man’s early years, it was shrouded in mystery. On board 256, the game becomes impassable due to a simple byte rollover. The level number was stored in a single byte (eight bits), allowing the game to track it as a number from 0 to 255. When you progress to board 256, the game adds 1 to 255, resulting in 0 (with the leading digit being discarded). Thus, the program attempts to draw level 0 for you. The game would actually succeed in doing so, if not for the algorithm that draws the little fruits in the corner:
The game attempts to use the early board algorithm, which says “Start with level 1, and draw the fruit for every level up to the current level.” This works great when that level number is 1 or 6, and not so great when that number is 0. The result is that the game draws 256 fruits, first along the intended row, then along the row below that, and then over the game board itself, starting in the upper right corner [S1]:
The byte rollover occurs again, so the fruit algorithm counter does eventually see it has reached “level 0”, and the board finishes loading, allowing you to play. The left side of the board is normal, but the right side (just over half the screen) is a semi-navigable soup of computer garbage. This has resulted in Pac-Man’s kill screen also being referred to as “the split screen”. This stray algorithm is also what creates the hidden dots on the split screen, which is why those dots reappear each time the board loads, when no other dots in the game regenerate that way.
More detailed explanations of the kill screen can be seen here:
https://www.youtube.com/watch?v=NKKfW8X9uYk
http://www.donhodges.com/how_high_can_you_get2.htm
In a 2016 profile with Great Big Story, Billy attempts to explain the split screen, starting at 1:20:
You play 255 boards. On 256 there’s enough memory left in the game for the left half of the board and not the right half.
https://www.youtube.com/watch?v=IoVvgSwPDYk
Here Billy was at a 2017 event in Minnesota:
There’s enough memory for the left half of the board, and not for the right half. And that’s exactly what it is.
https://www.youtube.com/watch?v=8u1Ez1SczYA
And from Exhibit A, at about 28:30:
We believe there’s not enough room on the board, in the memory for the entire board, and that’s why it’s garbled.
(Remember, if you wish to hear these for yourself, “Exhibits” A through E are linked in “Dot One”. A link to “Dot One” will be provided at the top of each installment in this series.)
Despite Billy’s confidence, it’s clear he doesn’t know the actual cause of the split screen. His answer might have been a decent guess back in the day, but this has been a known commodity for many years now. And it’s not as if his answer is quicker or easier to grasp than “The game tries to draw fruits everywhere.” [S2]
Going back to the ’80s, the question of the earliest discovery of the Pac-Man split screen has itself been a topic of some discussion:
https://forums.arcade-museum.com/threads/first-discovery-of-pac-man-kill-screen.394758/
For starters, an illustration of the split screen was included in the April 1982 book The Video Master’s Guide to Pac-Man, by Jim Sykora and John Birkner [S3]:
https://www.digitpress.com/library/books/book_vmg_pac-man.pdf
The authors seemed confident in the inevitability and impassability of the split screen (although they miscounted the number of “key” boards along the way). This impression was not shared by gamers at large around that time, as evidenced by newspaper reporting. Just prior to the publication of that book, on March 24, 1982, Redlands Daily Facts reported a Pac-Man score of 3,270,850 by Ken French. Ken lamented that he could have continued his game, had the machine not malfunctioned:
https://www.newspapers.com/newspage/698743430/
https://www.newspapers.com/newspage/698743434/
In April 1982, Kevin Fischer was said to have tallied five games north of three million before encountering what he called the game’s “ceiling” [S4]:
https://www.newspapers.com/newspage/688225694/
Another early Pac-Man master, Eric Schwibs, put up an impressive score just shy of three million on Saturday, May 1, 1982. The next week, the newspaper tried their best to explain what happened to Eric’s game [S5]:
https://www.newspapers.com/newspage/137619254/
The June 1982 issue of Electronic Games featured an early acknowledgment of the split screen, on page 32:
http://www.digitpress.com/library/magazines/electronic_games/electronic_games.htm
In late 1982, the split screen was even featured in a segment on the TV show Just Kidding, courtesy of Pac-Man ace Ricky Mori [S6]:
https://www.youtube.com/watch?v=EPN1GcLPG4s
The split screen phenomenon wasn’t widely understood right away. There are many references in contemporary newspapers and in early gaming magazines to the game “breaking down” or “freaking out”. In June 1982, a Charlotte News article titled “Pooped Pac-Man spoils record” told the story of 15-year-old Patrick Carr, who scored over 3 million before his game “just went crazy”:
https://www.newspapers.com/newspage/622648660/
Frank Breneman of Charlotte phoned Twin Galaxies with a score of 3,149,120 achieved at Tivoli Amusement Center in July 1982, with TG citing the previous record as 3,147,430 (lower than reported in the June article). In the Charlotte Observer, Breneman claimed to have two lives left as he hit the final screen, remarking “Half the screen just went out. The machine got hot, I guess.”
https://web.archive.org/web/20001003014325/http://twingalaxies.com/breneman.html
In August 1982, a Kentucky 14-year-old named Paul Reynolds racked up 3 million points on a Pac-Man cocktail cabinet before, as the Louisville Courier-Journal put it, the machine “went berserk”.
https://www.newspapers.com/newspage/110893071/
As reported in an August 24, 1982 article in the San Francisco Examiner, while playing on factory default settings (4 lives), the aforementioned Ricky Mori got to 3 million points before the game “coughed up its computerized lungs and called it quits”:
https://www.newspapers.com/newspage/461119198/
During 1982, TG’s officially recognized record on Pac-Man was (mostly) creeping up and up. As seen above, it was reported as 3,400,000 in Patrick Carr’s article in June, then 3,147,430 in Frank Breneman’s article in July (surpassed by Frank with 3,149,120). By August, TG’s official high score was 5,971,440, as seen in the coverage on Ricky Mori. Ricky was eager to try again on “a machine that can take the pressure”, confident he could beat that record if allowed to keep playing. The owner of the arcade Ricky played at, however, was skeptical of the authenticity of the 5.9 million score:
So if a perfect score on an original Pac-Man machine is 3,333,360, how exactly did Ken French score 5,971,440? [S7] How did Jeffrey Yee score 6,131,940, enough to receive a congratulatory letter from president Ronald Reagan? [S8] And how did Les Martin get into USA Today with a whopping score of 12, 719,060?
https://web.archive.org/web/20080409035024/https://www.twingalaxies.com/index.aspx?c=18&id=748
https://archive.org/details/joystik_magazine-1983-12/page/n63/mode/2up
Well, in Jeffrey Yee’s case, his poor performance on that segment of Just Kidding makes his claims seem especially dubious. Many in the classic gaming community have had fun with that video clip over the years:
It’s also hard to give Les Martin the benefit of the doubt when he specified that his 12.7 million point game was on a single token, and that it was achieved using a secret method of defeating the split screen [S9]:
https://www.newspapers.com/newspage/281480328/
However, while there was no shortage of fraudulent video game high scores in the ’80s, some of these higher-than-maximum Pac-Man scores used a different approach: When the game “failed” (i.e., when the player encountered the mysterious split screen), they simply put in another quarter and started again, stitching together each of their consecutive scores until they had lost the appropriate number of lives. [S10] (Original Pac-Man was before the advent of continues, where one could drop in another quarter to continue the same score.) The thinking was: Why should it be the player’s fault if their machine crapped out so deep into a solid game? This wasn’t necessarily “lying” or “cheating” (although it could be). Video games were a new thing. For many players, out there playing on their own, there were no official rules. Many considered it only fair to continue their marathon games in spite of such technical circumstances beyond their control. [S11]
Over time, the split screen phenomenon came to be more understood as an inevitable impasse. Some people, however, were a bit late on that news. Among them were the people at the Twin Galaxies International Scoreboard, in Ottumwa, Iowa. Future Pac-Man legend Tim Balderramos was the subject of a March 1983 article in the Rapid City Journal in South Dakota, featuring his Pac-Man score of 3,197,360. By this point, the official Twin Galaxies record was up to 9,980,420, by Doug Nelson. [S12] In attempting to reconcile such a high score with reports of the game “freaking out” at what Tim called “the end of the program”, a TG spokesperson suggested it was possible to simply clear that board and continue playing:
https://www.newspapers.com/newspage/350712815/
While Walter Day had always taken a keen interest in video games, for much of the finer knowledge of these games’ operation, it seems he was always reliant on the discoveries of younger players. In a 2017 interview with the Retro Hour, Walter recalled his travels across the country in 1981, prior to opening the Twin Galaxies arcades in Iowa and Missouri. Specifically, Walter recalled going from arcade to arcade, searching for a rumored Pac-Man master, hoping to meet this person and learn their gaming secrets (starting at about 33:30):
There was a rumor of someone who could get the highest score imaginable on Pac-Man, like hundreds of thousands. So I started going from arcade to arcade, trying to track down that rumor, until finally I found the arcade in a town called Sandy, Utah, where this guy allegedly could get these high scores. And when I came in and announced that I’d traveled far to try and find this Pac-Man expert, they all went cold on me and wouldn’t give me any information. They… for some reason, they wanted to protect the guy, or save his… or save him from having his tricks get out into the public or something like that. I wasn’t quite certain why, but they simply, you know, stopped me dead in my tracks. And I remember one guy smirking at me, cuz he… he knew I wanted this information and [to] meet this person and learn from him but they wouldn’t let me learn from him.
https://theretrohour.com/twin-galaxies-and-classic-arcades-with-walter-day-the-retro-hour-ep91/
Twin Galaxies would continue recognizing higher-than-maximum scores on Pac-Man into 1983, with Les Martin’s 12 million still considered the top score as of June:
http://www.videoparadise-sanjose.com/tg-all.htm
It would seem TG finally figured out that the split screen is indeed impassable in the summer of 1983, as reflected in their rules document for the Video Game Masters Tournament that August:
http://www.videoparadise-sanjose.com/tg-rules.htm
RANDY TUFTS
The split screen cannot be passed through normal game play. The game considers a board complete when a total of 244 dots (240 little dots plus four power pellets) are eaten, and there are not enough total dots on board 256 to achieve this. However, this didn’t stop gamers everywhere from fueling the pre-Internet rumor machine with all kinds of imaginative theories and guesses.
