Hi-
I have an OLED screen that uses display controller SEPS525. I use it in
SPI mode, in this mode it's not possible to read back from the SEPS525. I
have followed the instructions to integrate emWin in my system
(STM32) and I'm now able to use it with some success, using the
FlexColor driver F66718.
Since I cannot read back from the display controller, I settled for
using a display cache (which also makes sense for performance). However I
have an issue due to a shortage of available SRAM -- I have 32K, while
my display cache would need to hold a full 160x128x16bpp picture ~40K.
I have come to the conclusion that using a Memory Device instead would be a better
solution, as the latter allows usage of "banding". But that brings a
number of interrogations and concerns :
1) Reading the manual, I tend to conclude that there is no advantage in
using a display cache vs using Memory Devices (except code footprint
perhaps) : am I missing anything ? Is there a fundamental difference between both ?
2) In any case, "banding" is only available with Memory Devices (not with regular display cache) -- correct ?
3) From the manual I'm not very sure how to request usage of banding and I couldn't find an example in the emWin Samples page on the web site.
It seems I need to group all the GUI_xx operations I want to do on the selected rectangle into a callback function and call GUI_MEMDEV_Draw(),passing a pointer to this callback function as parameter GUI_CALLBACK_VOID_P * pfDraw ? Makes sense ?
Thanks for your help -
I have an OLED screen that uses display controller SEPS525. I use it in
SPI mode, in this mode it's not possible to read back from the SEPS525. I
have followed the instructions to integrate emWin in my system
(STM32) and I'm now able to use it with some success, using the
FlexColor driver F66718.
Since I cannot read back from the display controller, I settled for
using a display cache (which also makes sense for performance). However I
have an issue due to a shortage of available SRAM -- I have 32K, while
my display cache would need to hold a full 160x128x16bpp picture ~40K.
I have come to the conclusion that using a Memory Device instead would be a better
solution, as the latter allows usage of "banding". But that brings a
number of interrogations and concerns :
1) Reading the manual, I tend to conclude that there is no advantage in
using a display cache vs using Memory Devices (except code footprint
perhaps) : am I missing anything ? Is there a fundamental difference between both ?
2) In any case, "banding" is only available with Memory Devices (not with regular display cache) -- correct ?
3) From the manual I'm not very sure how to request usage of banding and I couldn't find an example in the emWin Samples page on the web site.
It seems I need to group all the GUI_xx operations I want to do on the selected rectangle into a callback function and call GUI_MEMDEV_Draw(),passing a pointer to this callback function as parameter GUI_CALLBACK_VOID_P * pfDraw ? Makes sense ?
Thanks for your help -