FAQ / BENCHMARK BASICS

Good questions, Clear answers

What is tested, what the scores mean and what is still being completed.

How often are results updated?

There is no fixed publication schedule. We aim to check for new grid versions once every three months and retest them where appropriate. The Test Date column shows when each grid was last tested; product release dates and benchmark test dates are different.

Why is a particular grid missing?

Coverage is limited to the supplied dataset and available test implementations. Absence is not a negative rating. Use Submit a grid to propose a product and provide official documentation and a representative use case.

How does testing work?

The supplied procedure starts each measurement on a freshly loaded page, repeats actions and selects a specified rounded result. Short actions use the third-smallest of 10-20 runs; long actions use the smallest of approximately three. Read Methodology for the environment and limitations.

Does the top score mean the best grid for every project?

No. The total depends on configured thresholds and selected scenarios. Framework fit, accessibility, license scope, server integration and your real workload may matter more than the overall score.

Are scrolling scores measured automatically?

The supplied Scroll blank and Scroll smooth values are manual ratings from 1 (best) to 6 (worst). They are different from the timed operations reported in milliseconds.

Can vendors pay for a better position?

The published testing principles prohibit paid placement and affiliate influence on rankings. Read the Testing rules page for the full policy.

Who operates CompareGrids?

CompareGrids is our business name for grid comparisons and integration services. Corporate guarantor and registration details are provided in Operator details on the About page and in the Legal notice.

Can I reproduce the benchmark from this download?

The package includes comparison templates, result data, scoring configuration, and FastGrid and TreeGrid test pages. Reproducing every published measurement also requires the corresponding product versions, datasets and test environment. Follow the methodology and verify each test implementation before comparing new runs with published results.

Are the listed prices current quotes?

No. They are estimates from the supplied dataset, rounded to 50 USD and based on the configured license scenario. Check current vendor pricing and contract terms before purchasing.

Full methodology → · Send a question or correction →