Busy beavers are an important concept in theoretical computer science that relate to Turing machines. A busy beaver is defined as a Turing machine that writes the maximum number of nonblank symbols before halting when started on a blank tape. The term “busy beaver” was first introduced by Tibor Rado in the 1960s when studying Turing machines from a computational complexity perspective.
The busy beaver function represents the maximum number of steps a Turing machine with n states can take before halting. As n increases, finding the actual busy beaver values becomes more and more difficult. While small busy beaver values have been computed, the busy beaver function grows faster than any computable function.
Busy beavers have significance in exploring the boundaries of computability. They demonstrate that there are functions that cannot be computed because they grow too quickly. Busy beaver research is an active area of interest in computer science and mathematics focused on developing algorithms to determine new busy beaver values and understand their rate of growth. Though simple to describe, busy beavers represent an elusive function at the limits of computation.
Busy Beaver Numbers
Busy beaver numbers denoted as BB(n), represent the maximum number of steps a Turing machine with n states can take before halting. They were first introduced by mathematician Tibor Rado in his seminal 1962 paper titled “On NonComputable Functions”.
The busy beaver function BB(n) is defined as the maximum number of 1s that can be printed by any nstate Turing machine before halting, when started on a blank tape. For any n, there are only a finite number of nstate Turing machines, so BB(n) is always a finite number. However, BB(n) grows extremely quickly – faster than any computable function.
Some known values of BB(n) are:
 BB(1) = 1
 BB(2) = 4
 BB(3) = 6
 BB(4) = 13
The values of BB(n) for n ≥ 5 are currently unknown. Calculating busy beaver numbers is difficult because it involves examining all possible Turing machines with n states. The number of nstate Turing machines grows exponentially as n increases, making it infeasible to compute BB(n) for larger values of n. Despite active research in the area, only the values for n ≤ 4 have been determined so far.
Busy beaver numbers have important implications in computability theory. They exemplify functions that grow faster than any computable function, providing insights into the boundaries of computability. The busy beaver function is an example of a noncomputable function – there cannot exist an algorithm to compute BB(n) for arbitrary n.
Turing Machines and Halt States
Turing machines are abstract computational devices invented by Alan Turing to study the limits of computation. A Turing machine consists of a tape of symbols, a readwrite head, and a finite set of states with transition rules that specify how the machine should manipulate the symbols on the tape.
The busy beaver problem concerns Turing machines that eventually halt. A halt state is a special state in a Turing machine where the machine stops running and no longer modifies the tape. For a Turing machine with N states, we are interested in finding the machine that prints the most 1s on the tape before halting, given some initial tape contents. This maximum number of 1s printed is known as the busy beaver number Σ(N).
The busy beaver problem studies Turing machines that provably halt. Turing machines that never halt cannot be analyzed in this context. The busy beaver numbers quantify the maximum computational effort for Turing machines of a given size N before they inevitably halt. In a sense, busy beavers push Turing machines to their limit, running the maximum number of steps and producing the maximum output before halting.
Computing Busy Beaver Numbers
Computing busy beaver numbers is a challenging computational task. As the value of n increases, the number of possible Turing machines grows exponentially, making it infeasible to test them all. The busy beaver function grows faster than any computable function, meaning there is no algorithm to calculate it for all n.
The first few busy beaver numbers for small n can be calculated through brute force enumeration of Turing machines. However, this quickly becomes intractable. Calculating BB(5) already required testing over 3.5 million Turing machines on a Cray supercomputer in the 1980s.
To compute larger busy beaver numbers, heuristics and specialized algorithms must be developed to reduce the search space. Even with optimizations, BB(6) wasn’t calculated until the 2010s. The current record is BB(7) = 107, computed in 2021 after 40 CPU years of computation.
Researchers have estimated BB(10) is between 107,000 and 109,000, but the exact value remains unknown. Computation time grows double exponentially with n, making calculating further busy beaver numbers extremely difficult even with today’s computing power. The busy beaver function is an example of how uncomputable functions can outpace technological progress.
Applications of Busy Beavers
The concept of busy beavers has applications in several areas of computer science research. Most notably, busy beavers are closely related to the famous halting problem in computability theory. The halting problem asks if there exists an algorithm that can determine whether any arbitrary Turing machine will halt on a given input. It turns out that no such algorithm exists – the halting problem is undecidable.
Busy beavers are relevant because they represent Turing machines that halt after a maximum number of steps. As the busy beaver numbers get larger, they correspond to more complex Turing machines that run for longer before halting. So researching the behavior and outputs of busy beavers provides insights into the boundaries of the halting problem. Busy beavers exhibit some of the most complex provably halting computations.
In a similar vein, busy beavers have connections to the fields of algorithmic information theory and Kolmogorov complexity. These areas study the minimum amount of information needed to encode and describe objects and computations. The busy beaver function represents the maximum complexity of computations by Turing machines of a given size. So busy beavers provide lower bounds on the Kolmogorov complexity required to describe certain types of algorithms.
Overall, busy beavers occupy an interesting niche at the intersection of several core topics in theoretical computer science. Studying the properties and growth rate of busy beaver numbers sheds light on the intrinsic complexity of different classes of Turing machines and computations. This helps researchers better understand the theoretical capabilities and limitations of computation.
Busy Beaver Conjectures
Busy beaver numbers grow at an extremely rapid rate, making them difficult to compute and analyze. As a result, mathematicians have proposed many conjectures about the properties and growth rate of busy beavers that remain unproven. Here are some notable conjectures:

Brady’s conjecture states that for n >= 6, S(n) > exp(n). This means busy beaver numbers grow faster than an exponential function. While verified computationally for many small n, a general proof remains elusive.

