Because there is no continuous rasterization of scaled up objects, it's important to have text in the same rendering space as the rest of the shapes. If you transform a group of shapes and want the text to be sharp, and why wouldn't you, it needs to regenerate the edges every frame.
Original Size

- Small
- Screenshot 2025-04-23 111803.jpg (634.6 KiB) Viewed 79 times
Bigly

- Big
- Screenshot 2025-04-23 111828.jpg (609.11 KiB) Viewed 79 times
It's the only way to make motion graphics work efficiently in Fusion. Everything else is making up hacks to sync up various transforms and font sizes to make it work.
P.S. There is no master opacity for any of the text. You can get around it with a mask input on normal Text nodes but inside the Shape system, none of them have a Mask input and any other way blows out the multi style layers.