Perfect Numbers

Take the number 6. Its proper divisors — the numbers that divide it evenly, not counting 6 itself — are 1, 2 and 3. Add them up: 1 + 2 + 3 = 6. The number rebuilds itself exactly from its own parts.

The Greeks found this so striking they called such numbers perfect. After 6 comes 28 (1 + 2 + 4 + 7 + 14), then 496, then 8128. Euclid wrote about them around 300 BC, and they have fascinated mathematicians ever since.

They are also rare and strange. The next perfect number after 8128 is 33,550,336 — there is nothing in between. And tucked inside this innocent-looking definition is a question that has stayed open for more than two thousand years: does an odd perfect number exist?

Pick any number and the tool below adds up its proper divisors. If the sum lands exactly on the number, it is perfect; less and it is deficient, more and it is abundant. Perfect numbers are the knife's edge between the two.

<p class="hint">{{hint}}</p>
<div class="row">
  <input id="num" type="number" min="1" value="28" />
  <button id="test" type="button">{{btn_test}}</button>
</div>
<div class="status" id="status">28 = 1 + 2 + 4 + 7 + 14 = 28 — perfect!</div>
<div class="gen">
  <label for="p">{{label_mersenne}}</label>
  <select id="p">
    <option value="2">2</option>
    <option value="3" selected>3</option>
    <option value="5">5</option>
    <option value="7">7</option>
    <option value="11">11</option>
    <option value="13">13</option>
  </select>
  <button id="make" type="button" class="ghost">{{btn_make}}</button>
</div>
<div class="status2" id="status2"></div>

* { box-sizing: border-box; }
body { font-family: system-ui, sans-serif; color: #222; margin: 0; }
.hint { font-size: .9rem; color: #444; margin: 0 0 .8rem; line-height: 1.45; }
.row, .gen { display: flex; gap: .5rem; align-items: center; flex-wrap: wrap; margin: .5rem 0; }
.gen label { font-size: .9rem; color: #1d3557; font-weight: 600; }
input, select { font: 600 15px system-ui, sans-serif; padding: .4rem .6rem; border: 1px solid #adb1b8;
        border-radius: 8px; }
input { width: 130px; }
.status, .status2 { font-size: 1rem; font-weight: 600; margin: .5rem 0; min-height: 1.4em; line-height: 1.4; }
.status.ok, .status2.ok { color: #0a7d33; }
.status.def { color: #1d6fb8; }
.status.ab { color: #c97a16; }
button { font: 600 14px system-ui, sans-serif; padding: .45rem .9rem; border: 1px solid #1d3557;
         background: #1d3557; color: #fff; border-radius: 8px; cursor: pointer; }
button.ghost { background: #fff; color: #1d3557; }
code { background: #eef2f6; padding: .1rem .35rem; border-radius: 5px; font-size: .92em; }

// Code not found

Then try the second button. Every even perfect number is built from a Mersenne prime — a prime of the form 2ᵖ − 1 — by the formula 2ᵖ⁻¹ × (2ᵖ − 1). This is the Euclid–Euler theorem: Euclid proved this recipe always yields a perfect number, and Euler (≈1750) proved every even perfect number must look exactly this way. Watch how a tiny prime explodes into a perfect number, and notice that no odd one ever appears.

How well do we actually understand perfect numbers? The answer splits sharply in two.

• Even perfect numbers: solved. The Euclid–Euler theorem ties them exactly to Mersenne primes. Euclid showed (≈300 BC) that 2ᵖ⁻¹(2ᵖ − 1) is perfect whenever 2ᵖ − 1 is prime; Leonhard Euler (≈1750) proved the converse — every even perfect number is of this form. So even perfect numbers and Mersenne primes are two names for the same list. As of today only 52 are known, each tied to a known Mersenne prime, the largest with tens of millions of digits.
• Odd perfect numbers: OPEN. Does even one exist? Nobody knows. This is one of the oldest unsolved problems in mathematics — older than P vs NP by millennia.
• What we do know is only a wall of constraints: if an odd perfect number exists it must exceed $10^{3000}$, have many distinct prime factors, and satisfy a long list of conditions. Every search has come up empty, yet no proof rules it out.

So perfect numbers live in two worlds at once: the even side is fully classified, while the odd side is a question mark that has resisted every mind for over two thousand years. Like the twin prime question, it is easy to state and brutally hard to settle.

A 2,000-year-old curiosity might sound purely ornamental, but its machinery reaches into very modern places:

• The hunt for Mersenne primes: because each even perfect number is a Mersenne prime in disguise, the search for new perfect numbers is the search for record-breaking primes. GIMPS (the Great Internet Mersenne Prime Search) has volunteers worldwide donate computing time to find them.
• Primality testing: Mersenne numbers have a special fast test (Lucas–Lehmer), a clean case study in why some numbers are easy to test and others hard — the same tension behind is N prime?.
• Distributed computing: GIMPS was one of the first great volunteer supercomputing projects, a blueprint later copied for protein folding and astronomy.
• Teaching number theory: perfect numbers are a perfect on-ramp — divisors, primes, proof and an honest open problem, all from a definition a child can grasp.

Understand perfect numbers and you have touched primes, divisor functions, and the living edge of mathematics where questions are still unanswered.

Perfect numbers are a small miracle of arithmetic: a handful of integers that reconstruct themselves perfectly from their divisors. On the even side, Euclid and Euler closed the book completely — each one is a Mersenne prime wearing a different hat, and we have found exactly 52.

But the odd side stays wide open. After more than two thousand years of searching, no one has found an odd perfect number, and no one has proven one cannot exist. It is a humbling reminder that mathematics still has ancient, simple-sounding questions it cannot answer — and that perfection, even in numbers, is harder to pin down than it looks. For more on the surprising hardness hidden in plain numbers, see is N prime?.