Does Eternity II have a solution?

Almost certainly yes. Statistical estimates of the piece set imply many perfect solutions exist; the difficulty is finding one in a search space of about 1.1 × 10^557 arrangements, not whether one exists.

Is Eternity II a scam? Can it even be solved?

The question comes up on Hacker News, on Discord, and it came up on the players' own mailing list back in 2007. A $2,000,000 prize, a deadline that passed with no winner, and a rule that you may not tell anyone you've cracked it: it does look suspicious.

Short answer

No, it is not a scam, and yes, a complete solution almost certainly exists. The competition ran from 28 July 2007 to noon on 31 December 2010 and expired with no winner, but that is a story about difficulty, not deception. The puzzle was built to have very few solutions, so close to one solution by the maker's account, that finding one among roughly 1.15×10⁵⁵⁷ arrangements (with the centre clue pinned; ~10⁵⁶⁰ without it) is astronomically hard. "Few solutions by design" is not the same claim as "no solution exists."

The community asked this first

You are in good company. On 6 August 2007, a player writing as jagbrain put it bluntly on the official list: "How do we know Eternity II isn't a big SCAM? ... Are there a notary in the Committee? If there is, the name must be public." That is the same worry people voice today: who holds the solution, and is anyone impartial checking? Here is what the record actually shows.

Three things people get wrong

“Even the creator has no solution.”

Half true, and it cuts the other way.

A Tomy leaflet quoted on the mailing list says "the ETERNITY II solution was printed between pages of random text whilst all parties were out of the room" (the leaflet was relayed by Dave Clark in a related message). The design deliberately avoided anyone keeping a readable master solution. That is a safeguard against insider cheating, not evidence that no solution exists. The puzzle was engineered around a solution; the makers chose not to hold it in plain sight.

“Only a quantum computer could solve it.”

A misframing.

Quantum hardware is not a magic key for this kind of search, and no quantum machine is needed for the claim "a solution exists" to hold. Eternity I, the 1999 predecessor, was solved within about 18 months by Alex Selby and Oliver Riordan on ordinary computers, using better reasoning rather than more raw speed. Eternity II was then hardened against the exact tricks that cracked Eternity I, which is why it is harder, not why it is impossible.

“It's cruel, or plain impossible.”

Neither.

The best partial board found so far matches 470 of the 480 interior edges (Joshua Blackwood, 2021). Ten unmatched edges out of 480 is a narrow gap, and progress has been real: 467 in 2008, 469 in 2020, 470 in 2021. A search this hard can go a very long time without a full hit while a solution still sits somewhere in the space. Hard is not the same as impossible.

Was there any real verification?

Yes, and it had a precedent that worked. Eternity I was checked by "specialist loss adjusters acting for the insurers," and the prize was paid. For Eternity II the plan was annual "scrutineering" dates when submitted boards would be judged (see the players' notes on the scrutiny schedule and the follow-up). The "don't tell anyone you've solved it" rule (raised on the list by simon.chapple) protected the prize from leaks, and it also fed the suspicion. A real solution had a real path to being verified and paid; nobody ever produced one before the deadline.

Few by design is not impossible

This is the point worth keeping. Selby and Riordan helped design Eternity II precisely so it would not have the abundance of near-solutions that let them beat Eternity I. Fewer target solutions means the search has almost nothing to home in on, so brute force and clever heuristics alike wander an enormous space. A solution exists by construction; what is missing is a method fast enough to find it. That is why the puzzle is famous, and why it is not a trick.

Read further

For the combinatorics behind that 10⁵⁵⁷, the NP-completeness of edge matching, and how solvers actually attack the board, see Algorithms and the Research pages.

For the current best boards, including the 470/480 record, see the records table on The Puzzle.

Common questions

Is Eternity II a scam?: No. It was a genuine contest with a real, escrowed $2,000,000 prize. The prize simply expired unclaimed in 2010 because no one found a complete solution in time — the puzzle is extraordinarily hard, not rigged.
Does Eternity II have a solution?: Almost certainly yes. Statistical estimates of the piece set imply many perfect solutions exist; the difficulty is finding one in a search space of about 1.1 × 10^557 arrangements, not whether one exists.