One technique for passing the split screen (outside of competitive play) is called “rack test” or “rack advance”. Arcade machines had a series of dip switches to govern things like difficulty settings, or how many quarters were to be charged for a single play, or to enable developer tools. On Pac-Man, when switch 7 was set to on, the game would auto-complete every board, allowing the player to skip ahead to any board they wanted to try out. [S13] Using rack advance to pass the split screen brings the player back to board 1, the cherry board, except the sprites are still moving at the ninth key speeds.
At least one player from that era, Les Martin, knew of the rack advance method. As the story goes, he claimed to know a secret way to pass the split screen, one which he did not want to show anyone. After arriving at the split screen, he asked everyone to leave the room as he performed his special magic trick. When they returned, they saw him playing on the cherry board, continuing his same score from before. [S14]
Meanwhile, throughout the year of 1983, Pac-Man was still a frequent topic of discussion in the pages of Joystik magazine. In the January issue, on the letters page, the editor remarked on what might be considered a “perfect game” of Pac-Man. [S15] Interestingly, they disregarded the repeating ninth key boards, framing the first twenty boards as the true perfect score challenge [S16]:
http://www.digitpress.com/library/magazines/joystik/joystik_jan83.pdf
Following that cue to page 70 brings us to a segment discussing the nature of the split screen:
By their language, it seems they understood there were hidden dots on the garbage side of the split screen (or at least had heard rumors to that effect), but didn’t know exactly how many or exactly where they were.
The author also tried to explain the nature and relevance of a byte rollover, even explicitly saying the fault occurs because the game is attempting to draw board “0”. The rumors about the split screen’s origin apparently continued to circulate, as months later Joystik printed notices in its October and November issues, reiterating the explanation of a byte rollover, and urging readers to understand that this was not some deliberate attempt by the game’s designers to limit high scores:
http://www.digitpress.com/library/magazines/joystik/joystik_nov83.pdf
Going back to the January issue, the inescapable fact of the split screen on the 256th board led to speculation on how best to approach the game competitively. One approach would be to expect players to get the maximum possible score on every board up to the split screen, for 3,326,600 points. Another would be to do what we would today call a speedrun to the final screen:
Some of this constructive talk was undone by an unfortunate letter printed in the April 1983 issue of Joystik, claiming that if you arrive at the 256th board with the numbers “256” in your score, you will pass the split screen and continue playing (a textbook example of a bogus video game rumor). [S17] Worse, rather than correcting the writer and reiterating what had already been explained previously, this rumor was affirmed by an editor who apparently did not read January’s explanation of a byte rollover:
http://www.digitpress.com/library/magazines/joystik/joystik_apr83.pdf
(Note: As we’ll see in a moment, a later issue would clarify that the editor used rack advance to test this theory, likely passing the split screen due to the rack advance itself.)
Fast forward a few months. That spurious letter in the April issue of Joystik had made the rounds, eventually being brought to the attention of a Canadian player and Pac-Man virtuoso by the name of Randy Tufts.
Randy wrote a letter to Joystik, printed in the September issue, where he explained his skepticism upon hearing this rumor. However, in his diligence, he tried setting up his game to enter the split screen at scores of 3,256,000, 3,225,600, and 3,025,600 (each with “256” in the score), and confirmed that the trick does not work at all:
https://www.digitpress.com/library/magazines/joystik/joystik_sep83.pdf
Stop for a moment and let that sink in. If what Randy claimed was true, he was such a master at the game that not only was he able to reach the split screen rather casually, but he was even able to do so while hitting specific target scores of his choosing, by skipping a desired number of fruits and/or ghosts along the way. And from everything we’ve heard from Randy’s colleagues, he was the real deal. Unfortunately, Randy provided no proof of these specific feats, but he did provide one photo we’ll see in a moment.
Randy continued his letter, elaborating on his thoughts that a “perfect game” of Pac-Man should entail getting every point (including fruits and ghosts) through the sixth key, which is one of the dreaded “one second” boards and the final board where the ghosts turn blue [S18]:
In fact, with the exception of his aforementioned runs testing the “256” theory of the split screen, Randy chose to just stop playing at the ninth key, as the long stretch of later boards were effectively no longer a challenge.
At first blush, this may appear similar to the sort of idle boasting we have heard from many players on many games. However, the Joystik editor chimed in, clarifying that they did indeed receive photographic proof for these perfect sixth key scores, as well as some photos of his split-screen games:
One unlabeled photo is provided in the magazine, showing Pac-Man parked in the standard safe spot on board 18 (the sixth key). The board has seven dots remaining, and the score is 70 points away from a perfect score through that board:
Photographs were the most compelling evidence of a score one could reasonably be expected to produce in that era. And unlike the alleged photos attributing an impossible time of 5.51 on Activision’s Dragster to multiple people (most notably Todd Rogers), in this case we actually get to see the photo and know that it unambiguously shows a perfect score through the sixth key minus the last seven dots (albeit with no lives remaining).
So we know Randy conquered all the blue time boards, and we have good reason to believe he could cruise to the split screen at will. Recall that, after blue time is over, the only points a player could miss are the “fruit” that appears twice each level, and collecting that is a simple matter. The only open question is how consistently Randy could do it without dying. This is not to say that Randy Tufts assuredly completed a modern perfect score of 3,333,360 – there’s no concrete proof that he did – but it is definitely to say he could have and would have, if that was the challenge set forth before him.
TODAY I LEARNED
It’s time to pause our jaunt through the ’80s, to watch the evolution of a lie in progress. The following is Billy as heard in Great Big Story in 2016, starting at about 1:50:
The designer of Pac-Man, not only did he never think anyone would get to the end, he didn’t know where the end was. He didn’t know at what point the game would run out of memory. Everything to him is completely baffling. They never saw the split screen until I sent it to them.
This claim was echoed by the unskeptical author of the video’s description:
https://www.youtube.com/watch?v=IoVvgSwPDYk
As we’ll discuss in a moment, Billy didn’t even begin playing Pac-Man competitively until the end of 1982. And we’ve already seen several English language newspapers from the spring and summer of 1982 reporting on the phenomenon. And of course, you have to consider the Japanese gaming community who are also trying to break the game. Are we really to believe the fact of the kill screen on board 256 didn’t get back to Namco in Japan? Do we really think no players anywhere sent Namco a letter asking about the screen? Do we think no distributors or arcade owners asked about this apparent flaw in their product? (Not that they would be unhappy that the game couldn’t be played indefinitely, but surely someone asked about it, right?) Are we supposed to believe the game developers didn’t at least once run through the rack advance feature they themselves built into the hardware of their game? Are we supposed to believe “everything” about the game is “baffling” to the people who literally made it, as if an early ’80s computer program is some sort of mystifying riddle box that can only be solved by some overly-confident teenage arcade-goer from Florida?
Common sense alone should bring a few chuckles and awkward stares in any room where Billy talks about being the one to introduce the split screen to Namco. But even worse, Billy often tells this story in the context of his trip to Japan in 1999. [S19] The idea that Namco were blissfully unaware of Pac-Man’s split screen for nineteen years should make anyone choke on their own unrestrained laughter.
But thankfully, for this one, we don’t need to rely on common sense alone. The split screen was among many features of original Pac-Man shown in the first edition of All About Namco, a book released by Namco in Japan in 1985, on page 29 [S20]:
http://archive.org/embed/namco-masterpiece-game-collection-600DPI
However, neither facts nor common sense stopped this story from gaining traction. Later, in February 2017, that Great Big Story video was the source for a Reddit TIL (“Today I Learned”), with the following title:
TIL the creators of Pac-Man did not now [sic] how the game ended as it was designed to run indefinitely. It was only when Billy Mitchell finished the game, after all its internal storage ran out, that they saw how it ended.
