Stm32f429 w/ili9341 using LTDC

jhegedus · ‎2015-07-26

Posted on July 27, 2015 at 00:09

Hello,

I'm not sure what happened to my first post, but everything I originally typed was lost.

At any rate, to the point, I can't seem to avoid a snowy background to any picture I produce on the 429Discovery LCD board using LTDC. If I write directly to the screen using SPI, a black background appears opaque/flat black and everything is perfect. If I write to a layer using LTDC, snow appears on the screen behind whatever color I'm trying to fill the framebuffer with.

I've tried various clock settings (Usually ending up at 6MHz max for the LTDC/LCD setup), various system clock/PLL settings, various LCD configuration settings (Register/data combos), various LTDC/layer settings. I can make minor changes to what makes the snow a little less noticeable, but it still remains.

I have used/tested the TM library at STDiscovery, and those colors/fills appear flat black and are using LTDC. I've tried comparing all the settings to my program, and everything appears to be parallel. I have the feeling I'm overlooking something silly.

I can indeed use the aforementioned library, but I had originally started doing this on my own. I got far enough to produce an image and can put a pixel on the screen, but the snow annoys me. I was hoping to finish what I started and then I can always go onto using a better library (It's a personal thing).

At any rate, please let me know if you need further information. I'm hoping someone might have experienced something similar and can tell me something I've done silly. Thanks in advance for any help you may provide!

jhegedus · ‎2015-07-27

Posted on July 27, 2015 at 13:33

OK, in case anyone runs into a similar issue, I've found the cause of the snow on the LCD. I had enabled the random number generator on the board and was using that during my initial testing of the Discovery. I also had I2c, SPI, and an onboard health LED setup. I disabled everything but the LCD functions and the colors appeared as they were supposed to. I had a feeling it was the RNG, so I started there, and sure enough the snow returned.

So, I'll look into their interaction further to see what happened there, but I'm pretty happy to know I'm not going crazy. I guess I should've started with a fresh program during this initial testing, but I suppose I may have wanted to use the RNG at some point along with the LCD, so good to know in the end.