BRT Community

Thanks for your reply.
I had similar ideas and did some checks along those lines as well.

PCLK seems to be OK and aligned with the display datasheet. I also tried changing it, but it didn't have any visible effect on the issue.
I checked the REG_UNDERRUN yesterday in both projects. In both cases it is increasing, but it looks like in the large display project it increments much faster. That could explain the observed problems — in the smaller display project I didn't notice anything wrong because everything appeared to render correctly.

I ran a few tests on a very minimal setup and everything works fine as long as I don't use fonts or large images from flash. For example(displaying form flash):

• 800×480 image in ASTC 12×12 works fine
• 580×480 image in ASTC 4×4 causes the underrun counter to increase, but the display still looks correct

For images this is manageable since I can load and display them from RAM_G, but the real issue is with fonts. I need a relatively large font with a wide character set, and my tests show that even a standard Calibri font at 45pt with only basic Latin characters already causes the underrun counter to increase.

Do you have any recommendations for handling larger fonts or fonts with more glyphs when reading from flash, without increasing underrun or causing display issues?

Hi,
I'm working with BT817 (STM32 + EVE-MCU-Dev) and multiple displays. On a 480×272 panel (PH480272T015-IAA15), everything works reliably — ASTC images from flash, fonts, stable rendering.
After moving to larger panels (MI0700COT-1 800×480 and LMT070WS027C 1024×600), I'm seeing unstable and non-deterministic behavior. The initialization and overall approach are analogous, but results are inconsistent. Depending on reset, the display may work correctly or show flickering, corrupted fonts/images, or ghosting (previous screen content partially visible).
Importantly, the issue also occurs with minimal code using only CMD_TESTCARD(), which is sometimes stable and sometimes shows artifacts. This suggests the problem is not related to display lists or assets.
Given that even testcard is affected, I suspect RGB timing (porches/sync), PCLK (frequency/polarity), CSPREAD, possible underruns, or signal integrity issues at higher pixel clocks. I've tried multiple timing configurations (both from datasheets and library defaults) as well as different PCLK settings, but without consistent improvement.
Has anyone seen similar behavior when moving from 480×272 to higher resolutions? Are there known-good timing settings for these panels?

Additionally, I have a few related questions:

• What is the practical maximum size/resolution of images that can be displayed smoothly on BT817 (especially when using ASTC 4×4 or assets from flash)?
• In flash FULL mode, is the SPI clock fixed by EVE, or can it be increased to improve throughput?
• For very large fonts (~400–500 pt), EVE Asset Builder seems to hit limits (glyph width issues) — what are recommended approaches to handle large fonts while keeping full character support?

Any suggestions or best practices would be greatly appreciated.
Thanks!

Hi,
I'm working with a BT817 and I've encountered a strange issue with ROM fonts.
According to the documentation, the BT817/8 has built-in ROM fonts with handles 16–34, and fonts 31–34 are large ROM fonts encoded in ASTC 8x8 format.
However, on my system:
EVE_CMD_TEXT(20, 10, 31, 0, "BT817 Flash Image");  // works
EVE_CMD_TEXT(20, 10, 32, 0, "BT817 Flash Image");  // BT817 freezes
Font 31 works correctly, but as soon as I use font 32 (or any handle >31), the BT817 co-processor appears to hang (command FIFO stops progressing).
External flash is working correctly and is in FAST/FULL mode — I can read images from it without any issues.
I'm not sure whether I'll actually need these ROM fonts since I'll probably use custom fonts anyway, but I'd like to understand whether this is a known issue, a configuration mistake on my side, or possibly a silicon/firmware bug.
Thanks in advance for any advice.