Types of Solitaire Games: More Than One Kind of Solitaire

When most people hear Solitaire, they picture Klondike: seven tableau columns, four foundations, a stock and waste, build down in alternating colours, and get every card onto the foundations.

That layout is the classic entry point. It is also only one ruleset in a larger family of patience games that share cards and solo play, then diverge on the engineering that matters: how cards may move, what must be true before a stack can travel, how much of the deal is visible, and what counts as progress toward a win.

Some variants are foundation builders. Others clear same-suit runs off the board. A few abandon tableau stacking for adjacency or arithmetic: Tri Peaks removes cards one rank above or below the waste; Pyramid clears pairs that sum to 13.

Same deck family. Different constraint systems.

The How to Play Solitaire guide documents the Klondike ruleset in full. The sections below map how the major variants differ as game structures, not as mood recommendations. For choosing by preference, the companion piece Which Solitaire Game Should You Play covers that separately.

Klondike Solitaire: Alternating Builds Under Partial Information

Klondike is the version most people mean when they say Solitaire.

One 52-card deck. Seven tableau columns dealt with a triangular face-down underlay (one through seven cards), only the top of each column face up. Tableau builds descend by rank and must alternate colour. Foundations climb Ace to King by suit. Empty columns accept a King only, or a legal run headed by a King. Properly ordered alternating sequences may move as a block; anything out of sequence stays locked in place until rebuilt.

That King-gated empty column and the alternating-colour build rule are the core movement constraints. They limit where cards can park and which stacks can relocate, which is why uncovering face-down cards often matters more than playing every legal move on sight.

The stock adds a second information channel. Draw 1 exposes the stock one card at a time; Draw 3 turns three and only the top waste card is playable, so intermediate cards stay buried until the waste clears. Same deck, same tableau grammar, different access topology. Draw 1 vs Draw 3 Solitaire spells out that stock-rule split. Play Klondike Solitaire uses the standard Solitaire Arcade Klondike layout.

FreeCell: Perfect Information With Explicit Buffer Capacity

FreeCell keeps tableau columns, alternating-colour descending builds, and Ace-to-King suit foundations, then removes the stock and the face-down tableau. All 52 cards deal face up into eight columns (four of seven, four of six). Four free cells each hold one card. Empty columns accept any exposed card, not only Kings.

The free cells are not flavour. They are finite single-card buffers. Cascade length in digital FreeCell is capped by how many free cells and empty columns are open: a longer in-sequence stack can move only when the empty-space budget can justify the same single-card path. Fill every free cell and the legal move graph collapses fast.

There is no hidden underlay to flip and no stock to reshuffle timing around. Complexity sits in space accounting: which buffer to spend, which empty column to keep open, which foundation climb frees a cascade without trapping a needed helper. FreeCell vs Solitaire compares that open-information ruleset with Klondike’s partial-information one.

Spider Solitaire: Two Decks, Suit-Gated Run Clearance

Spider scales the board before it changes the win condition. Two decks (104 cards), ten tableau columns, fifty cards held in a stock that deals one card onto every column at once. There are no Ace-up foundations during play. Progress is completing a descending King-to-Ace run in a single suit; each finished same-suit run leaves the tableau. Eight clears win the deal.

Build rules split movement from clearance. A card may sit on the next higher rank regardless of suit, so mixed descending piles are legal scaffolding. Only an unbroken same-suit sequence moves as a group or clears. Empty columns are unrestricted parking for any card or legal run, but you cannot deal from the stock while any column is empty: the deal rule forces the board to be full-width before ten new cards land.

Suit count is the combinatorial dial. In 1 Suit, every descending stack is already same-suit. In 2 Suit and especially 4 Suit, mixed scaffolding is cheap to build and expensive to untangle, because only clean same-suit runs relocate and clear. Spider 1 Suit vs 2 Suit vs 4 Suit breaks down how far that dial moves the puzzle.

