SmartcubeAnalyzer

SmartcubeAnalyzer is a tool designed to analyze and visualize data related to cube solving. It provides various statistical functions and visualizations to help users understand their performance and improve their solving skills.

Features

• Average Time: Displays the running average time of solves.
• Average Recognition and Execution: Shows the running average split by recognition and execution times.
• Count of Solves by How Long They Took: Visualizes the number of solves within specific time ranges.
• Average Turns Per Second (TPS): Displays the average TPS and TPS during execution.
• Average Turns: Shows the average number of turns per solve.
• Solve Efficiency: Shows move efficiency ratio (after cancelling redundant same-face moves; 100% = no wasted moves), OLL/PLL success rates, and a combined solve efficiency.
• Top 100 Fastest Solves: Lists the top 100 fastest solves given the current filters.
• Average Standard Deviation: Displays the running standard deviation of solve times.
• Percentage of Solves by Cross Color: Shows the percentage of solves starting with each cross color.
• Average solve time by inspection time: Visualizes the average solve time grouped by inspection time (e.g. the left bar is the 1/7 of solves with the lowest inspection time, the right bar the 1/7 with the most).
• Average Time by Step: Displays the average time taken for each step of the solve.
• Average Inspection Time: Shows the running average inspection time.
• Longest Daily Streaks: Shows how many days in a row you've achieved solves at each time threshold.
• Daily Fastest Solve: Shows the fastest solve for each day based on the selected filters.
• Current Records: Shows your current records for Single, Ao5, Ao12, Ao100, and Ao1000.
• OLL Edge Orientation (when OLL selected): Displays the percentage of OLL cases by edge orientation.
• PLL Corner Permutation (when PLL selected): Shows the percentage of PLL cases by corner permutation.
• Time Per Step Compared to Typical Solver (when CFOP with all steps): Compares your step times to those of a typical solver at your average; typical data is from Felix Zemdegs's CubeSkills blog.
• Percentage of the Solve Each Step Took (when 2+ steps selected): Doughnut chart of what percentage of your solve each step takes.
• Average Recognition Time and Execution Time per Case (when OLL or PLL only): Bar chart of recognition and execution time per last-layer algorithm, sorted by duration.
• Algorithm Practice (when OLL or PLL only): Per-case failure rate and move efficiency; "Failed" means core move count exceeded mode/average for that case; "Avg Wasted" shows redundant same-face moves that could be cancelled.
• Percentage of Good and Bad Solves (when all steps selected): Displays the percentage of solves considered good or bad based on user-defined criteria in the filter panel.
• History of Records (when all steps selected): Shows the history of personal bests (PBs) for single, Ao5, Ao12, Ao100, and Ao1000; only includes solves that meet the current filters.

Some charts appear only for certain method or step selections; the list above notes the main conditions in parentheses.

Supported data sources

SmartcubeAnalyzer can import and analyze solves exported from:

• Cubeast (cubeast.com)
• Acubemy (acubemy.com)

Both timer apps record:

• Detailed step splits
• Recognition and execution times
• Per-move timestamps

In addition:

• Acubemy stores gyro-based cube rotations
• Cubeast records inspection time

All sources are normalized so that metrics such as total time, recognition/execution splits, turns, and TPS are comparable across platforms.

How timings are interpreted

• Cubeast AUFs: Cubeast records AUF moves (U / U' / U2 / U3) at the start of a step as part of recognition. SmartcubeAnalyzer detects these leading AUFs in each step (except for Cross) and moves their duration from recognition into execution. This makes execution stats better reflect the actual turning you do, while still keeping total step time unchanged.
• Acubemy rotations: Acubemy records cube rotations in execution using gyro data, which can include spurious rotations that do not correspond to real turns and would otherwise inflate execution times. SmartcubeAnalyzer strips rotation moves from the solution so that turn counts, TPS, and step moves use only actual turns. When recomputing step recognition/execution, the timeline is already rotation-free; the time between the previous step’s last move and the current step’s first move (which in the raw data may include rotations) is assigned as recognition time for the current step, matching the same behavior used for Cubeast after AUF correction.

Installation

To install the dependencies, run:

npm install

Running the app

To start the development server, run:

npm start

This builds the app and starts a local development server (typically on http://localhost:3000), where you can explore the dashboards and visualizations.

Importing solve data

Once the app is running:

1. Export your solves from Cubeast or Acubemy.
2. Import the exported files into SmartcubeAnalyzer. You can upload one or more CSV files (from Cubeast and/or Acubemy) to Smartcube Analyzer and display your combined stats!
3. Explore the dashboards to drill into timings, TPS, and step-level performance.

Running Tests

To run the tests, use:

npm test

Contributing

Contributions are welcome! Please open an issue or submit a pull request.

License

This project is licensed under the MIT License.

Check it out!

To try it out for yourself, visit cuberplus.com.