And from there, the lie simply got repeated, becoming louder than the truth. A week after the Reddit post, it appeared on the site Kickass Facts, copying that Reddit text, and citing the GBS video:
https://www.kickassfacts.com/25-kickass-random-facts-list-312/
This particular block of text has since taken on a life of its own, appearing frequently in gaming trivia notes, sometimes with slight variations [S21]:
That text was also copied by the site Interesting Facts World (basically a bot site lifting text from Reddit TILs), though I’ll give them somewhat of a pass for this gem on their list of “Billy Mitchell Facts”:
https://www.interestingfactsworld.com/billy-mitchell-facts.html
What’s particularly interesting about this item is the evolution. Billy can claim to have shown the split screen to individuals who, themselves, had never seen it on a live Pac-Man machine (assuming even that claim is true). Then, Billy can tell the story in such a selective way as to give the impression that it was the company as a whole who had never “seen” the split screen until he showed “them”. From there, people repeat the story they believe they heard, which of course Billy fails to correct. And he can ultimately disavow responsibility for this misinformation, as he can claim that nothing he said was technically inaccurate.
This bit about Billy claiming to have shown Namco the split screen is not the only time Billy tries to insert himself into Pac-Man history. In a 2001 interview with NGenres, Billy claimed to have originated or discovered nearly all his tricks by himself (again, despite not playing the game seriously until the end of 1982):
https://web.archive.org/web/20011211122729/http://hub.ngenres.com/pacman_interview.html
Not surprisingly, a check of Joystik magazine provides nothing to support this pointless braggadocio. Billy’s name does appear in Joystik’s published high score lists, provided by Twin Galaxies. However, while the magazine did credit players for tips and strategies sent in to them, Billy’s name was not among them, for Pac-Man or any other game [S22]:
Similarly, there’s an amusing interaction in this 2004 episode of Retro Gaming Radio (with the Billy segment recorded at CGE in 2003). At 1:11:50, Billy brags about being the originator of published game guides:
Some people say “Did you read the book?” Read the book? I wrote the book!
https://archive.org/details/rgry6/06+Episode+2004-08+Part+01.mp3
The astute host Shane R. Monroe then asks Billy if he is indeed a published author, to which Billy’s answer is a lengthy “No”. Billy subsequently claimed to have provided material to authors who did write such books. Despite the existence of many such books from the ’80s, no such consultancy credits for Billy have been found. (Of course, it would be rather silly to think that the Billy Mitchell we know today, and as illustrated throughout this series, was so freely sharing his strategies in that early arcade era.)
One other odd thing about that line from Billy is that it sounds an awful lot like a story told by his Donkey Kong friend Steve Sanders (author of The Video Master’s Guide to Donkey Kong), as heard in the film Chasing Ghosts at about 36:30 [S23]:
When I was playing Donkey Kong, invariably, some kid would go “Man, you must’ve read the book!” And my buddy would say “He wrote the book.”
But hey, even if Billy did borrow his friend’s story, I’m sure it was an isolated incident.
Back in Pac-Man’s early days, a trio of gamers from Montana – Tom Asaki, Spencer Ouren, and Don Williams, otherwise known as the “Bozeman Think Tank” – took Pac-Man and Ms. Pac-Man analysis to another level. [S24] Asaki’s name is invoked by Billy in a story for how the name “Cruise Elroy” (a popular nickname for the ghost Blinky) came to be. These are Billy’s words from his Twitch stream on July 17, 2018, while referring to himself in the third person [S25]:
The red guy goes faster, typically called “Cruise Elroy”. It’s universally known as “Cruise Elroy”. And no one really knows where that began. And the truth is, the term “Cruise Elroy” was born out of a conversation between Billy Mitchell and Tom Asaki. And it was said a number of times, to a number of publications, and it just sticks. Basically, it means when the red guy begins going faster than the rest. But that’s where it was born. Born out of a conversation between me and Tom Asaki. You could ask Tom that.
Oh, I can, can I?
Actually, I don’t need to. In 2010, Tom Asaki posted some of his old Think Tank notes to the Classic Arcade Gaming forum, shedding light on the actual origin of the term “Cruise Elroy”, which was introduced to the group by Spencer Ouren prior to their contact with the Twin Galaxies International Scoreboard:
At the time he first said the term, I remember asking Spencer where he got it. It was just one of those spur-of-the-moment ideas. Though, it must have been born out of some experience of his, and we may never know those details. Cruise Elroy was already a part of our regular vocabulary long before we ever made contact with TGIS.
http://www.classicarcadegaming.com/forums/index.php?topic=3214.0
NOW YOU’RE PLAYING WITH PATTERNS
As discussed in the last installment, the danger and exhaustion of playing Pac-Man freehanded led to the development of prearranged patterns. Larry Walker, a Mississippi player, hit upon this secret to Pac-Man success as early as 1981, as reported in the Clarion Ledger in August of that year:
https://www.newspapers.com/newspage/181157663/
In the September 1982 issue of Joystik, still prior to the aforementioned discussions of the split screen, the magazine included a set of Pac-Man patterns assisting players in clearing several of the early boards. Many of these patterns were very crude, including some pinpoint turns, like this turn on the fruit spot, which the magazine referred to as a “quick-pluck maneuver”:
http://www.digitpress.com/library/magazines/joystik/joystik_sept82.pdf
That issue also included a pattern for the ninth key board, which gets repeated all the way to the split screen. Months later, the April ’83 issue included five more choices of ninth key patterns of varying degrees of difficulty and swag. Four of these patterns were captioned as follows:
http://www.digitpress.com/library/magazines/joystik/joystik_apr83.pdf
The advent of patterns is what turned Pac-Man from a simple chaotic maze game into a puzzle that could be definitively solved. In the previously mentioned San Francisco Examiner feature from 1982, Ricky Mori openly acknowledged his discovery and use of patterns, and the role they played in his ability to achieve scores north of 3 million (while still at that time lamenting that his machine was unable to “take the pressure” of extended play):
On the other hand, as we saw above, Randy Tufts claimed not to use patterns in his “perfect games” (through the sixth key), instead relying on grouping techniques. (Note that as said above, while Randy was reportedly capable of reaching the split screen at target scores, his typical game consisted of playing through the first 20 difficult boards, and skipping the monotony of boards 21 through 255, which would take an eternity without the use of ninth key patterns.) The reason Randy preferred grouping was because, as he said in a screenshot above, patterns “can fail because of one little hesitation”.
Indeed, the challenge of many Pac-Man patterns is that sharp turns and reversals are required. Especially in early days, patterns printed in magazines required negotiating short pauses and other joystick flicks as well, many of which had to be taken at precise moments. What were sought, and ultimately derived, were continuous motion patterns, eliminating these possible points of error. This left the player with only standard turns, which can be taken uniformly by buffering inputs – in other words, by turning the joystick in the intended direction before you arrive at an intersection, so that Pac-Man makes his turn at the first opportunity. [S26]
1982 saw a glut of printed Pac-Man guides, including Mastering Pac-Man by Ken Uston, Scoring Big at Pac-Man by Craig Kubey, and How to Win at Pac-Man by Penguin Books. [S27] Going back to the Sykora and Birkner book from the same year, an impressive-for-its-time ninth key pattern was included starting on numbered page 55:
https://www.digitpress.com/library/books/book_vmg_pac-man.pdf
Aside from a short pause above the ghost pen, and a couple late reversals, this was nearly a full continuous motion pattern for the ninth key board, published all the way back in April 1982. (We’ll be referring back to this pattern much, much later.)
While a reliable ninth key pattern can take a player from board 21 to the split screen, the most sought-after patterns of all were for the six “one second” boards. Even decades later, top Pac-Man players were very secretive about these patterns, which at one time would have required a lot of work (and either money or unlimited access to a Pac-Man cabinet set to free play) to devise. Even in 2005, patterns for those boards were still not readily available on the Internet [S28]:
https://atariage.com/forums/topic/73042-pac-man-patterning/
For the especially ostentatious, the aforementioned April 1983 Joystik included a bonus ninth key pattern which, while clearing the board, also allows the player to achieve swag points by passing straight through two of the ghosts:
This is made possible due to the slightly flawed way Pac-Man handles collision detection. In short, the game board is divided into invisible tiles. If Pac-Man’s center location travels from tile A to tile B on the exact same frame a ghost’s center location travels from tile B to tile A, the game never considers them as having collided, and they appear to pass right through each other. While a player can achieve this by accident, a prearranged pattern can be developed to include any number of these passes-through, allowing the trick to be performed on demand. [S29]
This level of deconstruction of the game was a little more than the manufacturers intended. [S30] In the United States, Pac-Man was licensed to Midway, a division of Bally Manufacturing. In a March 1982 Tampa Tribune profile on Brian Duncan and David Wilson, two University of South Florida students who had authored their own Pac-Man guide titled Pac-Man: Are You Ready?, Midway spokesperson Larry Berke sounded almost crestfallen at how their smash hit was being so handily defeated:
https://www.newspapers.com/newspage/335642492/
https://www.newspapers.com/newspage/335642524/
http://www.steverd.com/whatbook/pacmanru.htm
That same month, the Fort Worth Star-Telegram noted the rise of printed Pac-Man guides empowering players to dominate the game for hours on end, as well as Ms. Pac-Man’s role in liberating arcade owners’ money makers [S31]:
The missus was developed largely to combat a rising fever in video game parlor operators alarmed over the development of what they held to be the real Pac-Man Monster: A habitue of the booth who, educated through patience or reading, was able to plop down one quarter and play for hours.
https://www.newspapers.com/newspage/635439196/
https://www.newspapers.com/newspage/637872692/
The following year, in the 1983 Rapid City Journal feature on Tim Balderramos, Bally spokesperson Jim Jarocki obliquely acknowledged the split screen phenomenon on the original Pac-Man, while noting that later games in the series were designed with added randomization in the hopes of outmaneuvering aspiring pattern-plotters [S32]:
HIDDEN DOTS
On July 8, 1984, a newspaper credited a Pac-Man player with scoring 3.3 million. As noted, we’ve seen reporting of scores that high from previous years. But this was noteworthy in that the reporter made a point to specifically say that this player had collected “every conceivable point available in the game” – a direct claim which we had not yet seen.
