Performance#

What is Performance#

Performance primarily addresses the issue of how to make Lynx pages faster. Performance aims to achieve the dual objectives of efficient operation and smooth user experience. It encompasses two core dimensions: User Perceived Performance and Application Runtime Performance.

User Perceived Performance#

User Perceived Performance is a concept that measures users' subjective perception of a page being "fast" or "slow". It reflects users' perception of page loading and interaction response times, directly impacting user retention as it is the most direct subjective experience of application performance.

When users open a page, various reasons might prevent it from loading immediately, possibly leading them to exit prematurely, thinking they cannot access the page, resulting in a poor user experience. However, displaying a loading animation, friendly tips, or other engaging content can capture users' attention and create an impression of smooth application operation, thereby reducing users' perceived content loading time.

Optimizing user perceived performance is crucial for enhancing user retention. However, since it is a subjective feeling, measuring user perceived performance can be challenging. The Performance API offers metrics to assist in evaluation; more information can be found in Evaluating User Perceived Performance.

Application Runtime Performance#

Application runtime performance measures the concepts of loading and rendering speeds, serving as the technical foundation for supporting user experience. It can be represented by the time consumed at critical observable stages, which directly determine the speed at which page content is displayed, thereby affecting user perceived performance.

Poor application runtime performance manifests as "slow resource loading speed" or "slow page rendering speed," causing users to wait a long time to see page content. Pages that load slowly or are unresponsive can lead users to abandon further browsing.

Analyzing application runtime performance provides an objective basis for identifying and optimizing performance bottlenecks. By analyzing elapsed time at key stages such as initialization, resource preparation, and rendering processes, and through methods like modifying resource preloading, data partitioning, optimizing the rendering pipeline, or refining Lynx task processing logic, runtime performance can be improved. More information can be found in Evaluating Application Runtime Performance.