At a gaming convention in 2012, a board game designer challenged Eric Harshbarger to create dice that could determine a starting player without the risk of a draw.
The main objective was to swiftly establish the first player and kick off the game promptly.
“The aim was to save players time so they could get into their board game as swiftly as possible,” Harshbarger told As It Happens host Nil Köksal.
Harshbarger, a mathematician and senior lecturer at Auburn University in Alabama, found the challenge intriguing.
“It was a problem that didn’t have an obvious solution, which is something mathematicians are drawn to,” he explained.
Initially, Harshbarger conceived a dice set where each player had an equal chance of winning without the possibility of ties.
However, creating functional dice in reality proved to be a complex task.
Now, after almost 15 years, Harshbarger and his team have developed a set of dice called the Go First Dice for five players, ensuring a fair determination of the starting player with a guaranteed winner from a single roll.
A Persistent Challenge
Over time, Harshbarger took charge of the problem, maintaining a website to document progress as approximately 20 computer scientists and mathematicians worldwide contributed to the project.
They swiftly devised a set of four 12-sided dice that allowed two, three, or four players to roll and achieve a statistically impartial result.
“Solving it for two players was quick, three players took a day or two, and four players took two to three weeks, maybe a month,” said Harshbarger.
However, determining a solution for five players posed a greater challenge.
Harshbarger was confident in the mathematical possibility of the dice but struggled with translating the solution into a physically manufacturable die.
“I knew a solution with 1,440 sides was possible, but not practical,” he noted.
After exploring various theories and reducing the side count, a researcher from Australia proposed a more feasible solution: a set of five 120-sided dice.
The breakthrough attracted attention, leading to a Canadian researcher, Paul Meyer, joining the project within six months.
Canadian Breakthrough
Paul Meyer, a Canadian software engineer, contacted Harshbarger for his previous findings. Three months later, Meyer revealed his discovery using five 60-sided dice.
While many claimed to have solved the problem during the project, Meyer’s solution was verified and successful.
Excited, Harshbarger messaged Meyer, acknowledging the long search for this solution.