QED Logic quod erat demonstrandum

Logic puzzle techniques

Every Logic Ascent board is solved by applying these named techniques in order, cheapest first — never a guess, never brute force. The first three happen automatically on every marker you place; the rest are the real judgment calls, each shown below with a board that actually needs it.

Automatic, on every marker

  • Single cell. A row, column, or region with no marker yet and exactly one open cell left has no choice: that cell is forced.
  • Row/column exclusion. Placing a marker rules out the rest of its row, column, and region automatically — free, and available on every single mark, so it's never something to search for. Auto mode does this for you.
  • Adjacency clear. Placing a marker also rules out its touching neighbours, including diagonally, since no two markers may touch. Also free and automatic, on every mark, alongside row/column/region exclusion above.

Region confinement

A region's remaining open cells all sit in one row or column: that region's own marker has to land in that row or column, so every other open cell there — outside the region — can be ruled out.

A
B
C
D
E
F
G
H
1
2
3
4
5
6
7
8
1
2
3
4🐱
🐱
🐱
🐱
5
6
🐱
7
8
🐱
🐱
🐱

Every open cell left in this region sits in row 3, so its marker has to land there; that clears every other open cell in the row.

Try a board that needs this →

Line confinement

A row or column's remaining open cells all sit inside one region: that row or column's marker has to land in that region, so every other open cell in the region — outside that row or column — can be ruled out.

A
B
C
D
E
F
G
H
1
2
3
4
5
6
7
8
1🐱
2
3
4
5
🐱
🐱
🐱
6
🐱
7
8
🐱
🐱
🐱

Every open cell in row 8 belongs to the same region, so that region's marker has to be somewhere in this row; that clears every other open cell in the region.

Try a board that needs this →

Paired regions

Two regions, between them, have open cells only across the same two rows (or the same two columns): with one marker each, those two markers have to fill exactly those two rows or columns, so every other region's open cells there can be ruled out.

A
B
C
D
E
F
G
H
1
2
3
4
5
6
7
8
1
2
🐱
3
🐱
4
5
🐱
🐱
🐱
6
7
🐱
8🐱
🐱

Together, two regions have open cells only in columns C and D; with one marker each, those two markers have to fill exactly those two columns, so every other region's open cells there can be cleared.

Try a board that needs this →

Shape squeeze

A region's remaining candidates are shaped so that some cell outside the region touches every one of them: whichever candidate ends up marked will rule that cell out, so it can be ruled out now, before knowing which candidate wins.

A
B
C
D
E
F
G
H
1
2
3
4
5
6
7
8
1
2🐱
3
4
🐱
5
🐱
6
🐱
🐱
7
🐱
8
🐱
🐱

Every remaining candidate in this region touches C2 and B4 and C4, so whichever candidate ends up marked will clear them, and they can be ruled out now regardless of which one wins.

Try a board that needs this →

Forced chain

No simpler technique moves the board forward. Trying one specific cell as the marker and following its direct consequences leads to a contradiction — some row, column, or region left with no valid candidate — so that cell can never be the marker, and is ruled out for real.

A
B
C
D
E
F
G
H
1
2
3
4
5
6
7
8
1
2🐱
3
🐱
4
🐱
🐱
5
6
🐱
7
🐱
🐱
8🐱

Marking E2 would force C3 (the last open cell in column C), and then row 4 would be left with no open cell for its marker. So E2 can't be the marker there.

Try a board that needs this →