Tri Peaks: Dependency Peaks and Rank-Adjacent Clearance

Tri Peaks does not use Klondike-style tableau stacking or foundations.

Cards deal into three overlapping peaks. A card is playable only when nothing covers it. Removal requires adjacency of rank to the current waste card: one higher or one lower. Suit is irrelevant. A 7 takes a 6 or an 8; a Queen takes a Jack or a King. Clearing a card exposes whatever it covered, so the layout is a dependency graph more than a column builder.

There is no foundation sink. Cards leave the board into the waste chain. When the chain breaks, the stock supplies a new waste top and the adjacency search restarts. Wrap rules change the endpoints: some modes let Ace meet King; others treat Ace and King as hard edges that only meet 2 and Queen. That single connectivity rule reshapes which stock cards can bridge a dead end.

The engineering focus is sequencing under exposure constraints, not assembling alternating stacks. Tri Peaks Strategy: How to Build Longer Runs digs into preserving those adjacency chains.

Pyramid Solitaire: Sum-to-13 Matching Under Cover Constraints

Pyramid replaces sequence building with arithmetic pairing.

Twenty-eight cards form a triangle. Foundations do not exist. Suit and alternating colour do not matter. An exposed card pairs with another exposed card (or a stock/waste partner, depending on the exact house rules) when the ranks sum to 13: 6 with 7, 5 with 8, Queen with Ace. Kings remove alone as the self-pair of 13.

The overlapping deal is the real constraint layer. Cards higher in the pyramid stay illegal until both covering cards below them are gone. Visibility is not the same as access: you can see a useful 6 while it is still covered, and a matching 7 from the stock may arrive before that 6 is free. The matching rule is trivial; the cover graph decides whether pairs can actually fire.

Stock policy (how many passes, whether unused stock returns) changes how many partner opportunities exist before the deal locks. The Pyramid Solitaire strategy guide focuses on those access decisions once the sum-to-13 table is familiar.

Win Conditions Split the Family

The cleanest structural sort is what the rules treat as terminal progress.

Klondike and FreeCell both drain the board into four suit foundations built Ace upward. Spider never feeds Ace-up foundations during play; it ejects completed same-suit King-to-Ace runs. Tri Peaks and Pyramid clear cards off the layout entirely: Tri Peaks by rank adjacency from the waste, Pyramid by legal pairs through the cover graph.

Those sinks rewrite the legal move set. A foundation climb that helps Klondike does not exist in Pyramid. A mixed-suit descending pile that is useful scaffolding in Spider would be illegal as a group move in Klondike. Knowing one win condition does not transfer the others.

Information Models: Hidden Cards Versus Perfect Deals

Information design is a separate axis from the win condition.

Klondike buries most of the tableau under face-down cards and keeps the stock ordered but unseen until drawn. Spider also deals face-down underlays in the ten columns. Tri Peaks hides cards under overlapping parents until those parents clear. Pyramid shows every pyramid card but still encodes access through covers, and the stock remains unknown.

FreeCell is the opposite information model: the full deal is face up from the opening layout. Transparency does not remove difficulty; it relocates it into space and cascade planning. Is Every FreeCell Game Winnable? looks at how that perfect-information structure shapes FreeCell’s character.

How the Stock Is Wired Into Each Ruleset

The stock is not one reusable widget. Its contract changes per game.

In Klondike it is a sequential feed into the waste for tableau or foundation play. In Spider a deal writes one new card onto each of the ten columns simultaneously, with the empty-column lock preventing partial-board deals. In Tri Peaks it resets the waste anchor when an adjacency chain dies. In Pyramid it supplies candidate partners for exposed pyramid cards. FreeCell omits the stock entirely; every card is already on the tableau or in a free cell.

That wiring changes the math of timing. An early Spider deal can bury a near-complete same-suit run under ten new tops; When to Deal From the Stock in Spider Solitaire treats that as a structural timing problem, not a mood preference.

