AttinyStreamGfxApi VS gfx_demo

(fixed link: https://github.com/GoergPflug/AttinyStreamGfxApi)

GFX Library Documentation: https://www.codeproject.com/Articles/5302085/GFX-Forever-The-Complete-Guide-to-GFX-for-IoT

GFX is a feature rich graphics library that does jpg loading, truetype fonts, alpha blending, etc.

starting at line 573 here: https://github.com/codewitch-honey-crisis/gfx_demo/blob/master/lib/gfx/include/gfx_palette.hpp

basically draw to a bitmap and then blt that to the display. you can do the latter asynchronously, so you can load, then display, and while displaying, load the next one. it won't help immensely, but it will speed things up slightly. In GFX you'd use draw::bitmap_async. There's full documentation linked to from the link I gave you above - look in "Properly doing asynchronous draws" https://www.codeproject.com/Articles/5302085/GFX-Forever-The-Complete-Guide-to-GFX-for-IoT

I queue a variable number of transactions, either polling or non depending on context, using this code I wrote: https://github.com/codewitch-honey-crisis/gfx_demo/blob/master/src/esp-idf/drivers/common/spi_master.hpp

In case the OP doesn't respond, I'll field your question as best as I can: You can do the rendering using a bitmap flipping algorithm along with DMA. I also know it's possible to keep the framebuffer for that display in ram, with memory left over to do bitmap flipping because I do that very thing in https://github.com/codewitch-honey-crisis/gfx_demo - that device is a Lilygo-TTGO and I have several. The above code can run on it.