Free practice lab

Klondike Clock

Timed scored Klondike solitaire · no account required

Score0
5:00
Pass○○○

About this game

Timed scored Klondike solitaire

You get 300 seconds and a standard 52-card deck. Seven tableau columns are dealt at the start: the first column gets one card face up, the second gets one face down and one face up, continuing until the seventh column has six face down and one face up. The remaining 24 cards sit face down in the stock.

The clock does not start until your first move. Once it does, you build four foundation piles — one per suit, Ace through King — while rearranging the tableau. Tableau stacking requires opposite colors and descending rank. Empty tableau columns accept only Kings. You draw three cards at a time from the stock; each card scores 20 points the first time it appears. Moving any card to a foundation scores 85 points. Revealing a face-down tableau card also scores 20.

This trainer allows three free stock recycles. After that, each recycle costs 20 points. Because re-draws of already-seen cards score nothing, heavy cycling past the free passes quickly becomes a losing trade. Knowing when to stop drawing and work the tableau you have is one of the sharper edges in competitive play.

The time bonus only pays out if you finish the deal — either by solving all 52 cards or by hitting the Finish button. Running the clock to zero earns nothing from the time bonus regardless of how many foundation cards you have. The trainer's current formula is: seconds remaining × foundation card count × 0.2. With 60 seconds and 20 cards on the foundations, that is 240 additional points. On deals that are clearly unwinnable, finishing early rather than waiting for the clock is the correct play.

Klondike Clock runs the same scoring engine we used to build the strategy guide. Its seeded, repeatable deals let you replay the same layout, change one decision, and see whether the result actually improves.

Still being measured

The 0.2 time-bonus coefficient and paid-recycle rule come from the reference model, but recorded scores do not settle either constant cleanly. Both are named engine options so corrected values can replace them without rewriting the rules. Warning audio is also unverified.