Use case or problem
Canvas library is very slow compare to that is used in graph visualization or in drawio. They are many times more performant and does not slow down when <-moving->; when the map grows to very large sizes on mobile or another weak hardware (hundreds, thousands of nodes). I had to switch to fast drawio plugin, but it doesn’t link with other files, internal hotkeys and plugins don’t work there, and also it’s not integrated normally into the interface and I had to go back. Also, the search by nodes inside the canvas is really missing (available in graph lib). Ideally, when switching from the main search on the Vault, it should “scroll” (zoom) to a specific node like in regular md files.
Proposed solution
Switch from one lib of canvas rendering to another:
- more performant
- available search from the box
For me canvas is a very important feature, it has long replaced mind maps software for me. Also Obsidian compatible between desktop and mobile that add weight to be killer to another software with close functions.
Hello this is two FR in one. There is already one for search. Regarding canvas being slow, what and where exactly is slow?
Can you post a screen recording?
The search problem is a link in the same chain: unadapted library for the task of navigating on canvas (lacking basic search and very poor performance of interface redrawing). Also, one problem aggravates the other: a quick search would allow not moving back and forth in search of the right nodes. This is very inconvenient and forces you to keep looking for alternative solutions or go back to drawio. The first priority is to simply add a search that would quickly jump from node to node on the map that contain certain text - this would immediately solve more than half of the problem.
The case is simple: create hundreds of linked nodes with text and images and run it on low-end hardware (pc, mobile). And then try to move the map in different directions (the problem is especially clear at large scale).
drawio, for example, renders very fast regardless of the map size and hardware performance.
