Rendering text on GPU metal vs CPU core graphics

Question

I’m trying to decide if it’s a good idea to render text to video on low level metal GPU or just keep using core graphics to draw text on video frame.

Using the GPU seems like a couple months of work to get it right and even then is simple static rendering. Will I see significant gains in performance if I change to GPU? I’m doing animated video captions renderer app and the text on metal seems complex.

I already got the metal pipeline running with some drawing

Answer 1

Will I see significant gains in performance if I change to gpu ?

It depends on a lot of factors, but you'll get a significant improvement anyway. In my program I have to deal with high resolution video (16K, 60FPS), and without native text rendering, my program simply will not work.

Using the gpu seems like a couple months of work to get it right and even then is simple static rendering.

It wouldn't take that long, there are so many articles in the web, here are some of them:

Rendering Text in Metal

Rendering Text in OpenGL

Text Rendering Library