What's new

Two months of changes in one place. The algorithm databases grew the most: 2x2, big-cube and last-slot sets went public, every table gained a Consensus score, and every set got a printable cheat sheet.

The timer was redesigned with record flags, smart cubes read your turns far more reliably, and the whole site loads faster.

The 0.9 series was about the algorithm databases. OLL, PLL, and F2L now share one experience: 3D playback, sortable stats, and popularity taken from real solves instead of a hand-picked list.

Around them: guides in six languages, a Full F2L trainer mode, an activity feed that groups your solves into sessions, and an app you can install.

Ten days of things going public: the intuitive F2L guide, the glossary, the OLL and PLL databases, the 2-Look guide, and the solve analyzer. The beginner guide was rewritten, with a printable cheat sheet.

Accounts gained usernames and a settings menu, algorithms started grouping into named triggers, and Spanish became the first translation.

The algorithm side grew up: the F2L database opened to everyone, algorithms became rateable one by one, and a notation guide and notation settings arrived. Paste a sequence and Algorithm Lookup names the case.

Also: a public page of reconstructed pro solves, smart cubes that recognize their own brand, and solve import and export.

Seven weeks that turned one trainer into many: Corner Deduction, OLL and PLL recognition with weighted practice, Cross and X-cross, and the first smart-cube trainers. The timer picked up 2x2 through 7x7, ao100, and a solve calendar.

0.7 adds smart-cube solve reconstruction and F2L Lookahead, and drops the account requirement: practice saves on your device.

A small series around the timer and the last layer: a solve chart under the timer with ao5/ao12 trends and date filters, and a PLL recognition guide covering all 21 cases.

0.4 renames Algorithms to Guides and turns PLL recognition into a session you can drill.

F2L arrived, ratings gained a history you can scroll, and search (⌘K), public profiles, and this change log turned up.

The end of the series made guests first-class: solves and ratings save to the cloud before you sign up, then carry over when you do.

A short series: the whole app moved onto one design system, with per-case personal bests and a pass over the mobile timer and drill pages.

0.2 opens with F2L - all 41 cases, four slots each.

The old OLL trainer became an app: an account that syncs your ratings across devices, a PLL trainer beside the OLL one, a solves page, and case probabilities.

0.1 builds the shell around it - a rebuilt home page, dark mode, and guest solves that transfer when you sign up.