And you’ll never guess who it was!
https://www.newspapers.com/newspage/630478790/
“Wow”, you might be thinking. “Billy got 3.3 million on Pac-Man at a live event?”
Well, no. You see, that was the Video Game Masters Tournament, which at the time was how you qualified for Guinness. The following edition of Guinness cites this 1984 Miami event, along with several others held the same weekend around the country:
And it does list a Pac-Man score for Billy, but it wasn’t 3.3 million:
So it appears Billy bragged to the reporter about his previous high score while making it sound like that was the score he got that weekend at the event. Either that, or 2,478,610 was his real high score, and Billy just made up a score that sounded better. (Also, what’s up with that 2.4 million score? Did Billy not have a ninth key pattern?)
As for the Miami Herald’s reporting, the printed score of “3.3 million” was obviously rounded, as a semi-interested newspaper reporter is likely to do. But this still raises some questions: Did Billy actually achieve a perfect score in or prior to 1984? And if so, how was this “perfect score” defined? Did this “3.3 million” include the hidden dots or not? Or was this misreported? And if it was misreported, was it because the reporter misunderstood what was being told to him, or was it because someone was giving him bad information? Also (spoiler) if maxing out Pac-Man was supposed to be this herculean century-defining effort as we were all later led to believe, why was this reported in its day as being essentially a footnote?
Billy did not begin playing Pac-Man competitively until late 1982 at the very earliest. [S33] In this interview with the Kurt and Corey Show, Billy says he started playing Pac-Man “at the end of ’82” (heard at about 2:50) [S34]:
https://www.youtube.com/watch?v=gdd3uS-7NOE
And in this 2001 interview, Billy described the process of meeting Darren Olsen at the Life Magazine photo shoot in November 1982, which subsequently brought him to playing Ms. Pac-Man, and eventually Pac-Man [S35]:
https://web.archive.org/web/20011211122729/http://hub.ngenres.com/pacman_interview.html
In early January of 1983, just prior to the That’s Incredible! competition won by Ben Gold, Billy was already claiming the fourth-highest score on Pac-Man, without specifying what that score was [S36]:
https://www.newspapers.com/newspage/235549889/
https://www.newspapers.com/newspage/235550106/
While Billy may have started sparring with Darren Olsen, his true Pac-Man study partner in the 1980s was his local friend, Chris Ayra, a Pac-Man master in his own right.
There are few gamers on this Earth whom Billy will praise as much as he praises Ayra, as heard in this 2012 interview at about 1:30:
To be a little more honest with you, there was a guy, my very best friend. His name’s Chris Ayra, and I owe everything I am in the gaming world to him. Some games we played the same. Some games he played or I played, we didn’t really cross on. But the fact is the games that we did play, that we did cross on, I was always fighting to stay ahead of him, or I was fighting to catch him, or one or the other. And it went back and forth, each one of us, pushed, prodded the other one farther and farther. There wasn’t anybody that we played that we really competed with. We just competed with each other. And if it wasn’t for him, if it wasn’t for his skill, if it wasn’t for the rivalry that existed between us, very friendly, obviously, I would’ve never gotten to the gameplay level that I’m at. So when I say I owe everything I am in the gaming world to him, it’s true.
https://www.youtube.com/watch?v=_eiH2jWYgTA
Their mutual friend, Donkey Kong MAME technician Rob Childs, had no reservations declaring Ayra the superior Pac-Man player of the group:
https://www.youtube.com/watch?v=Zya6WY95wIA
Indeed, Ayra had a fair amount of Pac-Man related newspaper coverage of his own. In 1984, a local arcade owner was agog over Ayra’s ability to defeat Ms. Pac-Man:
https://www.newspapers.com/newspage/302216547/
https://www.newspapers.com/newspage/302216720/
(Note that Ms. Pac-Man’s kill screens are a different phenomenon than on original Pac-Man. We’ll see an example of this in “Dot Three”.)
Billy and Chris met in the spring of 1983… or at least, that’s what is commonly said. Chris affirms this in his signed statement with Billy’s September 2019 legal threat. However, in a recent interview, along with musician Jerry Buckner, Billy tells an elaborate and totally-not-made-up story about hearing the song “Pac-Man Fever” for the first time via a phone call from Chris (starting at 23:00 here):
But I wanted to tell you the story about the first time. A friend of mine, Chris, he and I learned the game together. And he’s a manager at a supermarket. And about 6:30 in the morning, before they were open, over the P.A., for the first time somebody played, and he heard the song “Pac-Man Fever”. And he’s running around the store, crazy, trying to figure out what it is, because he’s an absolute fanatic. And he calls me on the phone, and back then, it wasn’t so easy to get people on the phone like it is now, but he got me on the phone, and I could hear it blasting over the phone. And I just thought he was playing a trick on me. I didn’t think anything was going on. He was like “Yeah, there was this song!” And I go “There ain’t no song. I would’ve heard about it by now.” And he goes “No no, there’s this song.” So that was my first introduction to you, your song, everything you put together.
https://www.youtube.com/watch?v=CdATIvZn1vM
Billy is clearly talking about Chris Ayra, who he (by all other accounts) met in 1983, who he learned Pac-Man with, and who currently manages a supermarket. The problem, of course, is that the song “Pac-Man Fever” was released in 1981. The single peaked at #9 on Billboard in March of 1982, selling 1.2 million copies that year, so it wasn’t exactly a secret either.
At any rate, Billy and Chris started out grouping on Ms. Pac-Man, and then applied these techniques to the original game, which used many of the same mechanics as its sequel. From there, they started exploring original Pac-Man’s split screen, and tracking down the hidden dots required to potentially boost a score beyond the standard maximum of 3,332,820. Recall Billy’s words from Exhibit A, at about 2:20:
So we had it all mapped out. We had the absolute secret as to what the perfect score was, and my… me and my friend Chris, who did it with me, who’s… gets all the credit with me, told the formula, this secret, this score, and we gave it to Walter Day, as the scorekeeper. Nobody else knew it. We didn’t have anything to worry about. Nobody could figure this out. Not a chance. So, that was December of 1983.
So what was the secret that allowed Billy and Chris to solve the mystery of the hidden dots?
It’s simple: As teenagers (at least according to Billy), they owned a Pac-Man cabinet! Here’s Billy in a recent interview with Pac-Man Entertainment, at about 18:30:
In 1983, because some arcades were beginning to close, I was able to get an arcade cabinet. Yeah I did have one, in 1983 for a while, and that’s when we made the hardcore advances and strategies and scores and knowledge about perfect games and things like that.
https://www.youtube.com/watch?v=J0feW53grOI
So Billy went from not caring about Pac-Man at all in 1982, to him (i.e., his parents) buying a Pac-Man cabinet a year later. I suppose that’s one way to have an advantage over all those other arcade-goers. Despite Billy’s insistence that he doesn’t know anything about arcade hardware (and thus would have no idea why all his “direct feed” DK tapes turned out exactly like MAME), he clearly does have some experience with the inner workings of a Pac-Man cabinet. In Exhibit A, starting at about 14:20, Billy steps aside from the stage, goes behind the Pac-Man machine they have on display, and starts flipping through the switches to pause and to rack advance the game:
https://www.youtube.com/watch?v=MfNvVnG_cZk
The story of how Billy and Chris used this private machine to explore the split screen was discussed in Billy’s 2006 profile in Oxford American [S37]:
https://www.oxfordamerican.org/magazine/issue-53-spring-2006/the-perfect-man
The author seemed to be under the mistaken impression that Billy and Chris had to play through all 255 regular boards of Pac-Man each time they wanted to explore the split screen on their private machine. I can’t imagine who would have given him that impression.