Deck Count and Board Scale

Klondike, FreeCell, Tri Peaks and Pyramid are single-deck (52-card) systems. Spider is a two-deck (104-card) system with ten columns and eight clearance targets instead of four foundations.

More copies of each rank and suit increase the number of legal temporary builds and the number of same-suit assemblies required to finish. Scale here is not “harder by default.” It is a larger state space with a different clearance geometry.

Movement Legality: What May Travel Together

Tableau variants diverge sharply on stack mobility.

Klondike allows a multi-card move only when the lifted cards already form a legal alternating descending sequence, and empty columns stay King-gated. FreeCell’s multi-card moves are space-budgeted: free cells and empty columns determine how long a cascade can travel in one action. Spider lets mixed ranks sit together as single-card builds, but group travel and clearance require same-suit continuity.

Those movement predicates are why two boards that both “build down” play as different machines. The destination test (colour, suit, or rank-only) and the group-move test (ordered sequence, free-cell budget, or same-suit run) define the local physics of each game.

Parameter Twists Inside a Named Game

A game name still under-specifies the ruleset.

Klondike’s Draw 1 versus Draw 3 changes waste access without touching tableau grammar. Spider’s 1 Suit, 2 Suit and 4 Suit modes keep the same movement verbs while changing how often same-suit continuity is available. Pyramid house rules alter stock passes and exact clear definitions. Tri Peaks wrap versus no-wrap changes Ace/King connectivity on the adjacency chain.

Small parameter shifts can dominate strategy. Draw 3 Klondike is restrictive because intermediate waste cards stay unreachable until the top clears, which is why Best Draw 3 Solitaire Strategies treats that mode as its own access problem.

Adjacent Variants: Golf, Yukon, Forty Thieves

The family continues past the five games above, and the interesting differences are still structural.

Golf Solitaire is another rank-adjacency clearer: typically a 7×5 face-up tableau, play exposed cards one rank from the waste, suits ignored. Unlike Tri Peaks, the board is an open grid rather than overlapping peaks, empty columns grant no special parking power, and stock is usually a single pass with no redeal. Ace/King wrap modes change endpoint connectivity the same way Tri Peaks wrap rules do.

Yukon looks like a Klondike relative on the surface (tableau builds, foundations), but it relaxes the group-move predicate: you may lift a face-up card together with everything beneath it even when that lower group is not in sequence. That unsequenced group movement is the engineering point. It lets you relocate buried face-up cards as cargo, which Klondike forbids until the stack is rebuilt into a legal alternating run. There is typically no stock; the whole deal starts in the tableau.

Forty Thieves is a two-deck builder with a much tighter lane discipline. Tableau builds usually descend by suit (not alternating colour), often one card at a time, so open columns and carefully kept same-suit lanes matter more than colour scaffolding. Traditional rules give a single pass through the stock. The strict open-lane requirement is the constraint that defines the game: once a column is cluttered with the wrong suit, recovering a clean build path is expensive because mixed colour helpers are not legal the way they are in Klondike.

Around those sit further Klondike, Spider, FreeCell and patience variants. Most are tiny rule edits (stock passes, empty-column gates, suit filters) that still produce a new legal-move graph.

The Same Cards, Different Physics

Klondike, FreeCell, Tri Peaks and Pyramid can all begin from a shuffled 52-card deck. Spider adds a second deck. What follows is dictated by predicates: which cards may touch, which stacks may travel, what information is face up, and which objects count as cleared.

Klondike couples alternating builds with partial information and a sequential stock. FreeCell couples the same colour grammar with perfect information and finite free-cell buffers. Spider couples rank-only single builds with suit-gated group moves and two-deck clearance. Tri Peaks couples peak dependencies with ±1 waste adjacency. Pyramid couples a cover triangle with sum-to-13 removal.

That is the encyclopedia split: not which mood a game suits, but which constraints the rules enforce. Browse the live rulesets from Klondike Solitaire, FreeCell, Spider Solitaire, Tri Peaks or Pyramid Solitaire to see those physics in motion.

