Most web browsers nowadays support extensions that add, change, or remove certain functionality from the browser itself or websites.
Extensions may be installed to block advertisement or tracking, add functionality to websites, keep track of item prices, change the New Tab Page of the browser, or correct your spelling.
It should not come as a surprise that extensions that do get loaded consume some resources. 26 different Chrome extensions were analyzed recently to find out if they impact the browsing performance.
Popular extensions such as uBlock (Origin I suppose), Adblock Plus, HTTPS Everywhere, Grammarly, LastPass, or Google Dictionary were selected for the test.
The results confirm what many Internet users who use extensions probably know already:
- The average extension does not impact performance significantly.
- The more extensions you use, the higher the performance impact.
- Privacy tools improve performance on pages with lots of advertisement or tracking scripts even though they have a small initial cost.
- Some extensions, especially those with page-related actions, impact performance more.
Some extensions consume more CPU than others. Honey, a shopping extension, added 636ms of extra CPU time when installed. Grammarly, a spelling and grammar checker, more than 300ms, and Evernote Clipper and StayFocused, more than 200ms each.
Of all the tested content blockers, Ublock (again, I assume uBlock Origin is meant) used the least amount of CPU when loading tested sites. AdGuard, AdBlock, and AdBlock Plus all used more CPU in the conducted tests.
One privacy-focused extension, Privacy Badger, used less CPU than uBlock whereas Ghostery used more.
Extensions that don’t use best practices, e.g. an extension that is designed for a specific website but designed in a way so that it runs on all sites, impact performance more than they should. Chrome extension developers should also avoid running content scripts on document_start, and aim to keep the JavaScript that is included as small as possible.
It would be easy enough for the developer, usually, to address some or even all of these issues.
Closing Words
Researchers would have to repeat the test in a controlled environment and larger sample size to confirm the findings.
While Chrome has been at the center of the analysis, it is likely that the findings will hold true for other browsers as well.
The researcher did not look at memory usage or power consumption to find out if extensions impact those metrics as well.
A good advise is probably that you should try and run a minimal number of extensions in your browser of choice. Apart from improving performance, you may also improve privacy.
The post Surprise (not): Extensions may impact a browser’s performance appeared first on gHacks Technology News.