Back to the beginning of Exhibit A, at about 1:20, Billy elaborates on their split screen research (with the “clear sheet of plastic” this time becoming “a paper”):
And so the fact of the matter is, we got a machine, we laid a paper over it, and we actually played “Oops, there’s a wall.” We actually mapped it out. And then we said “Ooh, I just ate a dot.” “But it’s not there.” “Oh, the dot’s invisible.” But we heard it, we saw the points. So it counts as a dot. We went through this long, drawn out thing in order to determine what it was, and how many dots, and how many points were available.
They took an especial interest in two apparent dots on the garbage side which are visible, but do not actually register as dots. Billy sometimes tells a story of leaving his Pac-Man game on for a month, with Pac-Man vertically looping the board, going over the same dots over and over. Billy described this during a July 2019 Twitch stream (starting at 6:03:50 in the YouTube upload):
So people say “Here, eat these two.” They don’t eat. So I tried to find ways to have ’em eat, but again, they don’t eat. I went like that, and I thought if you run over ’em long enough, eventually they’ll eat. They don’t eat. So, one day I had it in my room like this, in the ’80s, and I left it like that and I went out of town. I came back a month later and it’s still doing that. So I think it’s safe to say they don’t eat.
https://www.youtube.com/watch?v=1sRM4GDud1Q
Gosh, I hope Mr. and Mrs. Mitchell weren’t too upset about that power bill!
As an odd aside, Billy has also claimed to have left his Pac-Man machine running for an entire month on two other occasions. [S38] One of these was in the park spot on a regular board, and one was on a special park spot on the split screen (i.e., with Pac-Man not in motion). Did he genuinely think something new would happen after, say, the first week? Or does he just like saying he left his game running for a month at any opportunity he can, because it sounds impressive and because nobody will be able to prove otherwise?
Going back to the two visible non-dots among the split screen garbage, Billy discussed them with G4TV (at 8:00):
Because there were two dots on the split screen that, although you could see them visually, you can’t eat ’em. So you hear all these silly things in the magazines, “Oh, you gotta run over them 256 times, then they’ll be eaten.”
https://www.youtube.com/watch?v=KGJGEo9I7B0
Wait, they talked about the dots in magazines? I thought these hidden dots were supposed to be Billy’s big secret. [S39]
If they shared this secret of hidden dots with Walter Day, it wasn’t reflected in Pac-Man’s official rules, at least not right away. The game rules for Pac-Man for the 1984 Video Game Masters Tournaments, which qualified players for the 1985 edition of Guinness, included these remarks which indicate no knowledge of or consideration for regenerating dots [S40]:
http://www.videoparadise-sanjose.com/1984vgmt-rules.htm
It’s debatable whether this could be construed as dissuading players from exploring the split screen. But it’s certainly confusing if one were to collect the regenerating dots. Do you report how many lives you had when you reached the split screen, or how many you had after sacrificing them all for extra points?
One might say “Well, the hidden dots were supposed to be a secret.” (Although that shouldn’t excuse the scorekeeper from actively giving competitors the impression the dots were out of bounds if they were, in fact, a secret weapon for knowledgeable players to use.) However, this secrecy argument doesn’t really hold up either. In 1998, Twin Galaxies printed its first physical record book, and on page 120, it just tells you what the perfect score is:
A-ha! There’s that 3.3 million score of Billy’s! Except… not. That Miami Herald piece was printed in 1984, and this score is listed as having occurred in his hometown two years later in 1986. In fact, we were unable to find any record of this score anywhere prior to the 1998 TG book. (Also, the Herald said Billy had gotten “every conceivable point available in the game”. A score of 3,312,100 is not perfect by any definition.) But hey, maybe Billy really did get to 3.3 million eventually. [S41]
(Note that, while the ’98 book included a tacit acknowledgment of the nine hidden dots, it instructed players to use 3+1 settings, allowing a maximum score of 3,333,180, 180 points shy of the true maximum. We’ll talk more about settings in our next installment.)
Billy was not the only one claiming he had a later-overlooked perfect score on Pac-Man in the ’80s. Chris Ayra said the same thing in a Twin Galaxies forum post in 2003, bringing the split screen’s hidden dots into the perfect score equation:
https://web.archive.org/web/20041216105206/https://www.twingalaxies.com/forums/viewtopic.php?p=10223
This post is actually a big bag of worms, for a few reasons. First, the reason Chris says he and Billy were unable to do perfect scores at the 1984 Guinness event – the one which the Miami Herald cited having a “Pac-Man competition”, where Billy claimed to have done a 3.3 million score but was credited with a 2.4 million score – was because, according to Ayra, they didn’t have a Pac-Man machine at all. Also, how did every Guinness event they attended year after year never have a Pac-Man machine? Wasn’t that game literally everywhere?
Second, Chris recalls the score being in 1987 while the TG record book (as seen above) cites it as 1986 (although they both have it in Alabama). But hey, maybe they’re both wrong. Third, Ayra reiterates what Billy said, that only the two of them plus Walter Day knew about the hidden dots until 1998. Fourth, his score, 3,324,730, is actually a bit odd mathematically. Given his story about missing a key and an energizer, and the fact that small dots are the only things that increment your points by 10, that “30” at the end says he failed to collect all the regular dots on the split screen, regardless of whether he went for the nine hidden dots or not. (In fact, Billy’s score ending in “00” would also indicate an inability to finish the split screen.) [S42]
Elements of this story were reiterated by Chris Ayra in the form of text messages to David Race, which David shared with us in our research:
But going back to the forum post, obviously the big bombshell here is this:
In late 83, Bill and I achieved Perfect scores on Pacman with Bill being first by a matter of weeks.
It’s not known what offline controversy or drama may have been occurring around the time of this post which compelled Ayra to publish this account so publicly and permanently. But he has repeated this basic story over the years. Current Pac-Man champion David Race recalls being told by Ayra that both he and Billy had achieved perfect scores on Pac-Man in the ’80s (as we’ll elaborate on in “Dot Four”). [S43]
In 2004, a CAGDC author identifying themselves only as “The Mole” recalls being told a similar thing by CAGDC administrator Mark Alpiger:
Indeed, Mark also told me that Chris had told him this when they talked via phone in the mid-80’s, but, Mark wasn’t into Pac-Man, and didn’t understand the significance of it. Indeed, he even forgot that Chris told him, until he reviewed his notes last year.
http://www.classicarcadegaming.com/ts/index.htm#07032004
Alpiger elaborated his skepticism about the notion that no one else could have figured out the secrets to a perfect score:
While I’ll admit that it’s not easy to develop patterns that allow a player to get all possible points, specifically when the ‘blue-time’ is short, it’s incomprehensible that, given the popularity of the game, that no one had, at the very least, a perfect score, up to the split-screen. I’ll admit, maybe they didn’t have perfect lives (tho, again, with decent patterns, it’s not overly difficult to do), and / or maybe they were too dumb to figure out about the nine hidden dots on the screen. Still, anyone with access to the machine could rack advance, and try out everything on the split-screen, and easily figure out the hidden dots, which are a big key to the perfect game…
In Dwayne Richard’s documentary The Perfect Fraudman, Pac-Man perfect score player Donald Hayes (seen in King of Kong, slamming his car trunk) also recalled conversing with Chris Ayra, and coming away with the impression he and Billy had done perfect scores (starting at 1:22:40):
I found that out first-hand. We actually discovered something, I’m actually sworn to secrecy… well, not really, but… We actually discovered something on Pac-Man on the reunion game that hasn’t been made public yet. So just even based on that, I know there are things that they keep to themselves, because that’s the way they are. They have these techniques or these secrets, and they want to keep ’em to themselves, for whatever reason. But yeah, they both had the knowledge, I’m convinced, Chris definitely had the knowledge back in the ’80s and he actually, from what I recall, I’ve talked to him a couple times on the phone, I’ve talked to him in person, and from what I recall he indicated that he had the knowledge and the know-how to do a perfect game, and I think actually did it, back in probably ’83 or ’84.
https://www.youtube.com/watch?v=BSFtDVuGxL8
Included in the Chris Ayra text messages David Race provided us was a reiteration of this claim, along with a specific date for this alleged achievement, and a reference to the scores being done at an open-all-night arcade:
Billy’s September 2019 legal threat to Twin Galaxies included a signed statement by Chris Ayra, where he entirely avoided the issue of whether a perfect score on Pac-Man was achieved prior to 1999 [S44]:
Of course, not surprisingly, Billy tells this story a little differently (at about 4:10):
Well the truth is – and I don’t share this often, so this is kinda cool – in 1983 I learned everything I needed to know that had to do with a perfect score. Because it’s not a matter of just eating and collecting every point. At the very end it comes down to literally like a detective game. You had to know and research and it was a lot of work. And me and my friend did it, me and my friend Chris. And I gained all and everything necessary to learn and execute that perfect score in December of ’83, him in January of ’84 a month later. And we just sat on our hands, we just, we showed Walter everything, and so he knew the secrets, even though he didn’t write it down.
https://www.youtube.com/watch?v=6jI6w9P6RUE
Read that again, and think about it. It feels like Billy almost gives the game away here. What exactly did you two do that was so special in December of ’83 and January of ’84, sir? Why did this “detective game” take Chris an extra month? Is he a slow reader?