Types of Solitaire Games: More Than One Kind of Solitaire

When most people hear Solitaire, they picture Klondike: seven tableau columns, four foundations, a stock and waste, build down in alternating colours, and get every card onto the foundations.

That layout is the classic entry point. It is also only one ruleset in a larger family of patience games that share cards and solo play, then diverge on the engineering that matters: how cards may move, what must be true before a stack can travel, how much of the deal is visible, and what counts as progress toward a win.

Some variants are foundation builders. Others clear same-suit runs off the board. A few abandon tableau stacking for adjacency or arithmetic: Tri Peaks removes cards one rank above or below the waste; Pyramid clears pairs that sum to 13.

Same deck family. Different constraint systems.

The How to Play Solitaire guide documents the Klondike ruleset in full. The sections below map how the major variants differ as game structures, not as mood recommendations. For choosing by preference, the companion piece Which Solitaire Game Should You Play covers that separately.

Klondike Solitaire: Alternating Builds Under Partial Information

Klondike is the version most people mean when they say Solitaire.

One 52-card deck. Seven tableau columns dealt with a triangular face-down underlay (one through seven cards), only the top of each column face up. Tableau builds descend by rank and must alternate colour. Foundations climb Ace to King by suit. Empty columns accept a King only, or a legal run headed by a King. Properly ordered alternating sequences may move as a block; anything out of sequence stays locked in place until rebuilt.

That King-gated empty column and the alternating-colour build rule are the core movement constraints. They limit where cards can park and which stacks can relocate, which is why uncovering face-down cards often matters more than playing every legal move on sight.

The stock adds a second information channel. Draw 1 exposes the stock one card at a time; Draw 3 turns three and only the top waste card is playable, so intermediate cards stay buried until the waste clears. Same deck, same tableau grammar, different access topology. Draw 1 vs Draw 3 Solitaire spells out that stock-rule split. Play Klondike Solitaire uses the standard Solitaire Arcade Klondike layout.

FreeCell: Perfect Information With Explicit Buffer Capacity

FreeCell keeps tableau columns, alternating-colour descending builds, and Ace-to-King suit foundations, then removes the stock and the face-down tableau. All 52 cards deal face up into eight columns (four of seven, four of six). Four free cells each hold one card. Empty columns accept any exposed card, not only Kings.

The free cells are not flavour. They are finite single-card buffers. Cascade length in digital FreeCell is capped by how many free cells and empty columns are open: a longer in-sequence stack can move only when the empty-space budget can justify the same single-card path. Fill every free cell and the legal move graph collapses fast.

There is no hidden underlay to flip and no stock to reshuffle timing around. Complexity sits in space accounting: which buffer to spend, which empty column to keep open, which foundation climb frees a cascade without trapping a needed helper. FreeCell vs Solitaire compares that open-information ruleset with Klondike’s partial-information one.

Spider Solitaire: Two Decks, Suit-Gated Run Clearance

Spider scales the board before it changes the win condition. Two decks (104 cards), ten tableau columns, fifty cards held in a stock that deals one card onto every column at once. There are no Ace-up foundations during play. Progress is completing a descending King-to-Ace run in a single suit; each finished same-suit run leaves the tableau. Eight clears win the deal.

Build rules split movement from clearance. A card may sit on the next higher rank regardless of suit, so mixed descending piles are legal scaffolding. Only an unbroken same-suit sequence moves as a group or clears. Empty columns are unrestricted parking for any card or legal run, but you cannot deal from the stock while any column is empty: the deal rule forces the board to be full-width before ten new cards land.

Suit count is the combinatorial dial. In 1 Suit, every descending stack is already same-suit. In 2 Suit and especially 4 Suit, mixed scaffolding is cheap to build and expensive to untangle, because only clean same-suit runs relocate and clear. Spider 1 Suit vs 2 Suit vs 4 Suit breaks down how far that dial moves the puzzle.

