P85 · Packing and covering · Classic · Hard

Squares in an equilateral triangle · n = 4

Pack n unit-side squares into a equilateral triangle. Contact is allowed; interiors cannot overlap. Minimise the container triangle side.

Instancen = 4
ObjectiveMinimize container triangle side
Best known, unproven4.154Source-precision interval: 4.154–4.155 (values inside match)exhibited site certificate 4.154700559Compiled by Erich Friedman; Found by Erich Friedman in 1997. Source expression: s = 3 + 2 / √3 = 4.154+.

Formal definition

  • ContainerThe editor and certificate normalize to the equilateral triangle with vertices (0,0), (1,0), (1/2,√3/2). This is a coordinate convention only; scores use the literature's unit-piece scale. Boundary contact is allowed.
  • SubmissionSubmit {radius, placements:[{x,y,turn},…]}, with decimal strings of at most nine places. x,y are centres; circles require turn=0, polygons use turn=tan(θ/2). Coordinates and turn lie in ±4; 0 < radius ≤ 4.
  • ConstraintsExactly n congruent shapes lie wholly in the container with pairwise disjoint interiors. Polygon rotations are independent, not limited to quarter turns.
  • PrecisionInput decimals are exact rationals; regular-polygon vertices are algebraic, not rounded templates. Feasibility has no floating-point tolerance. A finite-decimal certificate is not a continuous optimality proof.
  • Objectivecontainer triangle side = 1/(2 sin(pi/4) r), smaller is better. Pages, leaderboards and literature share these units; conversion is automatic.
Open the full editor
VERIFIED CONSTRUCTIONVERIFIED CONSTRUCTION

Getting a feel for it

Source and certificate

External targets are converted with their printed precision; decimal coordinate certificates are separate. Reconstruction loss never lowers the literature target, and a source construction is not automatically an optimality proof.

Source
The record-holding arrangement for Squares in an equilateral triangle n = 4, 4.154700559
Current leader

4.154700559

container triangle side

Matches the best known
Answer sourceErich Friedman
Solution methodPublished reference construction
Challenge this record Submit a proof / idea Share a proof or idea in the discussion. Accepted contributions can earn proof points.
Record holder's solver noteNo solver note yet (expand)

The record holder has not shared a solver note yet.

ANSWER FORMAT

How to write your answer

the equilateral triangle with vertices (0,0), (1,0), (1/2,√3/2)

Submit {radius, placements:[{x,y,turn},…]}, with decimal strings of at most nine places. x,y are centres; circles require turn=0, polygons use turn=tan(θ/2). Coordinates and turn lie in ±4; 0 < radius ≤ 4.

The current leader's answer

{
  "placements": [
    {
      "turn": "-0.414213562",
      "x": "0.259308766",
      "y": "0.120345617"
    },
    {
      "turn": "-0.414213562",
      "x": "0.500000000",
      "y": "0.120345617"
    },
    {
      "turn": "-0.414213562",
      "x": "0.740691234",
      "y": "0.120345617"
    },
    {
      "turn": "-0.414213562",
      "x": "0.500000000",
      "y": "0.361036851"
    }
  ],
  "radius": "0.170194403"
}
Submission format and technical detailsOpen this when you are ready to prepare a JSON answer

Instance parameters

{
  "n": 4,
  "sides": 5
}

The current leader's answer

{
  "placements": [
    {
      "turn": "-0.414213562",
      "x": "0.259308766",
      "y": "0.120345617"
    },
    {
      "turn": "-0.414213562",
      "x": "0.500000000",
      "y": "0.120345617"
    },
    {
      "turn": "-0.414213562",
      "x": "0.740691234",
      "y": "0.120345617"
    },
    {
      "turn": "-0.414213562",
      "x": "0.500000000",
      "y": "0.361036851"
    }
  ],
  "radius": "0.170194403"
}

Submit {radius, placements:[{x,y,turn},…]}, with decimal strings of at most nine places. x,y are centres; circles require turn=0, polygons use turn=tan(θ/2). Coordinates and turn lie in ±4; 0 < radius ≤ 4. · Verifier v1.0.0

DISCUSSION

Discussion

Talk strategy, share methods, ask why you are stuck. Every signed-in user can post. Posts carry your public byline, the same name your records use; the #number after it is the account's signup ordinal, so a name cannot be worn by someone else. New posts appear after an automated review.

No threads yet. Be the first to talk about this problem?