Cards are one additional layer to process

Am I the only one that thinks using cards as a UI design pattern is a way of dealing with shortage of screen space by replacing an appropriate amount of padding between elements with a background-colored grid?

The human eye can see a visual grid more easily, so yes it helps to split up areas quickly, but it also means, the eye needs to get involved with segmenting content int he first place, and then ignore them while parsing the content. On the other hand, if you have enough padding between elements the eye can totally focus on the content, while it is the higher functions of visual processing that clusters content into items.

So from a neurocognitive perspective, cards are more explicit but also need more computational power. Proper padding in contrast is more minimalistic as in not taking up too many resources of the eye and the brain thus leaving a stronger focus on the content itself.

If using more space really is impossible, I’d still go for subtle thin lines to delineate areas for clustering content into items than going for a card theme.

In my opinion, the card pattern make mostly sense when you have completely modular items that mix different kinds of modules in a single view — even more if the cards move around with lazy loading or drag and drop or screen resizing.