Tri Peaks: Dependency Peaks and Rank-Adjacent Clearance

Tri Peaks does not use Klondike-style tableau stacking or foundations.

Cards deal into three overlapping peaks. A card is playable only when nothing covers it. Removal requires adjacency of rank to the current waste card: one higher or one lower. Suit is irrelevant. A 7 takes a 6 or an 8; a Queen takes a Jack or a King. Clearing a card exposes whatever it covered, so the layout is a dependency graph more than a column builder.

There is no foundation sink. Cards leave the board into the waste chain. When the chain breaks, the stock supplies a new waste top and the adjacency search restarts. Wrap rules change the endpoints: some modes let Ace meet King; others treat Ace and King as hard edges that only meet 2 and Queen. That single connectivity rule reshapes which stock cards can bridge a dead end.

The engineering focus is sequencing under exposure constraints, not assembling alternating stacks. Tri Peaks Strategy: How to Build Longer Runs digs into preserving those adjacency chains.

Pyramid Solitaire: Sum-to-13 Matching Under Cover Constraints

Pyramid replaces sequence building with arithmetic pairing.

Twenty-eight cards form a triangle. Foundations do not exist. Suit and alternating colour do not matter. An exposed card pairs with another exposed card (or a stock/waste partner, depending on the exact house rules) when the ranks sum to 13: 6 with 7, 5 with 8, Queen with Ace. Kings remove alone as the self-pair of 13.

The overlapping deal is the real constraint layer. Cards higher in the pyramid stay illegal until both covering cards below them are gone. Visibility is not the same as access: you can see a useful 6 while it is still covered, and a matching 7 from the stock may arrive before that 6 is free. The matching rule is trivial; the cover graph decides whether pairs can actually fire.

Stock policy (how many passes, whether unused stock returns) changes how many partner opportunities exist before the deal locks. The Pyramid Solitaire strategy guide focuses on those access decisions once the sum-to-13 table is familiar.

Win Conditions Split the Family

The cleanest structural sort is what the rules treat as terminal progress.

Klondike and FreeCell both drain the board into four suit foundations built Ace upward. Spider never feeds Ace-up foundations during play; it ejects completed same-suit King-to-Ace runs. Tri Peaks and Pyramid clear cards off the layout entirely: Tri Peaks by rank adjacency from the waste, Pyramid by legal pairs through the cover graph.

Those sinks rewrite the legal move set. A foundation climb that helps Klondike does not exist in Pyramid. A mixed-suit descending pile that is useful scaffolding in Spider would be illegal as a group move in Klondike. Knowing one win condition does not transfer the others.

Information Models: Hidden Cards Versus Perfect Deals

Information design is a separate axis from the win condition.

Klondike buries most of the tableau under face-down cards and keeps the stock ordered but unseen until drawn. Spider also deals face-down underlays in the ten columns. Tri Peaks hides cards under overlapping parents until those parents clear. Pyramid shows every pyramid card but still encodes access through covers, and the stock remains unknown.

FreeCell is the opposite information model: the full deal is face up from the opening layout. Transparency does not remove difficulty; it relocates it into space and cascade planning. Is Every FreeCell Game Winnable? looks at how that perfect-information structure shapes FreeCell’s character.

How the Stock Is Wired Into Each Ruleset

The stock is not one reusable widget. Its contract changes per game.

In Klondike it is a sequential feed into the waste for tableau or foundation play. In Spider a deal writes one new card onto each of the ten columns simultaneously, with the empty-column lock preventing partial-board deals. In Tri Peaks it resets the waste anchor when an adjacency chain dies. In Pyramid it supplies candidate partners for exposed pyramid cards. FreeCell omits the stock entirely; every card is already on the tableau or in a free cell.

That wiring changes the math of timing. An early Spider deal can bury a near-complete same-suit run under ten new tops; When to Deal From the Stock in Spider Solitaire treats that as a structural timing problem, not a mood preference.