One of my research colleagues found this interesting as well:
Now it’s no surprise to any of us to hear Mitchell take all the credit of getting the knowledge first – that’s what he does. But when it suddenly brings into question previous statements you have to ask yourself – What’s going on here? Who do I believe? Particularly when you bring to the table Billy’s flawless memory! How can Mitchell claim that both he and Ayra told Walter Day about the dynamics of a perfect score in ’83 when Ayra didn’t “gain the knowledge” until the next year?
That said, there are also reasons to think the story of undeclared perfect scores in the ’80s isn’t true, or at least isn’t true the way it was told. The timeframes given by Billy and Chris don’t seem to match. The story that they couldn’t find a Pac-Man cabinet anywhere to do this score on (including at the 1984 Miami event where Billy was credited with a 2.4 million score) also doesn’t inspire confidence. Same for the story that TG’s closure kept them from submitting the score to anyone else, including a newspaper. Chris’s claim of them having done the scores at an all-night arcade would seem to be a reference to their nearby home arcade of Grand Prix in Dania Beach, Florida. It must’ve been tough to keep these big public scores a secret, which they wanted to keep secret because… reasons, I guess?
While Billy and Chris were both credited with scores in the ’80s that either reached the split screen or came close, neither demonstrated the ability to complete that split screen in competitive play. (This could in fact be a result of relying on freehand grouping techniques, which may not work properly when half the game’s layout is scrambled.) This isn’t to say Billy and Chris definitely could not finish the split screen, just that they did not demonstrate that ability in any documented way. There’s also the possibility that either of them could have achieved a perfect score by an earlier definition (for instance, perfect through the blue time boards, or up to the split screen but without the hidden dots), while not having achieved the later recognized perfect score of 3,333,360.
In fact, contrary to the many statements above relayed from Chris Ayra, in a June 2000 interview with the Back in Time webcast, he spoke of a recent rivalry finally inspiring them to achieve perfect scores, which he suggested he and Billy had not achieved previously (starting at about 1:02:30):
When we were challenged by Rick and John and them, the Canadians, a couple years back, that motivated us, and inspired us to higher levels, in which we… we knew we could accomplish, but we never did, cuz… like Bill said earlier, we never had any competition except one another. And there was other heights on, like Pac-Man, for example, Ms. Pac-Man, that we never quite reached. But once this rivalry came about a couple years ago, it inspired us and motivated us enough to finally achieve the perfect game, which Bill was first at it, and then later on I achieved it myself.
https://archive.org/details/bit06152000h
And even if Chris and Billy really did do perfect scores by today’s definition, it could be that the terms of these scores are not what Chris has claimed they were. It’s possible only one of them got the full perfect score, or that Ayra in fact got his first but in his later retellings deferred that honor to his more insecure friend.
Regardless of whether Billy or Chris did get a perfect score in the ’80s, it is reasonable to think they may have at least had that notion on their minds around December of 1983, months after the topic was discussed by Randy Tufts in the pages of Joystik magazine. Of course, Walter Day is always willing to take things a step further, as he did in this 2001 feature in RePlay Magazine:
But the person who is most luminary is Billy Mitchell, who pioneered the idea of getting a perfect game on Pac-Man.
Ah, so it was Billy’s idea all along.
BILL BASTABLE
Remember how Chris Ayra said he and Billy never told anyone other than Walter Day the secret of the nine dots?
Enter stage left, Bill Bastable, the true Pac-Man champion of the ’80s. Clearly a competitive person, Bastable was putting up top scores as early as 1982.
Don’t believe me? Ask Bally-Midway:
The letter is real (aside from my blocking out the old address, of course). The “perfect score” in question was a score of 3,332,820, achieved on September 6, 1982. But while the letter is real, we should also be true to what it represents. Bally-Midway said themselves they didn’t track high scores. They likely couldn’t have told you what exactly a “perfect score” on Pac-Man was at that time without someone else telling them first. (Granted, this was when official scorekeeper Twin Galaxies was still authorizing much higher scores on the belief that there was some mysterious way past the split screen.) Note also the reference to someone else already submitting a “perfect score” to them before Bastable. That said, this is a fantastic artifact of gaming history, and a nice acknowledgment directly from the U.S. manufacturers of the game. [S45]
This letter would become a point of controversy in 2017, resulting in a damage control campaign spearheaded by Billy’s friend Triforce. [S46] This included the following late night phone interview with Billy:
https://www.youtube.com/watch?v=M9EYBfhJIB8
Here we discover Billy Mitchell’s theory of the Bally-Midway letter, starting at about 21:10:
There’s not a prayer that Bill got a perfect score in 1982. None. Zero. He forgot to explain that to us in all the conversations we had? Okay. Without lying to you, he was nothing but a gentleman, okay. And he was a great player, okay. But he forgot to explain to us that he got a perfect score? He forgot to explain to us that he had a letter, okay? He showed us various pictures of different high scores, and he forgot to show us the one of a perfect score? Okay? I don’t believe in the tooth fairy, not for a long, long time, okay. In 1999, in July, about a week to two weeks after I did the perfect score, he called me, he congratulated me. We were on the phone for more than an hour. During that call, I put Chris on the call. It was a three-way call. There was nothing but pleasantries, okay. And again, Chris and he had far more communication than I did, okay. Bill was writing books, and putting together various things that were impressive in regards to Pac-Man and Ms. Pac-Man. He somehow forgot to share all that information? No. Okay, again, the tooth fairy left town a long time ago, okay. When somebody flattered me to the highest degree and decided to spend $100,000 doing a movie about me, or about me and Walter, okay, somehow convinced Bill how wicked we were. Well after that, unfortunately, we haven’t really been friends, but I hope the very best for Bill. He’s a great player. He’s a terrific player. You’re not going to convince me that he has that letter, and of all people in the world, he decided to turn it over to a select few people. Give me a break. Bally-Midway never wrote a letter like that, ever. I had discussions with Namco, because Namco is the one who licensed it to Bally, and I had discussions how they never, ever did anything, they didn’t have a promotional department, they didn’t do anything to foster or create any goodwill or legacy for Pac-Man. They were simply interested in collecting quarters. Everything you see about Pac-Man now, okay, every time there’s a cartoon, a toy, that was never ever done by Bally. It was only ever done by Namco, okay. That letter was created by somebody you know very well.
It’s funny, because there are so many more legitimate angles Billy could have gone with here. Billy could have pointed out that the definition of a “perfect” score on Pac-Man evolved over time. He could have said “I have video tapes confirming my perfect score, and Bastable does not.” (We’ll get to Billy’s tapes and what they do or do not include at a later time.)
But no! Billy goes right to “Fake news”. He accuses either Cat DeSpira or Dwayne Richard of manufacturing the letter on behalf of Bill Bastable to be used as fake evidence for evidently no other reason than to tear poor old innocent Billy Mitchell down. [S47]
Gosh, that sounds so familiar. Where have I heard that before?
Also, I fear Billy may have made up the bit about Bally-Midway never having a hand in anything other than “collecting quarters”. [S48] A 1982 New York Times piece on Pac-Man sublicensing quoted Bally-Midway marketing vice president Stanley Jarocki (after being unable to reach anyone at Namco America), noting that these tie-in products were expected to bring in more revenue than the game itself:
https://www.nytimes.com/1982/02/16/business/pac-man-sublicenses-extend-bally-s-profits.html
There was even – guess what – a Pac-Man cartoon, with a nod to Bally-Midway, and no mention of Namco [S49]:
You may have noticed the Bally-Midway letter does not explicitly list what score Bastable got. But don’t worry, further newspaper reporting gives his top score from the early ’80s as 3,332,820. That includes every available ghost, every fruit, and all of the normal dots on the split screen (though not the hidden dots). Here’s reporting on another perfect score – yes, he did this multiple times – in the Daily Register from Red Bank, New Jersey, on June 21, 1983 [S50]:
https://www.newspapers.com/newspage/517304401/
Even if for some reason you doubted these early scores due to lack of material proof, this demonstrates Bastable understood what the full perfect score was (minus the hidden dots) back in 1983. The following year, a very chillax looking Bastable reiterated this score in a July 12, 1984 profile in the Asbury Park Press, while also detailing his extensive knowledge of Ms. Pac-Man [S51]:
https://www.newspapers.com/newspage/145994579/
Furthermore, we have documentation back to 1982 that Bill Bastable knew of at least some of the hidden dots on the garbage side of the split screen. Here he was in the Daily Register, on November 18, 1982, in an article talking about the military draft:
http://www.newspapers.com/image/516877093
Bastable did collect three dots on one of his high score runs [S52], but he didn’t consider those extra dots to be particularly important. Around the time of the 2017 controversy, Bastable joined David Race for a phone interview for YouTube. At about 22:00, Bastable conveyed that his interest was in “whatever is actual Pac-Man, or looks like Pac-Man”, adding that, on the split screen, “the right half of the board doesn’t really look like Pac-Man”. At about 19:10, Bastable explained what he considered to be a perfect score:
I didn’t have the game at my house, and I wasn’t in a position to explore the right half of the board with the fact that I would have to go into the arcade, and I’d have to spend my money. When I first started out, it was about three or four bucks in the game. I got the cost of that down a little bit. And, you know, when I got to the blown board, I might have seen some dots under all that bullsh*t, but I didn’t want to get caught up in it, so I concentrated on everything else that’s a normal, you know, maze, or normal board, which means all of the left half of 256, and everything that preceded it. And I was able to do that six times while I was living at that Shader house at 112.
