I have been looking into code and if I understand code correctly new IntersectObserver is created for every component, that has inview action.

My scenario has one root element which has many (thousands) of child elements/components , which lazy load an image, when they get into view.

As per this article: https://www.bennadel.com/blog/3954-intersectionobserver-api-performance-many-vs-shared-in-angular-11-0-5.htm it looks that there is notable performance advantage for single IntersectObserver having many observed elements, comparing to many IntersectObservers, each one just with one observed element.

What do you think?