Support for random displays for which the user already has a driver #2623
Replies: 2 comments
-
The library has previously used that method to support ePaper displays. using an external hardware interface library. The ePaper support has been deprecated recently. However, this may work as a "sprite only functionality" setup file but I am not able to test it:
I will add to my long to do list. |
Beta Was this translation helpful? Give feedback.
-
Thanks Bodmer, this is exactly what I was looking for. With the recent proliferation of cheap esoteric displays it's good to have an option to detach TFT_eSPI graphics from display driver and data bus config. It would be great to have a standard interface which all devs could agree on as an option - like Linux framebuffer. Once everything is up and running with whatever display then optimizing for speed with partial graphics RAM updates should be straightforward albeit at the mercy of the supplied display driver implementation. If such a TFT_eSPI profile could contain a data set of 'need to update' screen coordinates then this could be easily passed to our glue code. I can see a profile like this easing the burden on you to constantly support new hardware. It could be the corner for 'all others - at your own risk'. And then I bet the manufacturers of dev boards featuring these larger displays will provide glue code as often for TFT_eSPI as they seem now to do for LVGL. LVGL / Square Line Studio is absolutely fantastic but is nowhere near as fast as using TFT_eSPI raw. You can see this using Gifs in LVGL compared to TFT_eSPI / Larry Bank's Gif player - of course taking into account the LVGL overhead of dealing with touch etc. Am I correct in thinking that TFT_eSPI code is used 'under the hood' of LVGL? If so then this could be the nugget needed for a driver / bus agnostic framework for those who want to stick with Bodmer no matter what. |
Beta Was this translation helpful? Give feedback.
-
It would be useful to have a user profile which is driver agnostic and simply prints to a psram buffer. This way TFT_eSPI could be used directly with the new AMOLED boards from Lilygo and others, which always have unfamiliar driver chips but have driver code which will accept 'push colours' from a specified frame buffer. This would also allow TFT_eSPI to be used with arbitrary QSPI , RGB parallel (ST7701S etc) and MIPI-DSI displays. For those who would rather not use LVGL this would be welcome I think.
Beta Was this translation helpful? Give feedback.
All reactions