https://www.youtube.com/watch?v=-bZbs7Ma9QE
It seems as though Bill Bastable was a bit too good for the rankings at Twin Galaxies in the 1980s. His 881,360 score on Ms. Pac-Man, reported in the newspaper in January 1985, was simply not recognized, while a later, lower score from Chris Ayra was the official TG world record until 1998. [S53] And yet, TG knew of Bill Bastable, as his name appears on a list of nominees for TG’s 1984 “Player of the Year” title:
http://www.classicarcadegaming.com/contests/Jan1985/extras/
We’ll leave it to you to decide who was advising Walter Day as to which Pac-Man and Ms. Pac-Man scores were worth acknowledging and which were not.
Chris Ayra did eventually make contact with Bastable, visiting him in late October 1986, leading to the Sarasota Herald-Tribune article linked above featuring a photo with the two of them together. Bastable tells the story in an interview done for Dwayne Richard’s documentary, The Perfect Fraudman, starting at about 50:50:
Well, what I thought it was originally was that, as long as I got through the left half of maze 256 and ate up all of what was obvious, that 3,332,820 I believe it is, used to be what I thought or what I perceived as a perfect score [on] original Pac-Man. Then… well, as I was playing back then, there was… I had met Chris Ayra and Bill Mitchell. It started to dawn on me that there was something else going on about the right half of maze 256, where the rainbow-colored letters and numbers are, where you have sixteen columns where you can move the muncher, and thirty-two rows other than these little walls. If somebody had a terrain map, they’d know where these are in particular within that particular arena.
Specifically, Ayra showed Bastable a photograph of the split screen, with a score higher than 3,332,820. While Ayra reportedly did not identify exactly where these extra points came from, Bastable was able to infer from Ayra’s tease that there were more points to be acquired on the broken half of Pac-Man’s final board. (One must wonder how many people Ayra showed off to while his friend Billy insists those hidden dots were their most closely guarded secret.)
Bill Bastable would go on to acquire a Pac-Man cabinet of his own for home use, allowing him to explore the right half of the split screen (which he often refers to as the “vanity board”), making exactly that sort of terrain map he described. Here, you can see his preserved notes, with one panel showing the navigable terrain, and the other panel showing the nine dots:
Bastable continued his account of his journey to Pac-Man supremacy:
Now, I didn’t know it at the time, that each muncher was worth 90 points and that, if you had five men and a bonus at 10,000 that you could have six times that 90, which is 540 points added into the 3,332,820 to get a grand total of 3,333,360. So once I got the game, I think it was either in ’87 or ’88, I don’t remember exact dates on these things, I set about trying to rack test it up there and learn what it is that I didn’t know about the right half of maze 256 so that I could complete the job or complete the mission of actually getting a perfect game.
There was just one other obstacle in Bastable’s way: The Pac-Man board he had acquired included a popular “jumper” modification. This revision used a solder connection to bridge two points on the circuit board, increasing the game’s difficulty by effectively eliminating five game levels from play. [S54] Since three of these eliminated levels included “blue time”, and since each “blue time” session nets the player 3,000 points, this reduced the maximum possible score attainable by a total of 36,000 points, resulting in a perfect score of only 3,297,360. (Since the fruit-drawing algorithm remains unchanged on the jumper version, the split screen still occurs on board 256.)
Thus, Bill Bastable did what any true Pac-Man champion would do: He smashed out a perfect score on the jumper version, before undoing the revision to go after the original Pac-Man perfect score as well. In answering our questions, Bastable offered his reasoning:
The jumper thing was already soldered, so I figured if I scratched it off, I wouldn’t have a chance to go after both of ’em.
Once he had achieved 3,297,360 on the jumper version, Bastable took a pair of scissors to scratch off the solder connection, restoring the board to its original functionality. This allowed him to pursue the original version’s perfect score of 3,333,360, which he achieved in 1988.
As a matter of fact, Bastable has a whole suite of evidence corroborating this 1988 perfect score. First, a photograph [S55]:
This was printed on photo paper dated 1988, as seen on the back:
Bill Herald, a reporter for the Sarasota Herald-Tribune, recalled his encounters with Bastable, as seen in Perfect Fraudman (at about 2:15:20):
There’s no doubt in my mind that he did all this. I mean, he showed me the pictures of it, I saw him in person a number of times, and I mean, he was just absolutely awesome. I can’t imagine anybody being any better at those video games, especially in the mid-’80s…
Bill Herald even kept a file with several notes, including the 1988 photo, and this message slip acknowledging both the standard and jumper perfect scores:
While this would not be satisfactory evidence for a video game world record today in the era of YouTube and Twitch streams, this was far, far above and beyond anything that was expected for the corroboration of video game achievements in the 1980s. There is simply no doubt whatsoever that Bill Bastable achieved a score of 3,333,360 on original Pac-Man.
Well, with one teeny tiny caveat, which unfortunately would disqualify the score competitively by today’s standards. The problem isn’t whether or not he achieved the score – he absolutely did – but rather how it was achieved. Bastable used another of Pac-Man’s dip switches (switch 8) to pause the game midway through. This is relevant for a couple reasons: First, it violates the common rule against artificially or externally pausing a game in progress, as well as the common rule against manipulating the game’s hardware during play. But it also confers an unfair advantage, in that a forced pause is meaningfully different than using the in-game safe spots to hide Pac-Man while the game continues running. As discussed in “Dot One”, those spots can be both temporary and fallible, whereas using the pause switch allows for a guaranteed freeze in the action for as long as you desire.
In terms of advantage, this break in the action is the difference between being able to get up, stretch, grab a snack, use the bathroom, and collect your thoughts before switching from “blue time” ghost collecting to the monotony of ninth key patterns, versus having to switch modes as the game before you continues running. Such breaks are allowed, but only within the confines of the game. To put it another way, the whole reason to use a pause switch is precisely because it is an added advantage.
Note that this was not “cheating” by Bill Bastable. He’s open about the fact that he paused the game in this fashion, both during his original version perfect score and his jumper revision perfect score. While he did present his score to a news reporter, he did not submit it to any adjudicating authority in violation of any rules explicit or implied. In fact, with Guinness and the Twin Galaxies scoreboard out of the picture, arcade score adjudication was in limbo altogether, with the Amusement Players Association perhaps the only outfit still attempting to track arcade high scores. And even if TG had still been accepting scores, this was done on more favorable settings, granting the player six lives instead of four, to maximize the points from the hidden dots, and thus would not have been recognized by TG anyway.
This wasn’t a score intended to be official, at a time when there was no such official body. Bill Bastable simply set out on his personal time to conquer a challenge on one of his favorite games, to the extent he was satisfied to do so. It’s not the kind of achievement that, if done today in the same fashion, could be recognized on the same list as players who did not take such a pause going into the ninth key stretch. But it was a perfect score of 3,333,360 on an original Pac-Man machine, all the way back in 1988. Whether you consider this the first achievement of a perfect score of Pac-Man, based on the evidence presented, is up to you.
One other important clarification: While the story of Bill Bastable pausing his machine has been told in a few different venues, that pause applies only to those two games in the late ’80s including the hidden dots. His previous six perfect scores from the early ’80s (minus the hidden dots), including the ones with newspaper documentation, were done at arcades, with no pausing involved.
Interestingly, Bastable himself does not believe his 1988 score to be the first ever maximum score on Pac-Man. [S56] He recalled his impression upon meeting Chris Arya in Perfect Fraudman, at about 53:30:
That’s the whole point. I’m convinced that they got one by 1983 or ’84. I don’t know the exact timeline, but I remember when Chris Ayra was here, he kind of showed me jokingly a picture of 3,333… or just the 333,090, or something like that. And he wanted to see how I was going to react to it, you know, to see how I was and stuff like that. I know that he had a perfect game, even though he showed me what he showed me… at the time that I’d seen him in October of ’86.