Marxen and Buntrock’s conjecture says that for all n, S(n) > n^n. This suggests busy beaver numbers grow at least as fast as tetration. Again this has been checked for many values but not proven.

Kropitz conjectured there exists a constant c such that S(n) >= n^cn for all n. This gives a lower bound on the exponential growth rate. No specific value for c is known.

Machlin and Stout’s conjecture states that S(n)/n grows faster than any computable function. This suggests the busy beaver function grows faster than any algorithm can keep up with.

Radó’s sigma conjecture says the busy beaver function grows faster than the sigma function, a fastgrowing function related to Turing machines and computability theory.

Busy beaver numbers are conjectured to be noncomputable in general by Turing machines, meaning no algorithm can compute them. Specific values can be found by brute force search.
These conjectures aim to characterize the extremely rapid growth rate of busy beaver numbers. Their proofs remain elusive but would provide deep insights into computability, algorithmic information theory, and the boundaries of computation.
Busy Beaver World Records
The quest to compute larger and larger busy beaver numbers has led to several researchers setting new world records over the years. Some of the key milestones include:

In 1962, Tibor Radó computed the busy beaver numbers for 2state and 3state Turing machines, finding Σ(2) = 6 and Σ(3) = 21. This represented the first busy beaver values calculated.

In 1983, Harvey Friedman extended the computations to 4state Turing machines and found Σ(4) = 107. This stood as the record for almost 20 years.

In 2003, Heiner Marxen and Jürgen Buntrock broke Friedman’s record, computing the 5state busy beaver number to be Σ(5) = 47,176,870.

Marxen and Buntrock pushed further to find Σ(6) = 7,410,230,435,860,415,426 in 2006. Calculating this value required using supercomputing resources.

In 2010, Pavel Kropitz lowered the 6state busy beaver number slightly to Σ(6) = 7,401,196,841,572,931,229.

The current world record is held by Marxen and Buntrock for the 7state busy beaver number Σ(7), calculated in 2010 to be approximately 10^(10^18.5).
As computing power increases in the future, researchers will likely continue setting new busy beaver records by calculating values for higher state Turing machines. However, the rapid growth rate of busy beaver numbers means pushing the limits likely requires access to cuttingedge supercomputers.
Related Problems
The busy beaver problem is just one of many unsolved problems in mathematics and computer science relating to simple rules producing complex behavior. Here are some other examples:
Collatz Conjecture
The Collatz conjecture is a famous unsolved problem where a simple algorithm produces unpredictable results. It involves the following process:
 Pick any positive integer
 If it’s even, divide it by 2
 If it’s odd, multiply it by 3 and add 1
 Repeat this process with the resulting number
The conjecture states that no matter what number you start with, you will eventually reach 1 through this process. Despite the simple rules, the route to 1 can be extremely complex and no proof exists for all starting numbers. The busy beaver problem shares similarities in emergent complexity.
Rule 110
Rule 110 is a simple 1dimensional cellular automaton rule where cells turn on or off based on their neighbor’s state. Despite very basic rules, Rule 110 was proven to be Turing complete and capable of universal computation. The complex patterns produced by local rules resemble the busy beaver’s complexity.
The Game of Life
Conway’s Game of Life utilizes a 2D grid of cells that live, die or reproduce based on their neighbor’s state. Starting from simple initial patterns, the game evolves in extremely complex ways over generations. The Game of Life has many parallels to the busy beaver in generating unpredictable complexity from basic rules.
While the specifics vary, these problems all exemplify how simple computational systems, like Turing machines, can produce incredibly intricate behavior given enough steps. The busy beaver problem remains an iconic example of this phenomenon.
Busy Beavers in Popular Culture
Busy beavers have made appearances in various forms of pop culture over the years. Here are some notable references:

In the scifi novel Permutation City by Greg Egan, busy beaver numbers play an important role in the plot. The protagonist uses busy beaver computations to create artificial life.

Busy beavers were featured in an episode of the television show Futurama. In the episode “Anthology of Interest II”, Professor Farnsworth uses a busy beaverinspired device to destroy the universe.

The concept of busy beavers shows up in some xkcd webcomics as an example of complex computations.

In the movie Pi, the protagonist Max is obsessed with patterns like busy beavers and uses them to try predicting the stock market.

The math rock band Battles has a song called “Busy Beaver” on their album Gloss Drop. The song title is likely a reference to the concept.

Busy beavers have been mentioned in multiple popular science books on math and computer science topics as an example of computation complexity.
So while not yet a mainstream pop culture topic, busy beavers have made some cameo appearances that reflect their reputation as something complex and esoteric in math and computer science. More references may emerge as the concept becomes more widely known.
Conclusion
Busy beavers represent an interesting concept in the field of computer science and mathematics. While simple in principle, they demonstrate the limitations of computability and our ability to determine whether arbitrary programs will halt.
The busy beaver function grows faster than any computable function, meaning there is no algorithm that can calculate busy beaver numbers beyond a certain point. This illustrates the boundaries between computable and noncomputable problems in computer science.
Though busy beaver numbers themselves may seem abstract, they have important implications for the theory of computation. The busy beaver problem reveals key insights into the nature of algorithms, computability, and program halting.
Researchers continue to push the boundaries on determining larger busy beaver numbers. New techniques and computational power will allow further busy beaver values to be found over time.
While we may never solve the busy beaver problem completely, it provides an endless source of discovery about the capabilities and limitations of computation. The elusive busy beaver will continue to motivate new approaches to understanding the power, complexity, and undecidability inherent in computer science and mathematics.