Deck Count and Board Scale

Klondike, FreeCell, Tri Peaks and Pyramid are single-deck (52-card) systems. Spider is a two-deck (104-card) system with ten columns and eight clearance targets instead of four foundations.

More copies of each rank and suit increase the number of legal temporary builds and the number of same-suit assemblies required to finish. Scale here is not “harder by default.” It is a larger state space with a different clearance geometry.

Movement Legality: What May Travel Together

Tableau variants diverge sharply on stack mobility.

Klondike allows a multi-card move only when the lifted cards already form a legal alternating descending sequence, and empty columns stay King-gated. FreeCell’s multi-card moves are space-budgeted: free cells and empty columns determine how long a cascade can travel in one action. Spider lets mixed ranks sit together as single-card builds, but group travel and clearance require same-suit continuity.

Those movement predicates are why two boards that both “build down” play as different machines. The destination test (colour, suit, or rank-only) and the group-move test (ordered sequence, free-cell budget, or same-suit run) define the local physics of each game.

Parameter Twists Inside a Named Game

A game name still under-specifies the ruleset.

Klondike’s Draw 1 versus Draw 3 changes waste access without touching tableau grammar. Spider’s 1 Suit, 2 Suit and 4 Suit modes keep the same movement verbs while changing how often same-suit continuity is available. Pyramid house rules alter stock passes and exact clear definitions. Tri Peaks wrap versus no-wrap changes Ace/King connectivity on the adjacency chain.

Small parameter shifts can dominate strategy. Draw 3 Klondike is restrictive because intermediate waste cards stay unreachable until the top clears, which is why Best Draw 3 Solitaire Strategies treats that mode as its own access problem.

Adjacent Variants: Golf, Yukon, Forty Thieves

The family continues past the five games above, and the interesting differences are still structural.

Golf Solitaire is another rank-adjacency clearer: typically a 7×5 face-up tableau, play exposed cards one rank from the waste, suits ignored. Unlike Tri Peaks, the board is an open grid rather than overlapping peaks, empty columns grant no special parking power, and stock is usually a single pass with no redeal. Ace/King wrap modes change endpoint connectivity the same way Tri Peaks wrap rules do.

Yukon looks like a Klondike relative on the surface (tableau builds, foundations), but it relaxes the group-move predicate: you may lift a face-up card together with everything beneath it even when that lower group is not in sequence. That unsequenced group movement is the engineering point. It lets you relocate buried face-up cards as cargo, which Klondike forbids until the stack is rebuilt into a legal alternating run. There is typically no stock; the whole deal starts in the tableau.

Forty Thieves is a two-deck builder with a much tighter lane discipline. Tableau builds usually descend by suit (not alternating colour), often one card at a time, so open columns and carefully kept same-suit lanes matter more than colour scaffolding. Traditional rules give a single pass through the stock. The strict open-lane requirement is the constraint that defines the game: once a column is cluttered with the wrong suit, recovering a clean build path is expensive because mixed colour helpers are not legal the way they are in Klondike.

Around those sit further Klondike, Spider, FreeCell and patience variants. Most are tiny rule edits (stock passes, empty-column gates, suit filters) that still produce a new legal-move graph.

The Same Cards, Different Physics

Klondike, FreeCell, Tri Peaks and Pyramid can all begin from a shuffled 52-card deck. Spider adds a second deck. What follows is dictated by predicates: which cards may touch, which stacks may travel, what information is face up, and which objects count as cleared.

Klondike couples alternating builds with partial information and a sequential stock. FreeCell couples the same colour grammar with perfect information and finite free-cell buffers. Spider couples rank-only single builds with suit-gated group moves and two-deck clearance. Tri Peaks couples peak dependencies with ±1 waste adjacency. Pyramid couples a cover triangle with sum-to-13 removal.

That is the encyclopedia split: not which mood a game suits, but which constraints the rules enforce. Browse the live rulesets from Klondike Solitaire, FreeCell, Spider Solitaire, Tri Peaks or Pyramid Solitaire to see those physics in motion.