In their text message conversation from 2017, David Race asked Chris Ayra about this picture he had shown to Bill Bastable, with Chris clarifying that it was one of multiple pictures he had taken during the split screen of his claimed perfect score:
In 2009, while researching for his documentaries, Dwayne recalled tracking down Bastable, reiterating Bastable’s own belief that he’d always been third to a perfect score (referring to Billy Mitchell as “Bill”):
https://forums.arcade-museum.com/threads/perfect-pac-man-game.89107/
Regardless, it should not be overlooked that Bill Bastable has both contemporary reporting and permanent evidence showing his ability to get a perfect score in the ’80s, his knowledge of what exactly the perfect score is (3,332,820 without the hidden dots), and his later awareness of all nine of those hidden dots. He clearly made an impression on people, as they were still talking about “Pac-Man Bill” on Usenet in 1998:
https://groups.google.com/g/rec.games.video.classic/c/5xCdJei2iog/m/yjj3fhbOSvUJ
https://groups.google.com/g/rec.games.video.arcade/c/Lj21U1Tw9f0/m/0SBDW2jhTVkJ
(Again, those were from 1998.) [S57]
Meanwhile, Billy’s and Chris’s secrecy doesn’t leave us a lot to go on other than their word. No photos of these supposed perfect scores from 1983 and 1984 have ever been published. They knew of hidden dots by the time Ayra met Bastable. But did Billy and Chris find all nine? Is it possible someone else told them about all nine dots before the TG book was printed a decade later in 1998? And if they were perfect score players like Bastable, why did their public scores indicate a failure to complete even the regular side of the split screen? At the end of the day, the permanent surviving record does not support Billy’s claims. (As you’ll see, that will become a theme of this series.)
Lastly, Billy enjoys painting himself as the original Pac-Man master, and anyone who came into contact with Twin Galaxies after him as a “Johnny-come-lately”. However, as we see, people were already obliterating the game in 1982, even before Billy traveled to the Life Magazine photo shoot. Bill Bastable in particular was already getting documented perfect scores (without the hidden dots) before Billy even took up the game competitively at all. [S58] Personally, I wouldn’t even care that Billy came along later. It makes no difference to me, but it apparently makes a big difference to him.
THE MEMORY GAME
Before we wrap up for the day, I wish to be clear on something. There will always be misremembered facts and discrepancies between historical accounts by different people, even when they witnessed all the same events and are all trying to tell the truth. The point is not to say “There are inconsistencies, therefore any particular thing is a lie.” The point is to start by identifying what the inconsistencies are, and what can be reasonably established. Then we can start to look at what is significant about these inconsistencies, whether these discrepancies are acknowledged by the relevant parties, and whether any effort has been made to correct subsequent retellings upon the discovery of new evidence.
For instance, did Billy Mitchell deliberately never mention Bill Bastable until the topic could no longer be avoided? Did Mitchell not ask whether Bastable had a photograph to prove his claim because he knew damn well Bastable did indeed have one? Did Mitchell not raise the question of whether Bastable’s 1982 letter reflected a score without the hidden dots (instead relying on some wild conspiracy theory) because the topic of changing definitions of a perfect score was something he wished to avoid? [S59]
In other words, what’s relevant is not the fact that there are inconsistencies. It would be astonishing if everyone’s recollection did fully match up after all these years. What matters is how these things relate to the larger picture.
Aside from that, the most important takeaway from today is that the idea that no one was capable of mastering Pac-Man until 1999 is, to be frank, laughable. Players like Randy Tufts had already provided proof of the completion of what they considered to be perfect games of Pac-Man. [S60] Regarding the inclusion of the hidden dots and the escalation of the maximum score, we’ve seen numerous players from the ’80s who would have achieved exactly that if that were the challenge set before them. And we’ve seen at least one who did do exactly that (albeit while using a disqualifying pause function). Far from being the Mt. Everest of video gaming, Pac-Man was a peculiar peak, with an obscure nook, which many climbers simply didn’t consider it worth their while to fully ascend… at least, not yet.
Do you like what you’re seeing so far? I hope so! Because in “Dot Three”, we’ll be taking this show to Funspot, where we’re going to Pac-Man like it’s 1999.
[…] over the last decade, either. If you’ve time to kill, you can read the series here—parts one, two, three, four, five, and six are currently up—and judge for yourself. Fair warning: Each section […]
I went to Riverview High School in Sarasota from 1980 – 1982. One day during this time period a friend told me that there was a pattern to Pac-Man, that someone in Sarasota knew the pattern, and that the highest score was over 3 million. It wasn’t a person I or anyone in my group of friends knew, it wasn’t someone at our high school. Apparently whoever it was played at a Pac-Man machine not at the local Aladdin’s Castle or Galaxy Lanes, but at some store over on Clark Rd.
I never thought much of it until I was reading on this site about the max possible score. I don’t know if this player in Sarasota achieved a “perfect” score, but there were definitely people in the early 80s hitting over 3 million.
Interesting! Thank you. I passed the tip along to others to take a look, see if maybe there’s any news coverage related to it. It’s crazy though, I’m sure a lot of players were putting up massive scores without ever seeking any serious press or reaching out to official scorekeepers, legends in their schools and arcades whose names are lost to time. But like you say, there was no shortage of people crushing 3-mil scores in ’82.
It’s actually a great and helpful piece of information. I’m glad that you shared this helpful info with us. Please stay us informed like this. Thank you for sharing.
Great page, thanks for sharing, very helpful info
FYI – many of the image links are not loading for me (and I assume others). It’s seems to be the same on every page. The non-working links seem to be linked to domain i1.wp.com ; deleting that part of the URL produces a working link.
E.g. Doesn’t work: https://i1.wp.com/perfectpacman.com/wp-content/uploads/2021/09/5M09-Walter-Day-and-John-Hardie.jpg
Works: http://perfectpacman.com/wp-content/uploads/2021/09/5M09-Walter-Day-and-John-Hardie.jpg?resize=225%2C300&ssl=1
Not familiar with Word Press at all but I assume it’s a simple config issue telling it to use this domain when it should just point to perfectpacman.com.
Amazing job reporting BTW. When journos don’t hold ‘power’ accountable we get a huge messes….. VG history should thank you for the clean up job.
Great in depth article with superb research but I didn’t see the name of Ken French anywhere. In late 1982 he was professionally recorded in ¾ U-matic tape getting to the highest score on any records you documented. The title was “Championship Pacman” and aired on select cable channels. He split the screen and each level is recorded and narrated. It’s the first real time documentation of this feat.
Ken is from Redbank CA and was flown to Pgh. PA to film this movie short for Warner Cable. I know because I was the producer and still have some reduced 1/2’” videotaped copies. It was edited by an Nick Spies a PBS Emmy winning editor.
For reference and verification go to an online pdf of Arcade Express It’s dated May 1983.
arcade_express_v1n20.pdf (digitpress.com) Or just search Arcade Express Championship Pacman.
The short article, Titled “Video Teaches Pac-man Strategy”. It’s on page 4 (four). I still have copies of this earliest accomplishment. So if anyone would like a video copy of the World’s first documented highest score and split screen done in 1982, your welcome to contact me.
Howdy! Thank you for your interest in the series.
Ken French is there.
> Just prior to the publication of that book, on March 24, 1982, Redlands Daily Facts reported a Pac-Man score of 3,270,850 by Ken French. Ken lamented that he could have continued his game, had the machine not malfunctioned
And I found the pdf you were referring to:
http://www.digitpress.com/library/newsletters/arcadeexpress/arcade_express_v1n20.pdf
Ken is also mentioned in the Dot Two “Supplemental” material. (Basically, extra stuff that couldn’t make the main series, but we didn’t want to just throw away, since other researchers may find it useful.)
https://perfectpacman.com/dot-two-supplemental/
The problem with Ken’s scores is, as we know today, the game tops out at 3,332,820 (or 3,333,360 if you eat all the hidden dots on the split screen). That makes his March 1982 score of 3,270,850 believable, but not his May claim of 4,186,400. On one hand, you could say he miscounted the million point rollovers, but you could also say he should have known when his 4m+ game took about as long as his previous 3m+ game. Then in June, his claimed top score became 5,971,440. Even if he somehow missed three million point rollovers (which would make the actual score 2,971,440), that score represents a big regression in game play from his other 3m+ marks. Surely he should have noticed.
I don’t have enough info at this time to judge whether he was stitching scores, as described in today’s piece (something which some players believed was okay, because the split screen wasn’t fully understood and because there were no formal rules at the time), or whether Ken made up the 5.9m score outright. We just know it’s impossible in standard play.
None of this is to say Ken wasn’t a fantastic player. You’d have to be to get 3,270,850, which I’m willing to believe is a genuine score. This is just to say there are unanswered questions about his scores.
You should put your footage on YouTube! People would love to see it. I know I would.
I went over this site and I think you have a lot of good information, saved to favorites (:.
Randy Tufts mentioned hidden dots in the July 1983 issue of Electronic Games saying, “some are not visible and others cannot be eaten.” Billy and Chris claim to have gained the knowledge that December. lol
https://archive.org/details/Electronic_Games_Volume_01_Number_17_1983-07_Reese_Communications_US/page/n19/
Mr. Werner,
You stated that if anyone would like a video copy of Ken French’s tutorial to contact you.
Well, I would definitely like to receive such a copy.
My email is: fastestpacman@gmail.com
Thank you,
David Race